Leping

Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 EUROPEAN COMMISSION Directorate-General for Justice and Consumers JUST.H – Horizontal Affairs H.3 – Budget, Programmes and Financial Management GRANT AGREEMENT Project 101144709 — EquiTech PREAMBLE This Agreement (‘the Agreement’) is between the following parties: on the one part, the European Union (‘EU’), represented by the European Commission (‘European Commission’ or ‘granting authority’), and on the other part, 1. ‘the coordinator’: SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE VOLINIKU KANTSELEI (SVV), PIC 917673448, established in Kiriku 4, TALLINN 10130, Estonia, and the following other beneficiaries, if they sign their ‘accession form’ (see Annex 3 and Article 40): 2. LYGIU GALIMYBIU KONTROLIERIAUS TARNYBA (OEOO), PIC 913111732, established in S KONARSKIO G 35, VILNIUS 03123, Lithuania, 3. TALLINNA TEHNIKAÜLIKOOL (TalTech), PIC 999842536, established in EHITAJATE TEE 5, TALLINN 19086, Estonia, 4. MAJANDUS JA KOMMUNIKATSIOONIMINISTEERIUM (MKM), PIC 963638450, established in SUUR-AMEERIKA 1, TALLINN 10122, Estonia, 5. MINISTRY OF JUSTICE (JUM), PIC 974582378, established in TONISMAAGI 5A, TALLINN 15191, Estonia, Unless otherwise specified, references to ‘beneficiary’ or ‘beneficiaries’ include the coordinator and affiliated entities (if any). If only one beneficiary signs the grant agreement (‘mono-beneficiary grant’), all provisions referring to the ‘coordinator’ or the ‘beneficiaries’ will be considered — mutatis mutandis — as referring to the beneficiary. The parties referred to above have agreed to enter into the Agreement. 1 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 By signing the Agreement and the accession forms, the beneficiaries accept the grant and agree to implement the action under their own responsibility and in accordance with the Agreement, with all the obligations and terms and conditions it sets out. The Agreement is composed of: Preamble Terms and Conditions (including Data Sheet) Annex 1 Description of the action1 Annex 2 Estimated budget for the action Annex 3 Accession forms (if applicable)2 Annex 3a Declaration on joint and several liability of affiliated entities (if applicable)3 Annex 4 Model for the financial statements Annex 5 Specific rules (if applicable) 1 Template published on Portal Reference Documents. 2 Template published on Portal Reference Documents. 3 Template published on Portal Reference Documents. 2 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 TERMS AND CONDITIONS TABLE OF CONTENTS GRANT AGREEMENT.................................................................................................................................................... 1 PREAMBLE........................................................................................................................................................................1 TERMS AND CONDITIONS...........................................................................................................................................3 DATASHEET...................................................................................................................................................................... 8 CHAPTER 1 GENERAL..............................................................................................................................................12 ARTICLE 1 — SUBJECT OF THE AGREEMENT ..................................................................................... 12 ARTICLE 2 — DEFINITIONS........................................................................................................................12 CHAPTER 2 ACTION................................................................................................................................................. 13 ARTICLE 3 — ACTION................................................................................................................................. 13 ARTICLE 4 — DURATION AND STARTING DATE...................................................................................13 CHAPTER 3 GRANT...................................................................................................................................................13 ARTICLE 5 — GRANT...................................................................................................................................13 5.1 Form of grant......................................................................................................................................13 5.2 Maximum grant amount..................................................................................................................... 14 5.3 Funding rate........................................................................................................................................14 5.4 Estimated budget, budget categories and forms of funding.............................................................. 14 5.5 Budget flexibility................................................................................................................................14 ARTICLE 6 — ELIGIBLE AND INELIGIBLE CONTRIBUTIONS............................................................ 14 6.1 and 6.2 General and specific eligibility conditions........................................................................... 14 6.3 Ineligible contributions.......................................................................................................................15 6.4 Consequences of non-compliance...................................................................................................... 15 CHAPTER 4 GRANT IMPLEMENTATION............................................................................................................ 15 SECTION 1 CONSORTIUM: BENEFICIARIES, AFFILIATED ENTITIES AND OTHER PARTICIPANTS........................................................................................................................................ 15 ARTICLE 7 — BENEFICIARIES................................................................................................................... 15 ARTICLE 8 — AFFILIATED ENTITIES....................................................................................................... 17 ARTICLE 9 — OTHER PARTICIPANTS INVOLVED IN THE ACTION................................................... 17 9.1 Associated partners.............................................................................................................................17 9.2 Third parties giving in-kind contributions to the action.................................................................... 17 9.3 Subcontractors.....................................................................................................................................18 9.4 Recipients of financial support to third parties..................................................................................18 3 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 ARTICLE 10 — PARTICIPANTS WITH SPECIAL STATUS....................................................................... 18 10.1 Non-EU participants......................................................................................................................... 18 10.2 Participants which are international organisations...........................................................................19 10.3 Pillar-assessed participants............................................................................................................... 19 SECTION 2 RULES FOR CARRYING OUT THE ACTION...........................................................................21 ARTICLE 11 — PROPER IMPLEMENTATION OF THE ACTION............................................................ 21 11.1 Obligation to properly implement the action................................................................................... 21 11.2 Consequences of non-compliance.................................................................................................... 21 ARTICLE 12 — CONFLICT OF INTERESTS.............................................................................................. 22 12.1 Conflict of interests.......................................................................................................................... 22 12.2 Consequences of non-compliance.................................................................................................... 22 ARTICLE 13 — CONFIDENTIALITY AND SECURITY............................................................................ 22 13.1 Sensitive information........................................................................................................................22 13.2 Classified information...................................................................................................................... 23 13.3 Consequences of non-compliance.................................................................................................... 23 ARTICLE 14 — ETHICS AND VALUES...................................................................................................... 23 14.1 Ethics.................................................................................................................................................23 14.2 Values................................................................................................................................................ 23 14.3 Consequences of non-compliance.................................................................................................... 24 ARTICLE 15 — DATA PROTECTION.......................................................................................................... 24 15.1 Data processing by the granting authority....................................................................................... 24 15.2 Data processing by the beneficiaries............................................................................................... 24 15.3 Consequences of non-compliance.................................................................................................... 25 ARTICLE 16 — INTELLECTUAL PROPERTY RIGHTS (IPR) — BACKGROUND AND RESULTS — ACCESS RIGHTS AND RIGHTS OF USE................................................................................ 25 16.1 Background and access rights to background..................................................................................25 16.2 Ownership of results.........................................................................................................................25 16.3 Rights of use of the granting authority on materials, documents and information received for policy, information, communication, dissemination and publicity purposes...................................25 16.4 Specific rules on IPR, results and background................................................................................ 26 16.5 Consequences of non-compliance.................................................................................................... 26 ARTICLE 17 — COMMUNICATION, DISSEMINATION AND VISIBILITY............................................ 27 17.1 Communication — Dissemination — Promoting the action........................................................... 27 17.2 Visibility — European flag and funding statement......................................................................... 27 17.3 Quality of information — Disclaimer..............................................................................................28 17.4 Specific communication, dissemination and visibility rules............................................................28 4 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 17.5 Consequences of non-compliance.................................................................................................... 28 ARTICLE 18 — SPECIFIC RULES FOR CARRYING OUT THE ACTION............................................... 28 SECTION 3 GRANT ADMINISTRATION......................................................................................................... 28 ARTICLE 19 — GENERAL INFORMATION OBLIGATIONS.................................................................... 28 19.1 Information requests......................................................................................................................... 28 19.2 Participant Register data updates..................................................................................................... 28 19.3 Information about events and circumstances which impact the action............................................28 19.4 Consequences of non-compliance.................................................................................................... 29 ARTICLE 20 — RECORD-KEEPING............................................................................................................ 29 20.1 Keeping records and supporting documents.................................................................................... 29 20.2 Consequences of non-compliance.................................................................................................... 29 ARTICLE 21 — REPORTING........................................................................................................................ 30 21.1 Continuous reporting........................................................................................................................ 30 21.2 Periodic reporting: Technical reports and financial statements....................................................... 30 21.3 Currency for financial statements and conversion into euros..........................................................31 21.4 Reporting language...........................................................................................................................31 21.5 Consequences of non-compliance.................................................................................................... 31 ARTICLE 22 — PAYMENTS AND RECOVERIES — CALCULATION OF AMOUNTS DUE................. 31 22.1 Payments and payment arrangements.............................................................................................. 31 22.2 Recoveries.........................................................................................................................................32 22.3 Amounts due.....................................................................................................................................32 22.4 Enforced recovery.............................................................................................................................36 22.5 Consequences of non-compliance.................................................................................................... 37 ARTICLE 23 — GUARANTEES....................................................................................................................38 23.1 Prefinancing guarantee..................................................................................................................... 38 23.2 Consequences of non-compliance.................................................................................................... 38 ARTICLE 24 — CERTIFICATES................................................................................................................... 38 ARTICLE 25 — CHECKS, REVIEWS, AUDITS AND INVESTIGATIONS — EXTENSION OF FINDINGS..................................................................................................................................... 38 25.1 Granting authority checks, reviews and audits................................................................................ 38 25.2 European Commission checks, reviews and audits in grants of other granting authorities..............40 25.3 Access to records for assessing simplified forms of funding.......................................................... 40 25.4 OLAF, EPPO and ECA audits and investigations........................................................................... 40 25.5 Consequences of checks, reviews, audits and investigations — Extension of findings...................41 25.6 Consequences of non-compliance.................................................................................................... 42 5 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 ARTICLE 26 — IMPACT EVALUATIONS................................................................................................... 42 26.1 Impact evaluation............................................................................................................................. 42 26.2 Consequences of non-compliance.................................................................................................... 42 CHAPTER 5 CONSEQUENCES OF NON-COMPLIANCE.................................................................................. 43 SECTION 1 REJECTIONS AND GRANT REDUCTION.................................................................................43 ARTICLE 27 — REJECTION OF CONTRIBUTIONS..................................................................................43 27.1 Conditions......................................................................................................................................... 43 27.2 Procedure.......................................................................................................................................... 43 27.3 Effects............................................................................................................................................... 43 ARTICLE 28 — GRANT REDUCTION........................................................................................................ 43 28.1 Conditions......................................................................................................................................... 43 28.2 Procedure.......................................................................................................................................... 44 28.3 Effects............................................................................................................................................... 44 SECTION 2 SUSPENSION AND TERMINATION............................................................................................44 ARTICLE 29 — PAYMENT DEADLINE SUSPENSION............................................................................. 44 29.1 Conditions......................................................................................................................................... 44 29.2 Procedure.......................................................................................................................................... 44 ARTICLE 30 — PAYMENT SUSPENSION...................................................................................................45 30.1 Conditions......................................................................................................................................... 45 30.2 Procedure.......................................................................................................................................... 45 ARTICLE 31 — GRANT AGREEMENT SUSPENSION..............................................................................46 31.1 Consortium-requested GA suspension............................................................................................. 46 31.2 EU-initiated GA suspension.............................................................................................................46 ARTICLE 32 — GRANT AGREEMENT OR BENEFICIARY TERMINATION......................................... 47 32.1 Consortium-requested GA termination............................................................................................ 48 32.2 Consortium-requested beneficiary termination................................................................................ 48 32.3 EU-initiated GA or beneficiary termination.................................................................................... 50 SECTION 3 OTHER CONSEQUENCES: DAMAGES AND ADMINISTRATIVE SANCTIONS............... 53 ARTICLE 33 — DAMAGES...........................................................................................................................53 33.1 Liability of the granting authority....................................................................................................53 33.2 Liability of the beneficiaries............................................................................................................ 53 ARTICLE 34 — ADMINISTRATIVE SANCTIONS AND OTHER MEASURES....................................... 53 SECTION 4 FORCE MAJEURE.......................................................................................................................... 53 ARTICLE 35 — FORCE MAJEURE.............................................................................................................. 53 CHAPTER 6 FINAL PROVISIONS...........................................................................................................................54 6 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 ARTICLE 36 — COMMUNICATION BETWEEN THE PARTIES...............................................................54 36.1 Forms and means of communication — Electronic management................................................... 54 36.2 Date of communication.................................................................................................................... 54 36.3 Addresses for communication.......................................................................................................... 55 ARTICLE 37 — INTERPRETATION OF THE AGREEMENT.....................................................................55 ARTICLE 38 — CALCULATION OF PERIODS AND DEADLINES..........................................................55 ARTICLE 39 — AMENDMENTS.................................................................................................................. 55 39.1 Conditions......................................................................................................................................... 55 39.2 Procedure.......................................................................................................................................... 55 ARTICLE 40 — ACCESSION AND ADDITION OF NEW BENEFICIARIES........................................... 56 40.1 Accession of the beneficiaries mentioned in the Preamble............................................................. 56 40.2 Addition of new beneficiaries.......................................................................................................... 56 ARTICLE 41 — TRANSFER OF THE AGREEMENT................................................................................. 57 ARTICLE 42 — ASSIGNMENTS OF CLAIMS FOR PAYMENT AGAINST THE GRANTING AUTHORITY.................................................................................................................................57 ARTICLE 43 — APPLICABLE LAW AND SETTLEMENT OF DISPUTES.............................................. 57 43.1 Applicable law..................................................................................................................................57 43.2 Dispute settlement............................................................................................................................ 57 ARTICLE 44 — ENTRY INTO FORCE.........................................................................................................58 7 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 DATA SHEET 1. General data Project summary: Project summary Public authorities’ uptake of AI and other algorithmic tools in data collection, processing and services is a rapid and, most probably, a non-reversible new reality. Globally, many governments use automated decision support tools or even automated (algorithm-based) decision-making (ADM) systems in their everyday operations. ADM is the process of making a decision or creating other legally relevant output through automated means with limited or no human involvement. The ADM systems may process large amounts of information, identify patterns, make predictions, or take various types of actions based on predefined criteria. The general objective of the project is to enhance the capacity of public authorities and their co-operation partners in Estonia and Lithuania in addressing risks of discrimination and bias in ADM systems. This will be achieved by in-depth research, developing support materials, trainings and a media campaign. This project seeks to contribute to prevention of discrimination and gender inequality within the European Union context by ensuring that the ADM systems used in the public sector are fair and non-discriminatory. The key activities include: 1. Research on gaps in policies and legal frameworks on AI in Estonia and Lithuania together with mapping of potential risks of discrimination and bias in ADM systems deployed by Estonian public authorities; 2. Development of guidelines, impact assessment checklist, training materials and online course for public sector employees and system developers to address potential risks for discrimination and bias in design and deployment of ADM systems; 3. Capacity building among project stakeholders and target groups through trainings, study trips, international conference and an academic workshop; 4.Raising public awareness on the risks of discrimination and bias in public sector ADM systems. Keywords: – To combat discrimination, bias, and intolerance in automated decision-making systems. Project number: 101144709 Project name: Improving response to risks of discrimination, bias and intolerance in automated decision-making systems to promote equality. Project acronym: EquiTech Call: CERV-2023-EQUAL Topic: CERV-2023-EQUAL Type of action: CERV Lump Sum Grants Granting authority: European Commission-EU Grant managed through EU Funding & Tenders Portal: Yes (eGrants) Project starting date: fixed date: 1 May 2024 Project end date: 30 April 2026 Project duration: 24 months Consortium agreement: Yes 2. Participants List of participants: N° Role Short name Legal name Ctry PIC Max grant amount 1 COO SVV SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE EE 917673448 168 142.00 VOLINIKU KANTSELEI 2 BEN OEOO LYGIU GALIMYBIU KONTROLIERIAUS TARNYBA LT 913111732 94 335.00 3 BEN TalTech TALLINNA TEHNIKAÜLIKOOL EE 999842536 139 731.00 4 BEN MKM MAJANDUS JA KOMMUNIKATSIOONIMINISTEERIUM EE 963638450 70 973.00 8 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 N° Role Short name Legal name Ctry PIC Max grant amount 5 BEN JUM MINISTRY OF JUSTICE EE 974582378 147 435.00 Total 620 616.00 Coordinator: – SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE VOLINIKU KANTSELEI (SVV) 3. Grant Maximum grant amount, total estimated eligible costs and contributions and funding rate: Maximum grant amount Maximum grant amount (Annex 2) (award decision) 620 616.00 620 616.00 Grant form: Lump Sum Grant mode: Action grant Budget categories/activity types: Lump sum contributions Cost eligibility options: n/a Budget flexibility: No 4. Reporting, payments and recoveries 4.1 Continuous reporting (art 21) Deliverables: see Funding & Tenders Portal Continuous Reporting tool 4.2 Periodic reporting and payments Reporting and payment schedule (art 21, 22): Reporting Payments Deadline Reporting periods Type Deadline Type (time to pay) RP No Month from Month to 30 days from entry into force/ financial guarantee Initial prefinancing (if required) – whichever is the latest 1 1 24 Periodic report 60 days after end Final payment 90 days from of reporting period receiving periodic report 9 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Prefinancing payments and guarantees: Prefinancing payment Prefinancing guarantee Guarantee Type Amount Division per participant amount Prefinancing 1 (initial) 496 492.80 n/a 1 - SVV n/a 2 - OEOO n/a 3 - TalTech n/a 4 - MKM n/a 5 - JUM n/a Reporting and payment modalities (art 21, 22): Mutual Insurance Mechanism (MIM): No Restrictions on distribution of initial prefinancing: The prefinancing may be distributed only if the minimum number of beneficiaries set out in the call condititions (if any) have acceded to the Agreement and only to beneficiaries that have acceded. Interim payment ceiling (if any): 100% of the maximum grant amount No-profit rule: n/a Late payment interest: ECB + 3.5% Bank account for payments: EE891010220034796011 EEUHEE2X Conversion into euros: n/a Reporting language: Language of the Agreement or other EU official language, if specified in the call conditions 4.3 Certificates (art 24): n/a 4.4 Recoveries (art 22) First-line liability for recoveries: Beneficiary termination: Beneficiary concerned Final payment: Coordinator After final payment: Beneficiary concerned Joint and several liability for enforced recoveries (in case of non-payment): Limited joint and several liability of other beneficiaries — up to the maximum grant amount of the beneficiary Joint and several liability of affiliated entities — n/a 5. Consequences of non-compliance, applicable law & dispute settlement forum Applicable law (art 43): Standard applicable law regime: EU law + law of Belgium Dispute settlement forum (art 43): 10 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Standard dispute settlement forum: EU beneficiaries: EU General Court + EU Court of Justice (on appeal) Non-EU beneficiaries: Courts of Brussels, Belgium (unless an international agreement provides for the enforceability of EU court judgements) 6. Other Specific rules (Annex 5): Yes Standard time-limits after project end: Confidentiality (for X years after final payment): 5 Record-keeping (for X years after final payment): 5 (or 3 for grants of not more than EUR 60 000) Reviews (up to X years after final payment): 5 (or 3 for grants of not more than EUR 60 000) Audits (up to X years after final payment): 5 (or 3 for grants of not more than EUR 60 000) Extension of findings from other grants to this grant (no later than X years after final payment): 5 (or 3 for grants of not more than EUR 60 000) Impact evaluation (up to X years after final payment): 5 (or 3 for grants of not more than EUR 60 000) 11 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 CHAPTER 1 GENERAL ARTICLE 1 — SUBJECT OF THE AGREEMENT This Agreement sets out the rights and obligations and terms and conditions applicable to the grant awarded for the implementation of the action set out in Chapter 2. ARTICLE 2 — DEFINITIONS For the purpose of this Agreement, the following definitions apply: Actions — The project which is being funded in the context of this Agreement. Grant — The grant awarded in the context of this Agreement. EU grants — Grants awarded by EU institutions, bodies, offices or agencies (including EU executive agencies, EU regulatory agencies, EDA, joint undertakings, etc.). Participants — Entities participating in the action as beneficiaries, affiliated entities, associated partners, third parties giving in-kind contributions, subcontractors or recipients of financial support to third parties. Beneficiaries (BEN) — The signatories of this Agreement (either directly or through an accession form). Affiliated entities (AE) — Entities affiliated to a beneficiary within the meaning of Article 187 of EU Financial Regulation 2018/10464 which participate in the action with similar rights and obligations as the beneficiaries (obligation to implement action tasks and right to charge costs and claim contributions). Associated partners (AP) — Entities which participate in the action, but without the right to charge costs or claim contributions. Purchases — Contracts for goods, works or services needed to carry out the action (e.g. equipment, consumables and supplies) but which are not part of the action tasks (see Annex 1). Subcontracting — Contracts for goods, works or services that are part of the action tasks (see Annex 1). In-kind contributions — In-kind contributions within the meaning of Article 2(36) of EU Financial 4 For the definition, see Article 187 Regulation (EU, Euratom) 2018/1046 of the European Parliament and of the Council of 18 July 2018 on the financial rules applicable to the general budget of the Union, amending Regulations (EU) No 1296/2013, (EU) No 1301/2013, (EU) No 1303/2013, (EU) No 1304/2013, (EU) No 1309/2013, (EU) No 1316/2013, (EU) No 223/2014, (EU) No 283/2014, and Decision No 541/2014/EU and repealing Regulation (EU, Euratom) No 966/2012 (‘EU Financial Regulation’) (OJ L 193, 30.7.2018, p. 1): “affiliated entities [are]: (a) entities that form a sole beneficiary [(i.e. where an entity is formed of several entities that satisfy the criteria for being awarded a grant, including where the entity is specifically established for the purpose of implementing an action to be financed by a grant)]; (b) entities that satisfy the eligibility criteria and that do not fall within one of the situations referred to in Article 136(1) and 141(1) and that have a link with the beneficiary, in particular a legal or capital link, which is neither limited to the action nor established for the sole purpose of its implementation”. 12 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Regulation 2018/1046, i.e. non-financial resources made available free of charge by third parties. Fraud — Fraud within the meaning of Article 3 of EU Directive 2017/13715 and Article 1 of the Convention on the protection of the European Communities’ financial interests, drawn up by the Council Act of 26 July 19956, as well as any other wrongful or criminal deception intended to result in financial or personal gain. Irregularities — Any type of breach (regulatory or contractual) which could impact the EU financial interests, including irregularities within the meaning of Article 1(2) of EU Regulation 2988/957. Grave professional misconduct — Any type of unacceptable or improper behaviour in exercising one’s profession, especially by employees, including grave professional misconduct within the meaning of Article 136(1)(c) of EU Financial Regulation 2018/1046. Applicable EU, international and national law — Any legal acts or other (binding or non-binding) rules and guidance in the area concerned. Portal — EU Funding & Tenders Portal; electronic portal and exchange system managed by the European Commission and used by itself and other EU institutions, bodies, offices or agencies for the management of their funding programmes (grants, procurements, prizes, etc.). CHAPTER 2 ACTION ARTICLE 3 — ACTION The grant is awarded for the action 101144709 — EquiTech (‘action’), as described in Annex 1. ARTICLE 4 — DURATION AND STARTING DATE The duration and the starting date of the action are set out in the Data Sheet (see Point 1). CHAPTER 3 GRANT ARTICLE 5 — GRANT 5.1 Form of grant 5 Directive (EU) 2017/1371 of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law (OJ L 198, 28.7.2017, p. 29). 6 OJ C 316, 27.11.1995, p. 48. 7 Council Regulation (EC, Euratom) No 2988/95 of 18 December 1995 on the protection of the European Communities financial interests (OJ L 312, 23.12.1995, p. 1). 13 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The grant is an action grant8 which takes the form of a lump sum grant for the completion of work packages. 5.2 Maximum grant amount The maximum grant amount is set out in the Data Sheet (see Point 3) and in the estimated budget (Annex 2). 5.3 Funding rate Not applicable 5.4 Estimated budget, budget categories and forms of funding The estimated budget for the action (lump sum breakdown) is set out in Annex 2. It contains the estimated eligible contributions for the action (lump sum contributions), broken down by participant and work package. Annex 2 also shows the types of contributions (forms of funding)9 to be used for each work package. 5.5 Budget flexibility Budget flexibility does not apply; changes to the estimated budget (lump sum breakdown) always require an amendment (see Article 39). Amendments for transfers between work packages are moreover possible only if: - the work packages concerned are not already completed (and declared in a financial statement) and - the transfers are justified by the technical implementation of the action. ARTICLE 6 — ELIGIBLE AND INELIGIBLE CONTRIBUTIONS 6.1 and 6.2 General and specific eligibility conditions Lump sum contributions are eligible (‘eligible contributions’), if: (a) they are set out in Annex 2 and (b) the work packages are completed and the work is properly implemented by the beneficiaries and/or the results are achieved, in accordance with Annex 1 and during in the period set out in Article 4 (with the exception of work/results relating to the submission of the final periodic report, which may be achieved afterwards; see Article 21) They will be calculated on the basis of the amounts set out in Annex 2. 8 For the definition, see Article 180(2)(a) EU Financial Regulation 2018/1046: ‘action grant’ means an EU grant to finance “an action intended to help achieve a Union policy objective”. 9 See Article 125 EU Financial Regulation 2018/1046. 14 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 6.3 Ineligible contributions ‘Ineligible contributions’ are: (a) lump sum contributions that do not comply with the conditions set out above (see Article 6.1 and 6.2) (b) lump sum contributions for activities already funded under other EU grants (or grants awarded by an EU Member State, non-EU country or other body implementing the EU budget), except for the following case: (i) Synergy actions: not applicable (c) other: (i) country restrictions for eligible costs: not applicable. 6.4 Consequences of non-compliance If a beneficiary declares lump sum contributions that are ineligible, they will be rejected (see Article 27). This may also lead to other measures described in Chapter 5. CHAPTER 4 GRANT IMPLEMENTATION SECTION 1 CONSORTIUM: BENEFICIARIES, AFFILIATED ENTITIES AND OTHER PARTICIPANTS ARTICLE 7 — BENEFICIARIES The beneficiaries, as signatories of the Agreement, are fully responsible towards the granting authority for implementing it and for complying with all its obligations. They must implement the Agreement to their best abilities, in good faith and in accordance with all the obligations and terms and conditions it sets out. They must have the appropriate resources to implement the action and implement the action under their own responsibility and in accordance with Article 11. If they rely on affiliated entities or other participants (see Articles 8 and 9), they retain sole responsibility towards the granting authority and the other beneficiaries. They are jointly responsible for the technical implementation of the action. If one of the beneficiaries fails to implement their part of the action, the other beneficiaries must ensure that this part is implemented by someone else (without being entitled to an increase of the maximum grant amount and subject to an amendment; see Article 39). The financial responsibility of each beneficiary in case of recoveries is governed by Article 22. The beneficiaries (and their action) must remain eligible under the EU programme funding the grant 15 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 for the entire duration of the action. Lump sum contributions will be eligible only as long as the beneficiary and the action are eligible. The internal roles and responsibilities of the beneficiaries are divided as follows: (a) Each beneficiary must: (i) keep information stored in the Portal Participant Register up to date (see Article 19) (ii) inform the granting authority (and the other beneficiaries) immediately of any events or circumstances likely to affect significantly or delay the implementation of the action (see Article 19) (iii) submit to the coordinator in good time: - the prefinancing guarantees (if required; see Article 23) - the financial statements and certificates on the financial statements (CFS): not applicable - the contribution to the deliverables and technical reports (see Article 21) - any other documents or information required by the granting authority under the Agreement (iv) submit via the Portal data and information related to the participation of their affiliated entities. (b) The coordinator must: (i) monitor that the action is implemented properly (see Article 11) (ii) act as the intermediary for all communications between the consortium and the granting authority, unless the Agreement or granting authority specifies otherwise, and in particular: - submit the prefinancing guarantees to the granting authority (if any) - request and review any documents or information required and verify their quality and completeness before passing them on to the granting authority - submit the deliverables and reports to the granting authority - inform the granting authority about the payments made to the other beneficiaries (report on the distribution of payments; if required, see Articles 22 and 32) (iii) distribute the payments received from the granting authority to the other beneficiaries without unjustified delay (see Article 22). The coordinator may not delegate or subcontract the above-mentioned tasks to any other beneficiary or third party (including affiliated entities). However, coordinators which are public bodies may delegate the tasks set out in Point (b)(ii) last 16 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 indent and (iii) above to entities with ‘authorisation to administer’ which they have created or which are controlled by or affiliated to them. In this case, the coordinator retains sole responsibility for the payments and for compliance with the obligations under the Agreement. Moreover, coordinators which are ‘sole beneficiaries’10 (or similar, such as European research infrastructure consortia (ERICs)) may delegate the tasks set out in Point (b)(i) to (iii) above to one of their members. The coordinator retains sole responsibility for compliance with the obligations under the Agreement. The beneficiaries must have internal arrangements regarding their operation and co-ordination, to ensure that the action is implemented properly. If required by the granting authority (see Data Sheet, Point 1), these arrangements must be set out in a written consortium agreement between the beneficiaries, covering for instance: - the internal organisation of the consortium - the management of access to the Portal - different distribution keys for the payments and financial responsibilities in case of recoveries (if any) - additional rules on rights and obligations related to background and results (see Article 16) - settlement of internal disputes - liability, indemnification and confidentiality arrangements between the beneficiaries. The internal arrangements must not contain any provision contrary to this Agreement. ARTICLE 8 — AFFILIATED ENTITIES Not applicable ARTICLE 9 — OTHER PARTICIPANTS INVOLVED IN THE ACTION 9.1 Associated partners Not applicable 9.2 Third parties giving in-kind contributions to the action Other third parties may give in-kind contributions to the action (i.e. personnel, equipment, other goods, works and services, etc. which are free-of-charge), if necessary for the implementation. Third parties giving in-kind contributions do not implement any action tasks. They may not charge contributions to the action (no lump sum contributions) and the costs for the in-kind contributions are not eligible (may not be included in the estimated budget in Annex 2). 10 For the definition, see Article 187(2) EU Financial Regulation 2018/1046: “Where several entities satisfy the criteria for being awarded a grant and together form one entity, that entity may be treated as the sole beneficiary, including where it is specifically established for the purpose of implementing the action financed by the grant.” 17 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The third parties and their in-kind contributions should be set out in Annex 1. 9.3 Subcontractors Subcontractors may participate in the action, if necessary for the implementation. Subcontractors must implement their action tasks in accordance with Article 11. The beneficiaries’ costs for subcontracting are considered entirely covered by the lump sum contributions for implementing the work packages (irrespective of the actual subcontracting costs incurred, if any). The beneficiaries must ensure that their contractual obligations under Articles 11 (proper implementation), 12 (conflict of interest), 13 (confidentiality and security), 14 (ethics), 17.2 (visibility), 18 (specific rules for carrying out action), 19 (information) and 20 (record-keeping) also apply to the subcontractors. The beneficiaries must ensure that the bodies mentioned in Article 25 (e.g. granting authority, OLAF, Court of Auditors (ECA), etc.) can exercise their rights also towards the subcontractors. 9.4 Recipients of financial support to third parties If the action includes providing financial support to third parties (e.g. grants, prizes or similar forms of support), the beneficiaries must ensure that their contractual obligations under Articles 12 (conflict of interest), 13 (confidentiality and security), 14 (ethics), 17.2 (visibility), 18 (specific rules for carrying out action), 19 (information) and 20 (record-keeping)also apply to the third parties receiving the support (recipients). The beneficiaries must also ensure that the bodies mentioned in Article 25 (e.g. granting authority, OLAF, Court of Auditors (ECA), etc.) can exercise their rights also towards the recipients. ARTICLE 10 — PARTICIPANTS WITH SPECIAL STATUS 10.1 Non-EU participants Participants which are established in a non-EU country (if any) undertake to comply with their obligations under the Agreement and: - to respect general principles (including fundamental rights, values and ethical principles, environmental and labour standards, rules on classified information, intellectual property rights, visibility of funding and protection of personal data) - for the submission of certificates under Article 24: use qualified external auditors which are independent and comply with comparable standards as those set out in EU Directive 2006/43/EC11 - for the controls under Article 25: allow for checks, reviews, audits and investigations (including on-the-spot checks, visits and inspections) by the bodies mentioned in that Article (e.g. granting authority, OLAF, Court of Auditors (ECA), etc.). 11 Directive 2006/43/EC of the European Parliament and of the Council of 17 May 2006 on statutory audits of annual accounts and consolidated accounts or similar national regulations (OJ L 157, 9.6.2006, p. 87). 18 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Special rules on dispute settlement apply (see Data Sheet, Point 5). 10.2 Participants which are international organisations Participants which are international organisations (IOs; if any) undertake to comply with their obligations under the Agreement and: - to respect general principles (including fundamental rights, values and ethical principles, environmental and labour standards, rules on classified information, intellectual property rights, visibility of funding and protection of personal data) - for the submission of certificates under Article 24: to use either independent public officers or external auditors which comply with comparable standards as those set out in EU Directive 2006/43/EC - for the controls under Article 25: to allow for the checks, reviews, audits and investigations by the bodies mentioned in that Article, taking into account the specific agreements concluded by them and the EU (if any). For such participants, nothing in the Agreement will be interpreted as a waiver of their privileges or immunities, as accorded by their constituent documents or international law. Special rules on applicable law and dispute settlement apply (see Article 43 and Data Sheet, Point 5). 10.3 Pillar-assessed participants Pillar-assessed participants (if any) may rely on their own systems, rules and procedures, in so far as they have been positively assessed and do not call into question the decision awarding the grant or breach the principle of equal treatment of applicants or beneficiaries. ‘Pillar-assessment’ means a review by the European Commission on the systems, rules and procedures which participants use for managing EU grants (in particular internal control system, accounting system, external audits, financing of third parties, rules on recovery and exclusion, information on recipients and protection of personal data; see Article 154 EU Financial Regulation 2018/1046). Participants with a positive pillar assessment may rely on their own systems, rules and procedures, in particular for: - record-keeping (Article 20): may be done in accordance with internal standards, rules and procedures - currency conversion for financial statements (Article 21): may be done in accordance with usual accounting practices - guarantees (Article 23): for public law bodies, prefinancing guarantees are not needed - certificates (Article 24): - certificates on the financial statements (CFS): may be provided by their regular internal or external auditors and in accordance with their internal financial regulations and procedures 19 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 - certificates on usual accounting practices (CoMUC): are not needed if those practices are covered by an ex-ante assessment and use the following specific rules, for: - recoveries (Article 22): in case of financial support to third parties, there will be no recovery if the participant has done everything possible to retrieve the undue amounts from the third party receiving the support (including legal proceedings) and non-recovery is not due to an error or negligence on its part - checks, reviews, audits and investigations by the EU (Article 25): will be conducted taking into account the rules and procedures specifically agreed between them and the framework agreement (if any) - impact evaluation (Article 26): will be conducted in accordance with the participant’s internal rules and procedures and the framework agreement (if any) - grant agreement suspension (Article 31): certain costs incurred during grant suspension are eligible (notably, minimum costs necessary for a possible resumption of the action and costs relating to contracts which were entered into before the pre-information letter was received and which could not reasonably be suspended, reallocated or terminated on legal grounds) - grant agreement termination (Article 32): the final grant amount and final payment will be calculated taking into account also costs relating to contracts due for execution only after termination takes effect, if the contract was entered into before the pre-information letter was received and could not reasonably be terminated on legal grounds - liability for damages (Article 33.2): the granting authority must be compensated for damage it sustains as a result of the implementation of the action or because the action was not implemented in full compliance with the Agreement only if the damage is due to an infringement of the participant’s internal rules and procedures or due to a violation of third parties’ rights by the participant or one of its employees or individual for whom the employees are responsible. Participants whose pillar assessment covers procurement and granting procedures may also do purchases, subcontracting and financial support to third parties (Article 6.2) in accordance with their internal rules and procedures for purchases, subcontracting and financial support. Participants whose pillar assessment covers data protection rules may rely on their internal standards, rules and procedures for data protection (Article 15). The participants may however not rely on provisions which would breach the principle of equal treatment of applicants or beneficiaries or call into question the decision awarding the grant, such as in particular: - eligibility (Article 6) - consortium roles and set-up (Articles 7-9) - security and ethics (Articles 13, 14) 20 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 - IPR (including background and results, access rights and rights of use), communication, dissemination and visibility (Articles 16 and 17) - information obligation (Article 19) - payment, reporting and amendments (Articles 21, 22 and 39) - rejections, reductions, suspensions and terminations (Articles 27, 28, 29-32) If the pillar assessment was subject to remedial measures, reliance on the internal systems, rules and procedures is subject to compliance with those remedial measures. Participants whose assessment has not yet been updated to cover (the new rules on) data protection may rely on their internal systems, rules and procedures, provided that they ensure that personal data is: - processed lawfully, fairly and in a transparent manner in relation to the data subject - collected for specified, explicit and legitimate purposes and not further processed in a manner that is incompatible with those purposes - adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed - accurate and, where necessary, kept up to date - kept in a form which permits identification of data subjects for no longer than is necessary for the purposes for which the data is processed and - processed in a manner that ensures appropriate security of the personal data. Participants must inform the coordinator without delay of any changes to the systems, rules and procedures that were part of the pillar assessment. The coordinator must immediately inform the granting authority. Pillar-assessed participants that have also concluded a framework agreement with the EU, may moreover — under the same conditions as those above (i.e. not call into question the decision awarding the grant or breach the principle of equal treatment of applicants or beneficiaries) — rely on provisions set out in that framework agreement. SECTION 2 RULES FOR CARRYING OUT THE ACTION ARTICLE 11 — PROPER IMPLEMENTATION OF THE ACTION 11.1 Obligation to properly implement the action The beneficiaries must implement the action as described in Annex 1 and in compliance with the provisions of the Agreement, the call conditions and all legal obligations under applicable EU, international and national law. 11.2 Consequences of non-compliance 21 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 12 — CONFLICT OF INTERESTS 12.1 Conflict of interests The beneficiaries must take all measures to prevent any situation where the impartial and objective implementation of the Agreement could be compromised for reasons involving family, emotional life, political or national affinity, economic interest or any other direct or indirect interest (‘conflict of interests’). They must formally notify the granting authority without delay of any situation constituting or likely to lead to a conflict of interests and immediately take all the necessary steps to rectify this situation. The granting authority may verify that the measures taken are appropriate and may require additional measures to be taken by a specified deadline. 12.2 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28) and the grant or the beneficiary may be terminated (see Article 32). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 13 — CONFIDENTIALITY AND SECURITY 13.1 Sensitive information The parties must keep confidential any data, documents or other material (in any form) that is identified as sensitive in writing (‘sensitive information’) — during the implementation of the action and for at least until the time-limit set out in the Data Sheet (see Point 6). If a beneficiary requests, the granting authority may agree to keep such information confidential for a longer period. Unless otherwise agreed between the parties, they may use sensitive information only to implement the Agreement. The beneficiaries may disclose sensitive information to their personnel or other participants involved in the action only if they: (a) need to know it in order to implement the Agreement and (b) are bound by an obligation of confidentiality. The granting authority may disclose sensitive information to its staff and to other EU institutions and bodies. It may moreover disclose sensitive information to third parties, if: 22 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 (a) this is necessary to implement the Agreement or safeguard the EU financial interests and (b) the recipients of the information are bound by an obligation of confidentiality. The confidentiality obligations no longer apply if: (a) the disclosing party agrees to release the other party (b) the information becomes publicly available, without breaching any confidentiality obligation (c) the disclosure of the sensitive information is required by EU, international or national law. Specific confidentiality rules (if any) are set out in Annex 5. 13.2 Classified information The parties must handle classified information in accordance with the applicable EU, international or national law on classified information (in particular, Decision 2015/44412 and its implementing rules). Deliverables which contain classified information must be submitted according to special procedures agreed with the granting authority. Action tasks involving classified information may be subcontracted only after explicit approval (in writing) from the granting authority. Classified information may not be disclosed to any third party (including participants involved in the action implementation) without prior explicit written approval from the granting authority. Specific security rules (if any) are set out in Annex 5. 13.3 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 14 — ETHICS AND VALUES 14.1 Ethics The action must be carried out in line with the highest ethical standards and the applicable EU, international and national law on ethical principles. Specific ethics rules (if any) are set out in Annex 5. 14.2 Values The beneficiaries must commit to and ensure the respect of basic EU values (such as respect for 12 Commission Decision 2015/444/EC, Euratom of 13 March 2015 on the security rules for protecting EU classified information (OJ L 72, 17.3.2015, p. 53). 23 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 human dignity, freedom, democracy, equality, the rule of law and human rights, including the rights of minorities). Specific rules on values (if any) are set out in Annex 5. 14.3 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 15 — DATA PROTECTION 15.1 Data processing by the granting authority Any personal data under the Agreement will be processed under the responsibility of the data controller of the granting authority in accordance with and for the purposes set out in the Portal Privacy Statement. For grants where the granting authority is the European Commission, an EU regulatory or executive agency, joint undertaking or other EU body, the processing will be subject to Regulation 2018/172513. 15.2 Data processing by the beneficiaries The beneficiaries must process personal data under the Agreement in compliance with the applicable EU, international and national law on data protection (in particular, Regulation 2016/67914). They must ensure that personal data is: - processed lawfully, fairly and in a transparent manner in relation to the data subjects - collected for specified, explicit and legitimate purposes and not further processed in a manner that is incompatible with those purposes - adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed - accurate and, where necessary, kept up to date - kept in a form which permits identification of data subjects for no longer than is necessary for the purposes for which the data is processed and - processed in a manner that ensures appropriate security of the data. 13 Regulation (EU) 2018/1725 of the European Parliament and of the Council of 23 October 2018 on the protection of natural persons with regard to the processing of personal data by the Union institutions, bodies, offices and agencies and on the free movement of such data, and repealing Regulation (EC) No 45/2001 and Decision No 1247/2002/EC (OJ L 295, 21.11.2018, p. 39). 14 Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (‘GDPR’) (OJ L 119, 4.5.2016, p. 1). 24 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The beneficiaries may grant their personnel access to personal data only if it is strictly necessary for implementing, managing and monitoring the Agreement. The beneficiaries must ensure that the personnel is under a confidentiality obligation. The beneficiaries must inform the persons whose data are transferred to the granting authority and provide them with the Portal Privacy Statement. 15.3 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 16 — INTELLECTUAL PROPERTY RIGHTS (IPR) — BACKGROUND AND RESULTS —ACCESS RIGHTS AND RIGHTS OF USE 16.1 Background and access rights to background The beneficiaries must give each other and the other participants access to the background identified as needed for implementing the action, subject to any specific rules in Annex 5. ‘Background’ means any data, know-how or information — whatever its form or nature (tangible or intangible), including any rights such as intellectual property rights — that is: (a) held by the beneficiaries before they acceded to the Agreement and (b) needed to implement the action or exploit the results. If background is subject to rights of a third party, the beneficiary concerned must ensure that it is able to comply with its obligations under the Agreement. 16.2 Ownership of results The granting authority does not obtain ownership of the results produced under the action. ‘Results’ means any tangible or intangible effect of the action, such as data, know-how or information, whatever its form or nature, whether or not it can be protected, as well as any rights attached to it, including intellectual property rights. 16.3 Rights of use of the granting authority on materials, documents and information received for policy, information, communication, dissemination and publicity purposes The granting authority has the right to use non-sensitive information relating to the action and materials and documents received from the beneficiaries (notably summaries for publication, deliverables, as well as any other material, such as pictures or audio-visual material, in paper or electronic form) for policy information, communication, dissemination and publicity purposes — during the action or afterwards. The right to use the beneficiaries’ materials, documents and information is granted in the form of a royalty-free, non-exclusive and irrevocable licence, which includes the following rights: 25 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 (a) use for its own purposes (in particular, making them available to persons working for the granting authority or any other EU service (including institutions, bodies, offices, agencies, etc.) or EU Member State institution or body; copying or reproducing them in whole or in part, in unlimited numbers; and communication through press information services) (b) distribution to the public (in particular, publication as hard copies and in electronic or digital format, publication on the internet, as a downloadable or non-downloadable file, broadcasting by any channel, public display or presentation, communicating through press information services, or inclusion in widely accessible databases or indexes) (c) editing or redrafting (including shortening, summarising, inserting other elements (e.g. meta-data, legends, other graphic, visual, audio or text elements), extracting parts (e.g. audio or video files), dividing into parts, use in a compilation) (d) translation (e) storage in paper, electronic or other form (f) archiving, in line with applicable document-management rules (g) the right to authorise third parties to act on its behalf or sub-license to third parties the modes of use set out in Points (b), (c), (d) and (f), if needed for the information, communication and publicity activity of the granting authority and (h) processing, analysing, aggregating the materials, documents and information received and producing derivative works. The rights of use are granted for the whole duration of the industrial or intellectual property rights concerned. If materials or documents are subject to moral rights or third party rights (including intellectual property rights or rights of natural persons on their image and voice), the beneficiaries must ensure that they comply with their obligations under this Agreement (in particular, by obtaining the necessary licences and authorisations from the rights holders concerned). Where applicable, the granting authority will insert the following information: “© – [year] – [name of the copyright owner]. All rights reserved. Licensed to the [name of granting authority] under conditions.” 16.4 Specific rules on IPR, results and background Specific rules regarding intellectual property rights, results and background (if any) are set out in Annex 5. 16.5 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such a breach may also lead to other measures described in Chapter 5. 26 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 ARTICLE 17 — COMMUNICATION, DISSEMINATION AND VISIBILITY 17.1 Communication — Dissemination — Promoting the action Unless otherwise agreed with the granting authority, the beneficiaries must promote the action and its results by providing targeted information to multiple audiences (including the media and the public), in accordance with Annex 1 and in a strategic, coherent and effective manner. Before engaging in a communication or dissemination activity expected to have a major media impact, the beneficiaries must inform the granting authority. 17.2 Visibility — European flag and funding statement Unless otherwise agreed with the granting authority, communication activities of the beneficiaries related to the action (including media relations, conferences, seminars, information material, such as brochures, leaflets, posters, presentations, etc., in electronic form, via traditional or social media, etc.), dissemination activities and any infrastructure, equipment, vehicles, supplies or major result funded by the grant must acknowledge the EU support and display the European flag (emblem) and funding statement (translated into local languages, where appropriate): The emblem must remain distinct and separate and cannot be modified by adding other visual marks, brands or text. Apart from the emblem, no other visual identity or logo may be used to highlight the EU support. When displayed in association with other logos (e.g. of beneficiaries or sponsors), the emblem must be displayed at least as prominently and visibly as the other logos. For the purposes of their obligations under this Article, the beneficiaries may use the emblem without first obtaining approval from the granting authority. This does not, however, give them the right to 27 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 exclusive use. Moreover, they may not appropriate the emblem or any similar trademark or logo, either by registration or by any other means. 17.3 Quality of information — Disclaimer Any communication or dissemination activity related to the action must use factually accurate information. Moreover, it must indicate the following disclaimer (translated into local languages where appropriate): “Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or [name of the granting authority]. Neither the European Union nor the granting authority can be held responsible for them.” 17.4 Specific communication, dissemination and visibility rules Specific communication, dissemination and visibility rules (if any) are set out in Annex 5. 17.5 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 18 — SPECIFIC RULES FOR CARRYING OUT THE ACTION Not applicable SECTION 3 GRANT ADMINISTRATION ARTICLE 19 — GENERAL INFORMATION OBLIGATIONS 19.1 Information requests The beneficiaries must provide — during the action or afterwards and in accordance with Article 7 — any information requested in order to verify eligibility of the lump sum contributions declared, proper implementation of the action and compliance with the other obligations under the Agreement. The information provided must be accurate, precise and complete and in the format requested, including electronic format. 19.2 Participant Register data updates The beneficiaries must keep — at all times, during the action or afterwards — their information stored in the Portal Participant Register up to date, in particular, their name, address, legal representatives, legal form and organisation type. 19.3 Information about events and circumstances which impact the action 28 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The beneficiaries must immediately inform the granting authority (and the other beneficiaries) of any of the following: (a) events which are likely to affect or delay the implementation of the action or affect the EU’s financial interests, in particular: (i) changes in their legal, financial, technical, organisational or ownership situation (including changes linked to one of the exclusion grounds listed in the declaration of honour signed before grant signature) (ii) linked action information: not applicable (b) circumstances affecting: (i) the decision to award the grant or (ii) compliance with requirements under the Agreement. 19.4 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 20 — RECORD-KEEPING 20.1 Keeping records and supporting documents The beneficiaries must — at least until the time-limit set out in the Data Sheet (see Point 6) — keep records and other supporting documents to prove the proper implementation of the action (proper implementation of the work and/or achievement of the results as described in Annex 1) in line with the accepted standards in the respective field (if any); beneficiaries do not need to keep specific records on the actual costs incurred. The records and supporting documents must be made available upon request (see Article 19) or in the context of checks, reviews, audits or investigations (see Article 25). If there are on-going checks, reviews, audits, investigations, litigation or other pursuits of claims under the Agreement (including the extension of findings; see Article 25), the beneficiaries must keep these records and other supporting documentation until the end of these procedures. The beneficiaries must keep the original documents. Digital and digitalised documents are considered originals if they are authorised by the applicable national law. The granting authority may accept non-original documents if they offer a comparable level of assurance. 20.2 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, lump sum contributions insufficiently substantiated will be ineligible (see Article 6) and will be rejected (see Article 27), and the grant may be reduced (see Article 28). 29 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Such breaches may also lead to other measures described in Chapter 5. ARTICLE 21 — REPORTING 21.1 Continuous reporting The beneficiaries must continuously report on the progress of the action (e.g. deliverables, milestones, outputs/outcomes, critical risks, indicators, etc; if any), in the Portal Continuous Reporting tool and in accordance with the timing and conditions it sets out (as agreed with the granting authority). Standardised deliverables (e.g. progress reports not linked to payments, reports on cumulative expenditure, special reports, etc; if any) must be submitted using the templates published on the Portal. 21.2 Periodic reporting: Technical reports and financial statements In addition, the beneficiaries must provide reports to request payments, in accordance with the schedule and modalities set out in the Data Sheet (see Point 4.2): - for additional prefinancings (if any): an additional prefinancing report - for interim payments (if any) and the final payment: a periodic report The prefinancing and periodic reports include a technical and financial part. The technical part includes an overview of the action implementation. It must be prepared using the template available in the Portal Periodic Reporting tool. The financial part of the additional prefinancing report includes a statement on the use of the previous prefinancing payment. The financial part of the periodic report includes: - the financial statement (consolidated statement for the consortium) - the explanation on the use of resources (or detailed cost reporting table): not applicable - the certificates on the financial statements (CFS): not applicable. The financial statement must contain the lump sum contributions indicated in Annex 2, for the work packages that were completed during the reporting period. For the last reporting period, the beneficiaries may exceptionally also declare partial lump sum contributions for work packages that were not completed (e.g. due to force majeure or technical impossibility). Lump sum contributions which are not declared in a financial statement will not be taken into account by the granting authority. By signing the financial statement (directly in the Portal Periodic Reporting tool), the coordinator confirms (on behalf of the consortium) that: - the information provided is complete, reliable and true 30 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 - the lump sum contributions declared are eligible (in particular, the work packages have been completed, that the work has been properly implemented and/or the results were achieved in accordance with Annex 1; see Article 6) - the proper implementation and/or achievement can be substantiated by adequate records and supporting documents (see Article 20) that will be produced upon request (see Article 19) or in the context of checks, reviews, audits and investigations (see Article 25). In case of recoveries (see Article 22), beneficiaries will be held responsible also for the lump sum contributions declared for their affiliated entities (if any). 21.3 Currency for financial statements and conversion into euros The financial statements must be drafted in euro. 21.4 Reporting language The reporting must be in the language of the Agreement, unless otherwise agreed with the granting authority (see Data Sheet, Point 4.2). 21.5 Consequences of non-compliance If a report submitted does not comply with this Article, the granting authority may suspend the payment deadline (see Article 29) and apply other measures described in Chapter 5. If the coordinator breaches its reporting obligations, the granting authority may terminate the grant or the coordinator’s participation (see Article 32) or apply other measures described in Chapter 5. ARTICLE 22 — PAYMENTS AND RECOVERIES — CALCULATION OF AMOUNTS DUE 22.1 Payments and payment arrangements Payments will be made in accordance with the schedule and modalities set out in the Data Sheet (see Point 4.2). They will be made in euro to the bank account indicated by the coordinator (see Data Sheet, Point 4.2) and must be distributed without unjustified delay (restrictions may apply to distribution of the initial prefinancing payment; see Data Sheet, Point 4.2). Payments to this bank account will discharge the granting authority from its payment obligation. The cost of payment transfers will be borne as follows: - the granting authority bears the cost of transfers charged by its bank - the beneficiary bears the cost of transfers charged by its bank - the party causing a repetition of a transfer bears all costs of the repeated transfer. Payments by the granting authority will be considered to have been carried out on the date when they are debited to its account. 31 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 22.2 Recoveries Recoveries will be made, if — at beneficiary termination, final payment or afterwards — it turns out that the granting authority has paid too much and needs to recover the amounts undue. The general liability regime for recoveries (first-line liability) is as follows: At final payment, the coordinator will be fully liable for recoveries, even if it has not been the final recipient of the undue amounts. At beneficiary termination or after final payment, recoveries will be made directly against the beneficiaries concerned. Beneficiaries will be fully liable for repaying the debts of their affiliated entities. In case of enforced recoveries (see Article 22.4): - the beneficiaries will be jointly and severally liable for repaying debts of another beneficiary under the Agreement (including late-payment interest), if required by the granting authority (see Data Sheet, Point 4.4) - affiliated entities will be held liable for repaying debts of their beneficiaries under the Agreement (including late-payment interest), if required by the granting authority (see Data Sheet, Point 4.4). 22.3 Amounts due 22.3.1 Prefinancing payments The aim of the prefinancing is to provide the beneficiaries with a float. It remains the property of the EU until the final payment. For initial prefinancings (if any), the amount due, schedule and modalities are set out in the Data Sheet (see Point 4.2). For additional prefinancings (if any), the amount due, schedule and modalities are also set out in the Data Sheet (see Point 4.2). However, if the statement on the use of the previous prefinancing payment shows that less than 70% was used, the amount set out in the Data Sheet will be reduced by the difference between the 70% threshold and the amount used. Prefinancing payments (or parts of them) may be offset (without the beneficiaries’ consent) against amounts owed by a beneficiary to the granting authority — up to the amount due to that beneficiary. For grants where the granting authority is the European Commission or an EU executive agency, offsetting may also be done against amounts owed to other Commission services or executive agencies. Payments will not be made if the payment deadline or payments are suspended (see Articles 29 and 30). 22.3.2 Amount due at beneficiary termination — Recovery In case of beneficiary termination, the granting authority will determine the provisional amount due for the beneficiary concerned. 32 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 This will be done on the basis of work packages already completed in previous interim payments. Payments for ongoing/not yet completed work packages which the beneficiary was working on before termination (if any) will therefore be made only later on, with the next interim or final payments when those work packages have been completed. The amount due will be calculated in the following step: Step 1 — Calculation of the total accepted EU contribution Step 1 — Calculation of the total accepted EU contribution The granting authority will first calculate the ‘accepted EU contribution’ for the beneficiary, on the basis of the beneficiary’s lump sum contributions for the work packages which were approved in previous interim payments. After that, the granting authority will take into account grant reductions (if any). The resulting amount is the ‘total accepted EU contribution’ for the beneficiary. The balance is then calculated by deducting the payments received (if any; see report on the distribution of payments in Article 32), from the total accepted EU contribution: {total accepted EU contribution for the beneficiary minus {prefinancing and interim payments received (if any)}}. If the balance is negative, it will be recovered in accordance with the following procedure: The granting authority will send a pre-information letter to the beneficiary concerned: - formally notifying the intention to recover, the amount due, the amount to be recovered and the reasons why and - requesting observations within 30 days of receiving notification. If no observations are submitted (or the granting authority decides to pursue recovery despite the observations it has received), it will confirm the amount to be recovered and ask this amount to be paid to the coordinator (confirmation letter). 22.3.3 Interim payments Interim payments reimburse the eligible lump sum contributions claimed for work packages implemented during the reporting periods (if any). Interim payments (if any) will be made in accordance with the schedule and modalities set out the Data Sheet (see Point 4.2). Payment is subject to the approval of the periodic report and the work packages declared. Their approval does not imply recognition of compliance, authenticity, completeness or correctness of their content. Incomplete work packages and work packages that have not been delivered or cannot be approved will be rejected (see Article 27). 33 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The interim payment will be calculated by the granting authority in the following steps: Step 1 — Calculation of the total accepted EU contribution Step 2 — Limit to the interim payment ceiling Step 1 — Calculation of the total accepted EU contribution The granting authority will first calculate the ‘accepted EU contribution’ for the action for the reporting period, by calculating the lump sum contributions for the approved work packages. After that, the granting authority will take into account grant reductions from beneficiary termination (if any). The resulting amount is the ‘total accepted EU contribution’. Step 2 — Limit to the interim payment ceiling The resulting amount is then capped to ensure that the total amount of prefinancing and interim payments (if any) does not exceed the interim payment ceiling set out in the Data Sheet (see Point 4.2). Interim payments (or parts of them) may be offset (without the beneficiaries’ consent) against amounts owed by a beneficiary to the granting authority — up to the amount due to that beneficiary. For grants where the granting authority is the European Commission or an EU executive agency, offsetting may also be done against amounts owed to other Commission services or executive agencies. Payments will not be made if the payment deadline or payments are suspended (see Articles 29 and 30). 22.3.4 Final payment — Final grant amount — Revenues and Profit — Recovery The final payment (payment of the balance) reimburses the remaining eligible lump sum contributions claimed for the implemented work packages (if any). The final payment will be made in accordance with the schedule and modalities set out in the Data Sheet (see Point 4.2). Payment is subject to the approval of the final periodic report and the work packages declared. Their approval does not imply recognition of compliance, authenticity, completeness or correctness of their content. Work packages (or parts of them) that have not been delivered or cannot be approved will be rejected (see Article 27). The final grant amount for the action will be calculated in the following steps: Step 1 — Calculation of the total accepted EU contribution Step 2 — Limit to the maximum grant amount Step 3 — Reduction due to the no-profit rule Step 1 — Calculation of the total accepted EU contribution 34 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The granting authority will first calculate the ‘accepted EU contribution’ for the action for all reporting periods, by calculating the lump sum contributions for the approved work packages. After that, the granting authority will take into account grant reductions (if any). The resulting amount is the ‘total accepted EU contribution’. Step 2 — Limit to the maximum grant amount Not applicable Step 3 — Reduction due to the no-profit rule Not applicable The balance (final payment) is then calculated by deducting the total amount of prefinancing and interim payments already made (if any), from the final grant amount: {final grant amount minus {prefinancing and interim payments made (if any)}}. If the balance is positive, it will be paid to the coordinator. The final payment (or part of it) may be offset (without the beneficiaries’ consent) against amounts owed by a beneficiary to the granting authority — up to the amount due to that beneficiary. For grants where the granting authority is the European Commission or an EU executive agency, offsetting may also be done against amounts owed to other Commission services or executive agencies. Payments will not be made if the payment deadline or payments are suspended (see Articles 29 and 30). If the balance is negative, it will be recovered in accordance with the following procedure: The granting authority will send a pre-information letter to the coordinator: - formally notifying the intention to recover, the final grant amount, the amount to be recovered and the reasons why - requesting observations within 30 days of receiving notification. If no observations are submitted (or the granting authority decides to pursue recovery despite the observations it has received), it will confirm the amount to be recovered (confirmation letter), together with a debit note with the terms and date for payment. If payment is not made by the date specified in the debit note, the granting authority will enforce recovery in accordance with Article 22.4. 22.3.5 Audit implementation after final payment — Revised final grant amount — Recovery If — after the final payment (in particular, after checks, reviews, audits or investigations; see 35 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Article 25) — the granting authority rejects lump sum contributions (see Article 27) or reduces the grant (see Article 28), it will calculate the revised final grant amount for the beneficiary concerned. The beneficiary revised final grant amount will be calculated in the following step: Step 1 — Calculation of the revised total accepted EU contribution Step 1 — Calculation of the revised total accepted EU contribution The granting authority will first calculate the ‘revised accepted EU contribution’ for the beneficiary, by calculating the ‘revised accepted contributions’. After that, it will take into account grant reductions (if any). The resulting ‘revised total accepted EU contribution’ is the beneficiary revised final grant amount. If the revised final grant amount is lower than the beneficiary’s final grant amount (i.e. its share in the final grant amount for the action), it will be recovered in accordance with the following procedure: The beneficiary final grant amount (i.e. share in the final grant amount for the action) is calculated as follows: {{total accepted EU contribution for the beneficiary divided by total accepted EU contribution for the action} multiplied by final grant amount for the action}. The granting authority will send a pre-information letter to the beneficiary concerned: - formally notifying the intention to recover, the amount to be recovered and the reasons why and - requesting observations within 30 days of receiving notification. If no observations are submitted (or the granting authority decides to pursue recovery despite the observations it has received), it will confirm the amount to be recovered (confirmation letter), together with a debit note with the terms and the date for payment. Recoveries against affiliated entities (if any) will be handled through their beneficiaries. If payment is not made by the date specified in the debit note, the granting authority will enforce recovery in accordance with Article 22.4. 22.4 Enforced recovery If payment is not made by the date specified in the debit note, the amount due will be recovered: (a) by offsetting the amount — without the coordinator or beneficiary’s consent — against any amounts owed to the coordinator or beneficiary by the granting authority. In exceptional circumstances, to safeguard the EU financial interests, the amount may be offset before the payment date specified in the debit note. 36 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 For grants where the granting authority is the European Commission or an EU executive agency, debts may also be offset against amounts owed by other Commission services or executive agencies. (b) by drawing on the financial guarantee(s) (if any) (c) by holding other beneficiaries jointly and severally liable (if any; see Data Sheet, Point 4.4) (d) by holding affiliated entities jointly and severally liable (if any, see Data Sheet, Point 4.4) (e) by taking legal action (see Article 43) or, provided that the granting authority is the European Commission or an EU executive agency, by adopting an enforceable decision under Article 299 of the Treaty on the Functioning of the EU (TFEU) and Article 100(2) of EU Financial Regulation 2018/1046. The amount to be recovered will be increased by late-payment interest at the rate set out in Article 23.5, from the day following the payment date in the debit note, up to and including the date the full payment is received. Partial payments will be first credited against expenses, charges and late-payment interest and then against the principal. Bank charges incurred in the recovery process will be borne by the beneficiary, unless Directive 2015/236615 applies. For grants where the granting authority is an EU executive agency, enforced recovery by offsetting or enforceable decision will be done by the services of the European Commission (see also Article 43). 22.5 Consequences of non-compliance 22.5.1 If the granting authority does not pay within the payment deadlines (see above), the beneficiaries are entitled to late-payment interest at the reference rate applied by the European Central Bank (ECB) for its main refinancing operations in euros, plus the percentage specified in the Data Sheet (Point 4.2). The ECB reference rate to be used is the rate in force on the first day of the month in which the payment deadline expires, as published in the C series of the Official Journal of the European Union. If the late-payment interest is lower than or equal to EUR 200, it will be paid to the coordinator only on request submitted within two months of receiving the late payment. Late-payment interest is not due if all beneficiaries are EU Member States (including regional and local government authorities or other public bodies acting on behalf of a Member State for the purpose of this Agreement). If payments or the payment deadline are suspended (see Articles 29 and 30), payment will not be considered as late. 15 Directive (EU) 2015/2366 of the European Parliament and of the Council of 25 November 2015 on payment services in the internal market, amending Directives 2002/65/EC, 2009/110/EC and 2013/36/EU and Regulation (EU) No 1093/2010, and repealing Directive 2007/64/EC (OJ L 337, 23.12.2015, p. 35). 37 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 Late-payment interest covers the period running from the day following the due date for payment (see above), up to and including the date of payment. Late-payment interest is not considered for the purposes of calculating the final grant amount. 22.5.2 If the coordinator breaches any of its obligations under this Article, the grant may be reduced (see Article 28) and the grant or the coordinator may be terminated (see Article 32). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 23 — GUARANTEES 23.1 Prefinancing guarantee If required by the granting authority (see Data Sheet, Point 4.2), the beneficiaries must provide (one or more) prefinancing guarantee(s) in accordance with the timing and the amounts set out in the Data Sheet. The coordinator must submit them to the granting authority in due time before the prefinancing they are linked to. The guarantees must be drawn up using the template published on the Portal and fulfil the following conditions: (a) be provided by a bank or approved financial institution established in the EU or — if requested by the coordinator and accepted by the granting authority — by a third party or a bank or financial institution established outside the EU offering equivalent security (b) the guarantor stands as first-call guarantor and does not require the granting authority to first have recourse against the principal debtor (i.e. the beneficiary concerned) and (c) remain explicitly in force until the final payment and, if the final payment takes the form of a recovery, until five months after the debit note is notified to a beneficiary. They will be released within the following month. 23.2 Consequences of non-compliance If the beneficiaries breach their obligation to provide the prefinancing guarantee, the prefinancing will not be paid. Such breaches may also lead to other measures described in Chapter 5. ARTICLE 24 — CERTIFICATES Not applicable ARTICLE 25 — CHECKS, REVIEWS, AUDITS AND INVESTIGATIONS — EXTENSION OF FINDINGS 25.1 Granting authority checks, reviews and audits 38 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 25.1.1 Internal checks The granting authority may — during the action or afterwards — check the proper implementation of the action and compliance with the obligations under the Agreement, including assessing lump sum contributions, deliverables and reports. 25.1.2 Project reviews The granting authority may carry out reviews on the proper implementation of the action and compliance with the obligations under the Agreement (general project reviews or specific issues reviews). Such project reviews may be started during the implementation of the action and until the time-limit set out in the Data Sheet (see Point 6). They will be formally notified to the coordinator or beneficiary concerned and will be considered to start on the date of the notification. If needed, the granting authority may be assisted by independent, outside experts. If it uses outside experts, the coordinator or beneficiary concerned will be informed and have the right to object on grounds of commercial confidentiality or conflict of interest. The coordinator or beneficiary concerned must cooperate diligently and provide — within the deadline requested — any information and data in addition to deliverables and reports already submitted. The granting authority may request beneficiaries to provide such information to it directly. Sensitive information and documents will be treated in accordance with Article 13. The coordinator or beneficiary concerned may be requested to participate in meetings, including with the outside experts. For on-the-spot visits, the beneficiary concerned must allow access to sites and premises (including to the outside experts) and must ensure that information requested is readily available. Information provided must be accurate, precise and complete and in the format requested, including electronic format. On the basis of the review findings, a project review report will be drawn up. The granting authority will formally notify the project review report to the coordinator or beneficiary concerned, which has 30 days from receiving notification to make observations. Project reviews (including project review reports) will be in the language of the Agreement, unless otherwise agreed with the granting authority (see Data Sheet, Point 4.2). 25.1.3 Audits The granting authority may carry out audits on the proper implementation of the action and compliance with the obligations under the Agreement. Such audits may be started during the implementation of the action and until the time-limit set out in the Data Sheet (see Point 6). They will be formally notified to the beneficiary concerned and will be considered to start on the date of the notification. The granting authority may use its own audit service, delegate audits to a centralised service or use 39 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 external audit firms. If it uses an external firm, the beneficiary concerned will be informed and have the right to object on grounds of commercial confidentiality or conflict of interest. The beneficiary concerned must cooperate diligently and provide — within the deadline requested — any information (including complete accounts, individual salary statements or other personal data) to verify compliance with the Agreement. Sensitive information and documents will be treated in accordance with Article 13. For on-the-spot visits, the beneficiary concerned must allow access to sites and premises (including for the external audit firm) and must ensure that information requested is readily available. Information provided must be accurate, precise and complete and in the format requested, including electronic format. On the basis of the audit findings, a draft audit report will be drawn up. The auditors will formally notify the draft audit report to the beneficiary concerned, which has 30 days from receiving notification to make observations (contradictory audit procedure). The final audit report will take into account observations by the beneficiary concerned and will be formally notified to them. Audits (including audit reports) will be in the language of the Agreement, unless otherwise agreed with the granting authority (see Data Sheet, Point 4.2). 25.2 European Commission checks, reviews and audits in grants of other granting authorities Where the granting authority is not the European Commission, the latter has the same rights of checks, reviews and audits as the granting authority. 25.3 Access to records for assessing simplified forms of funding The beneficiaries must give the European Commission access to their statutory records for the periodic assessment of simplified forms of funding which are used in EU programmes. 25.4 OLAF, EPPO and ECA audits and investigations The following bodies may also carry out checks, reviews, audits and investigations — during the action or afterwards: - the European Anti-Fraud Office (OLAF) under Regulations No 883/201316 and No 2185/9617 - the European Public Prosecutor’s Office (EPPO) under Regulation 2017/1939 16 Regulation (EU, Euratom) No 883/2013 of the European Parliament and of the Council of 11 September 2013 concerning investigations conducted by the European Anti-Fraud Office (OLAF) and repealing Regulation (EC) No 1073/1999 of the European Parliament and of the Council and Council Regulation (Euratom) No 1074/1999 (OJ L 248, 18/09/2013, p. 1). 17 Council Regulation (Euratom, EC) No 2185/96 of 11 November 1996 concerning on-the-spot checks and inspections carried out by the Commission in order to protect the European Communities' financial interests against fraud and other irregularities (OJ L 292, 15/11/1996, p. 2). 40 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 - the European Court of Auditors (ECA) under Article 287 of the Treaty on the Functioning of the EU (TFEU) and Article 257 of EU Financial Regulation 2018/1046. If requested by these bodies, the beneficiary concerned must provide full, accurate and complete information in the format requested (including complete accounts, individual salary statements or other personal data, including in electronic format) and allow access to sites and premises for on-the-spot visits or inspections — as provided for under these Regulations. To this end, the beneficiary concerned must keep all relevant information relating to the action, at least until the time-limit set out in the Data Sheet (Point 6) and, in any case, until any ongoing checks, reviews, audits, investigations, litigation or other pursuits of claims have been concluded. 25.5 Consequences of checks, reviews, audits and investigations — Extension of findings 25.5.1 Consequences of checks, reviews, audits and investigations in this grant Findings in checks, reviews, audits or investigations carried out in the context of this grant may lead to rejections (see Article 27), grant reduction (see Article 28) or other measures described in Chapter 5. Rejections or grant reductions after the final payment will lead to a revised final grant amount (see Article 22). Findings in checks, reviews, audits or investigations during the action implementation may lead to a request for amendment (see Article 39), to change the description of the action set out in Annex 1. Checks, reviews, audits or investigations that find systemic or recurrent errors, irregularities, fraud or breach of obligations in any EU grant may also lead to consequences in other EU grants awarded under similar conditions (‘extension to other grants’). Moreover, findings arising from an OLAF or EPPO investigation may lead to criminal prosecution under national law. 25.5.2 Extension from other grants Findings of checks, reviews, audits or investigations in other grants may be extended to this grant, if: (a) the beneficiary concerned is found, in other EU grants awarded under similar conditions, to have committed systemic or recurrent errors, irregularities, fraud or breach of obligations that have a material impact on this grant and (b) those findings are formally notified to the beneficiary concerned — together with the list of grants affected by the findings — within the time-limit for audits set out in the Data Sheet (see Point 6). The granting authority will formally notify the beneficiary concerned of the intention to extend the findings and the list of grants affected. If the extension concerns rejections of lump sum contributions: the notification will include: (a) an invitation to submit observations on the list of grants affected by the findings (b) the request to submit revised financial statements for all grants affected 41 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 (c) the correction rate for extrapolation, established on the basis of the systemic or recurrent errors, to calculate the amounts to be rejected, if the beneficiary concerned: (i) considers that the submission of revised financial statements is not possible or practicable or (ii) does not submit revised financial statements. If the extension concerns grant reductions: the notification will include: (a) an invitation to submit observations on the list of grants affected by the findings and (b) the correction rate for extrapolation, established on the basis of the systemic or recurrent errors and the principle of proportionality. The beneficiary concerned has 60 days from receiving notification to submit observations, revised financial statements or to propose a duly substantiated alternative correction method/rate. On the basis of this, the granting authority will analyse the impact and decide on the implementation (i.e. start rejection or grant reduction procedures, either on the basis of the revised financial statements or the announced/alternative method/rate or a mix of those; see Articles 27 and 28). 25.6 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, lump sum contributions insufficiently substantiated will be ineligible (see Article 6) and will be rejected (see Article 27), and the grant may be reduced (see Article 28). Such breaches may also lead to other measures described in Chapter 5. ARTICLE 26 — IMPACT EVALUATIONS 26.1 Impact evaluation The granting authority may carry out impact evaluations of the action, measured against the objectives and indicators of the EU programme funding the grant. Such evaluations may be started during implementation of the action and until the time-limit set out in the Data Sheet (see Point 6). They will be formally notified to the coordinator or beneficiaries and will be considered to start on the date of the notification. If needed, the granting authority may be assisted by independent outside experts. The coordinator or beneficiaries must provide any information relevant to evaluate the impact of the action, including information in electronic format. 26.2 Consequences of non-compliance If a beneficiary breaches any of its obligations under this Article, the granting authority may apply the measures described in Chapter 5. 42 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 CHAPTER 5 CONSEQUENCES OF NON-COMPLIANCE SECTION 1 REJECTIONS AND GRANT REDUCTION ARTICLE 27 — REJECTION OF CONTRIBUTIONS 27.1 Conditions The granting authority will — at interim payment, final payment or afterwards — reject any lump sum contributions which are ineligible (see Article 6), in particular following checks, reviews, audits or investigations (see Article 25). The rejection may also be based on the extension of findings from other grants to this grant (see Article 25). Ineligible lump sum contributions will be rejected. 27.2 Procedure If the rejection does not lead to a recovery, the granting authority will formally notify the coordinator or beneficiary concerned of the rejection, the amounts and the reasons why. The coordinator or beneficiary concerned may — within 30 days of receiving notification — submit observations if it disagrees with the rejection (payment review procedure). If the rejection leads to a recovery, the granting authority will follow the contradictory procedure with pre-information letter set out in Article 22. 27.3 Effects If the granting authority rejects lump sum contributions, it will deduct them from the lump sum contributions declared and then calculate the amount due (and, if needed, make a recovery; see Article 22). ARTICLE 28 — GRANT REDUCTION 28.1 Conditions The granting authority may — at beneficiary termination, final payment or afterwards — reduce the grant for a beneficiary, if: (a) the beneficiary (or a person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed: (i) substantial errors, irregularities or fraud or (ii) serious breach of obligations under this Agreement or during its award (including improper implementation of the action, non-compliance with the call conditions, submission of false information, failure to provide required information, breach of ethics or security rules (if applicable), etc.), or (b) the beneficiary (or a person having powers of representation, decision-making or control, or 43 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 person essential for the award/implementation of the grant) has committed — in other EU grants awarded to it under similar conditions — systemic or recurrent errors, irregularities, fraud or serious breach of obligations that have a material impact on this grant (extension of findings; see Article 25.5). The amount of the reduction will be calculated for each beneficiary concerned and proportionate to the seriousness and the duration of the errors, irregularities or fraud or breach of obligations, by applying an individual reduction rate to their accepted EU contribution. 28.2 Procedure If the grant reduction does not lead to a recovery, the granting authority will formally notify the coordinator or beneficiary concerned of the reduction, the amount to be reduced and the reasons why. The coordinator or beneficiary concerned may — within 30 days of receiving notification — submit observations if it disagrees with the reduction (payment review procedure). If the grant reduction leads to a recovery, the granting authority will follow the contradictory procedure with pre-information letter set out in Article 22. 28.3 Effects If the granting authority reduces the grant, it will deduct the reduction and then calculate the amount due (and, if needed, make a recovery; see Article 22). SECTION 2 SUSPENSION AND TERMINATION ARTICLE 29 — PAYMENT DEADLINE SUSPENSION 29.1 Conditions The granting authority may — at any moment — suspend the payment deadline if a payment cannot be processed because: (a) the required report (see Article 21) has not been submitted or is not complete or additional information is needed (b) there are doubts about the amount to be paid (e.g. ongoing extension procedure, queries about eligibility, need for a grant reduction, etc.) and additional checks, reviews, audits or investigations are necessary, or (c) there are other issues affecting the EU financial interests. 29.2 Procedure The granting authority will formally notify the coordinator of the suspension and the reasons why. The suspension will take effect the day the notification is sent. If the conditions for suspending the payment deadline are no longer met, the suspension will be lifted — and the remaining time to pay (see Data Sheet, Point 4.2) will resume. 44 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 If the suspension exceeds two months, the coordinator may request the granting authority to confirm if the suspension will continue. If the payment deadline has been suspended due to the non-compliance of the report and the revised report is not submitted (or was submitted but is also rejected), the granting authority may also terminate the grant or the participation of the coordinator (see Article 32). ARTICLE 30 — PAYMENT SUSPENSION 30.1 Conditions The granting authority may — at any moment — suspend payments, in whole or in part for one or more beneficiaries, if: (a) a beneficiary (or a person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed or is suspected of having committed: (i) substantial errors, irregularities or fraud or (ii) serious breach of obligations under this Agreement or during its award (including improper implementation of the action, non-compliance with the call conditions, submission of false information, failure to provide required information, breach of ethics or security rules (if applicable), etc.), or (b) a beneficiary (or a person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed — in other EU grants awarded to it under similar conditions — systemic or recurrent errors, irregularities, fraud or serious breach of obligations that have a material impact on this grant (extension of findings; see Article 25.5). If payments are suspended for one or more beneficiaries, the granting authority will make partial payment(s) for the part(s) not suspended. If suspension concerns the final payment, the payment (or recovery) of the remaining amount after suspension is lifted will be considered to be the payment that closes the action. 30.2 Procedure Before suspending payments, the granting authority will send a pre-information letter to the beneficiary concerned: - formally notifying the intention to suspend payments and the reasons why and - requesting observations within 30 days of receiving notification. If the granting authority does not receive observations or decides to pursue the procedure despite the observations it has received, it will confirm the suspension (confirmation letter). Otherwise, it will formally notify that the procedure is discontinued. At the end of the suspension procedure, the granting authority will also inform the coordinator. The suspension will take effect the day after the confirmation notification is sent. 45 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 If the conditions for resuming payments are met, the suspension will be lifted. The granting authority will formally notify the beneficiary concerned (and the coordinator) and set the suspension end date. During the suspension, no prefinancing will be paid to the beneficiaries concerned. For interim payments, the periodic reports for all reporting periods except the last one (see Article 21) must not contain any financial statements from the beneficiary concerned (or its affiliated entities). The coordinator must include them in the next periodic report after the suspension is lifted or — if suspension is not lifted before the end of the action — in the last periodic report. ARTICLE 31 — GRANT AGREEMENT SUSPENSION 31.1 Consortium-requested GA suspension 31.1.1 Conditions and procedure The beneficiaries may request the suspension of the grant or any part of it, if exceptional circumstances — in particular force majeure (see Article 35) — make implementation impossible or excessively difficult. The coordinator must submit a request for amendment (see Article 39), with: - the reasons why - the date the suspension takes effect; this date may be before the date of the submission of the amendment request and - the expected date of resumption. The suspension will take effect on the day specified in the amendment. Once circumstances allow for implementation to resume, the coordinator must immediately request another amendment of the Agreement to set the suspension end date, the resumption date (one day after suspension end date), extend the duration and make other changes necessary to adapt the action to the new situation (see Article 39) — unless the grant has been terminated (see Article 32). The suspension will be lifted with effect from the suspension end date set out in the amendment. This date may be before the date of the submission of the amendment request. During the suspension, no prefinancing will be paid. Moreover, no work may be done. Ongoing work packages must be interrupted and no new work packages may be started. 31.2 EU-initiated GA suspension 31.2.1 Conditions The granting authority may suspend the grant or any part of it, if: (a) a beneficiary (or a person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed or is suspected of having committed: (i) substantial errors, irregularities or fraud or 46 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 (ii) serious breach of obligations under this Agreement or during its award (including improper implementation of the action, non-compliance with the call conditions, submission of false information, failure to provide required information, breach of ethics or security rules (if applicable), etc.), or (b) a beneficiary (or a person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed — in other EU grants awarded to it under similar conditions — systemic or recurrent errors, irregularities, fraud or serious breach of obligations that have a material impact on this grant (extension of findings; see Article 25.5) (c) other: (i) linked action issues: not applicable (ii) additional GA suspension grounds: not applicable. 31.2.2 Procedure Before suspending the grant, the granting authority will send a pre-information letter to the coordinator: - formally notifying the intention to suspend the grant and the reasons why and - requesting observations within 30 days of receiving notification. If the granting authority does not receive observations or decides to pursue the procedure despite the observations it has received, it will confirm the suspension (confirmation letter). Otherwise, it will formally notify that the procedure is discontinued. The suspension will take effect the day after the confirmation notification is sent (or on a later date specified in the notification). Once the conditions for resuming implementation of the action are met, the granting authority will formally notify the coordinator a lifting of suspension letter, in which it will set the suspension end date and invite the coordinator to request an amendment of the Agreement to set the resumption date (one day after suspension end date), extend the duration and make other changes necessary to adapt the action to the new situation (see Article 39) — unless the grant has been terminated (see Article 32). The suspension will be lifted with effect from the suspension end date set out in the lifting of suspension letter. This date may be before the date on which the letter is sent. During the suspension, no prefinancing will be paid. Moreover, no work may be done. Ongoing work packages must be interrupted and no new work packages may be started. The beneficiaries may not claim damages due to suspension by the granting authority (see Article 33). Grant suspension does not affect the granting authority’s right to terminate the grant or a beneficiary (see Article 32) or reduce the grant (see Article 28). ARTICLE 32 — GRANT AGREEMENT OR BENEFICIARY TERMINATION 47 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 32.1 Consortium-requested GA termination 32.1.1 Conditions and procedure The beneficiaries may request the termination of the grant. The coordinator must submit a request for amendment (see Article 39), with: - the reasons why - the date the consortium ends work on the action (‘end of work date’) and - the date the termination takes effect (‘termination date’); this date must be after the date of the submission of the amendment request. The termination will take effect on the termination date specified in the amendment. If no reasons are given or if the granting authority considers the reasons do not justify termination, it may consider the grant terminated improperly. 32.1.2 Effects The coordinator must — within 60 days from when termination takes effect — submit a periodic report (for the open reporting period until termination). The granting authority will calculate the final grant amount and final payment on the basis of the report submitted and taking into account the lump sum contributions for activities implemented before the end of work date (see Article 22). Partial lump sum contributions for work packages that were not completed (e.g. due to technical reasons) may exceptionally be taken into account. If the granting authority does not receive the report within the deadline, only lump sum contributions which are included in an approved periodic report will be taken into account (no contributions if no periodic report was ever approved). Improper termination may lead to a grant reduction (see Article 28). After termination, the beneficiaries’ obligations (in particular Articles 13 (confidentiality and security), 16 (IPR), 17 (communication, dissemination and visibility), 21 (reporting), 25 (checks, reviews, audits and investigations), 26 (impact evaluation), 27 (rejections), 28 (grant reduction) and 42 (assignment of claims)) continue to apply. 32.2 Consortium-requested beneficiary termination 32.2.1 Conditions and procedure The coordinator may request the termination of the participation of one or more beneficiaries, on request of the beneficiary concerned or on behalf of the other beneficiaries. The coordinator must submit a request for amendment (see Article 39), with: - the reasons why 48 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 - the opinion of the beneficiary concerned (or proof that this opinion has been requested in writing) - the date the beneficiary ends work on the action (‘end of work date’) - the date the termination takes effect (‘termination date’); this date must be after the date of the submission of the amendment request. If the termination concerns the coordinator and is done without its agreement, the amendment request must be submitted by another beneficiary (acting on behalf of the consortium). The termination will take effect on the termination date specified in the amendment. If no information is given or if the granting authority considers that the reasons do not justify termination, it may consider the beneficiary to have been terminated improperly. 32.2.2 Effects The coordinator must — within 60 days from when termination takes effect — submit: (i) a report on the distribution of payments to the beneficiary concerned (ii) a termination report from the beneficiary concerned, for the open reporting period until termination, containing an overview of the progress of the work (iii) a second request for amendment (see Article 39) with other amendments needed (e.g. reallocation of the tasks and the estimated budget of the terminated beneficiary; addition of a new beneficiary to replace the terminated beneficiary; change of coordinator, etc.). The granting authority will calculate the amount due to the beneficiary on the basis of the reports submitted in previous interim payments (i.e. beneficiary’s lump sum contributions for completed and approved work packages). Lump sum contributions for ongoing/not yet completed work packages will have to be included in the periodic report for the next reporting periods when those work packages have been completed. If the granting authority does not receive the report on the distribution of payments within the deadline, it will consider that: - the coordinator did not distribute any payment to the beneficiary concerned and that - the beneficiary concerned must not repay any amount to the coordinator. If the second request for amendment is accepted by the granting authority, the Agreement is amended to introduce the necessary changes (see Article 39). If the second request for amendment is rejected by the granting authority (because it calls into question the decision awarding the grant or breaches the principle of equal treatment of applicants), the grant may be terminated (see Article 32). Improper termination may lead to a reduction of the grant (see Article 31) or grant termination (see Article 32). 49 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 After termination, the concerned beneficiary’s obligations (in particular Articles 13 (confidentiality and security), 16 (IPR), 17 (communication, dissemination and visibility), 21 (reporting), 25 (checks, reviews, audits and investigations), 26 (impact evaluation), 27 (rejections), 28 (grant reduction) and 42 (assignment of claims)) continue to apply. 32.3 EU-initiated GA or beneficiary termination 32.3.1 Conditions The granting authority may terminate the grant or the participation of one or more beneficiaries, if: (a) one or more beneficiaries do not accede to the Agreement (see Article 40) (b) a change to the action or the legal, financial, technical, organisational or ownership situation of a beneficiary is likely to substantially affect the implementation of the action or calls into question the decision to award the grant (including changes linked to one of the exclusion grounds listed in the declaration of honour) (c) following termination of one or more beneficiaries, the necessary changes to the Agreement (and their impact on the action) would call into question the decision awarding the grant or breach the principle of equal treatment of applicants (d) implementation of the action has become impossible or the changes necessary for its continuation would call into question the decision awarding the grant or breach the principle of equal treatment of applicants (e) a beneficiary (or person with unlimited liability for its debts) is subject to bankruptcy proceedings or similar (including insolvency, winding-up, administration by a liquidator or court, arrangement with creditors, suspension of business activities, etc.) (f) a beneficiary (or person with unlimited liability for its debts) is in breach of social security or tax obligations (g) a beneficiary (or person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has been found guilty of grave professional misconduct (h) a beneficiary (or person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed fraud, corruption, or is involved in a criminal organisation, money laundering, terrorism-related crimes (including terrorism financing), child labour or human trafficking (i) a beneficiary (or person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) was created under a different jurisdiction with the intent to circumvent fiscal, social or other legal obligations in the country of origin (or created another entity with this purpose) (j) a beneficiary (or person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed: (i) substantial errors, irregularities or fraud or 50 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 (ii) serious breach of obligations under this Agreement or during its award (including improper implementation of the action, non-compliance with the call conditions, submission of false information, failure to provide required information, breach of ethics or security rules (if applicable), etc.) (k) a beneficiary (or person having powers of representation, decision-making or control, or person essential for the award/implementation of the grant) has committed — in other EU grants awarded to it under similar conditions — systemic or recurrent errors, irregularities, fraud or serious breach of obligations that have a material impact on this grant (extension of findings; see Article 25.5) (l) despite a specific request by the granting authority, a beneficiary does not request — through the coordinator — an amendment to the Agreement to end the participation of one of its affiliated entities or associated partners that is in one of the situations under points (d), (f), (e), (g), (h), (i) or (j) and to reallocate its tasks, or (m) other: (i) linked action issues: not applicable (ii) additional GA termination grounds: not applicable. 32.3.2 Procedure Before terminating the grant or participation of one or more beneficiaries, the granting authority will send a pre-information letter to the coordinator or beneficiary concerned: - formally notifying the intention to terminate and the reasons why and - requesting observations within 30 days of receiving notification. If the granting authority does not receive observations or decides to pursue the procedure despite the observations it has received, it will confirm the termination and the date it will take effect (confirmation letter). Otherwise, it will formally notify that the procedure is discontinued. For beneficiary terminations, the granting authority will — at the end of the procedure — also inform the coordinator. The termination will take effect the day after the confirmation notification is sent (or on a later date specified in the notification; ‘termination date’). 32.3.3 Effects (a) for GA termination: The coordinator must — within 60 days from when termination takes effect — submit a periodic report (for the last open reporting period until termination). The granting authority will calculate the final grant amount and final payment on the basis of the report submitted and taking into account the lump sum contributions for activities implemented before termination takes effect (see Article 22). Partial lump sum contributions 51 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 for work packages that were not completed (e.g. due to technical reasons) may exceptionally be taken into account. If the grant is terminated for breach of the obligation to submit reports, the coordinator may not submit any report after termination. If the granting authority does not receive the report within the deadline, only lump sum contributions which are included in an approved periodic report will be taken into account (no contributions if no periodic report was ever approved). Termination does not affect the granting authority’s right to reduce the grant (see Article 28) or to impose administrative sanctions (see Article 34). The beneficiaries may not claim damages due to termination by the granting authority (see Article 33). After termination, the beneficiaries’ obligations (in particular Articles 13 (confidentiality and security), 16 (IPR), 17 (communication, dissemination and visibility), 21 (reporting), 25 (checks, reviews, audits and investigations), 26 (impact evaluation), 27 (rejections), 28 (grant reduction) and 42 (assignment of claims)) continue to apply. (b) for beneficiary termination: The coordinator must — within 60 days from when termination takes effect — submit: (i) a report on the distribution of payments to the beneficiary concerned (ii) a termination report from the beneficiary concerned, for the open reporting period until termination, containing an overview of the progress of the work (iii) a request for amendment (see Article 39) with any amendments needed (e.g. reallocation of the tasks and the estimated budget of the terminated beneficiary; addition of a new beneficiary to replace the terminated beneficiary; change of coordinator, etc.). The granting authority will calculate the amount due to the beneficiary on the basis of the reports submitted in previous interim payments (i.e. beneficiary’s lump sum contributions for completed and approved work packages). Lump sum contributions for ongoing/not yet completed work packages will have to be included in the periodic report for the next reporting periods when those work packages have been completed. If the granting authority does not receive the report on the distribution of payments within the deadline, it will consider that: - the coordinator did not distribute any payment to the beneficiary concerned and that - the beneficiary concerned must not repay any amount to the coordinator. If the request for amendment is accepted by the granting authority, the Agreement is amended to introduce the necessary changes (see Article 39). 52 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 If the request for amendment is rejected by the granting authority (because it calls into question the decision awarding the grant or breaches the principle of equal treatment of applicants), the grant may be terminated (see Article 32). After termination, the concerned beneficiary’s obligations (in particular Articles 13 (confidentiality and security), 16 (IPR), 17 (communication, dissemination and visibility), 21 (reporting), 25 (checks, reviews, audits and investigations), 26 (impact evaluation), 27 (rejections), 28 (grant reduction) and 42 (assignment of claims)) continue to apply. SECTION 3 OTHER CONSEQUENCES: DAMAGES AND ADMINISTRATIVE SANCTIONS ARTICLE 33 — DAMAGES 33.1 Liability of the granting authority The granting authority cannot be held liable for any damage caused to the beneficiaries or to third parties as a consequence of the implementation of the Agreement, including for gross negligence. The granting authority cannot be held liable for any damage caused by any of the beneficiaries or other participants involved in the action, as a consequence of the implementation of the Agreement. 33.2 Liability of the beneficiaries The beneficiaries must compensate the granting authority for any damage it sustains as a result of the implementation of the action or because the action was not implemented in full compliance with the Agreement, provided that it was caused by gross negligence or wilful act. The liability does not extend to indirect or consequential losses or similar damage (such as loss of profit, loss of revenue or loss of contracts), provided such damage was not caused by wilful act or by a breach of confidentiality. ARTICLE 34 — ADMINISTRATIVE SANCTIONS AND OTHER MEASURES Nothing in this Agreement may be construed as preventing the adoption of administrative sanctions (i.e. exclusion from EU award procedures and/or financial penalties) or other public law measures, in addition or as an alternative to the contractual measures provided under this Agreement (see, for instance, Articles 135 to 145 EU Financial Regulation 2018/1046 and Articles 4 and 7 of Regulation 2988/9518). SECTION 4 FORCE MAJEURE ARTICLE 35 — FORCE MAJEURE 18 Council Regulation (EC, Euratom) No 2988/95 of 18 December 1995 on the protection of the European Communities financial interests (OJ L 312, 23.12.1995, p. 1). 53 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 A party prevented by force majeure from fulfilling its obligations under the Agreement cannot be considered in breach of them. ‘Force majeure’ means any situation or event that: - prevents either party from fulfilling their obligations under the Agreement, - was unforeseeable, exceptional situation and beyond the parties’ control, - was not due to error or negligence on their part (or on the part of other participants involved in the action), and - proves to be inevitable in spite of exercising all due diligence. Any situation constituting force majeure must be formally notified to the other party without delay, stating the nature, likely duration and foreseeable effects. The parties must immediately take all the necessary steps to limit any damage due to force majeure and do their best to resume implementation of the action as soon as possible. CHAPTER 6 FINAL PROVISIONS ARTICLE 36 — COMMUNICATION BETWEEN THE PARTIES 36.1 Forms and means of communication — Electronic management EU grants are managed fully electronically through the EU Funding & Tenders Portal (‘Portal’). All communications must be made electronically through the Portal in accordance with the Portal Terms and Conditions and using the forms and templates provided there (except if explicitly instructed otherwise by the granting authority). Communications must be made in writing and clearly identify the grant agreement (project number and acronym). Communications must be made by persons authorised according to the Portal Terms and Conditions. For naming the authorised persons, each beneficiary must have designated — before the signature of this Agreement — a ‘legal entity appointed representative (LEAR)’. The role and tasks of the LEAR are stipulated in their appointment letter (see Portal Terms and Conditions). If the electronic exchange system is temporarily unavailable, instructions will be given on the Portal. 36.2 Date of communication The sending date for communications made through the Portal will be the date and time of sending, as indicated by the time logs. The receiving date for communications made through the Portal will be the date and time the communication is accessed, as indicated by the time logs. Formal notifications that have not been accessed within 10 days after sending, will be considered to have been accessed (see Portal Terms and Conditions). 54 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 If a communication is exceptionally made on paper (by e-mail or postal service), general principles apply (i.e. date of sending/receipt). Formal notifications by registered post with proof of delivery will be considered to have been received either on the delivery date registered by the postal service or the deadline for collection at the post office. If the electronic exchange system is temporarily unavailable, the sending party cannot be considered in breach of its obligation to send a communication within a specified deadline. 36.3 Addresses for communication The Portal can be accessed via the Europa website. The address for paper communications to the granting authority (if exceptionally allowed) is the official mailing address indicated on its website. For beneficiaries, it is the legal address specified in the Portal Participant Register. ARTICLE 37 — INTERPRETATION OF THE AGREEMENT The provisions in the Data Sheet take precedence over the rest of the Terms and Conditions of the Agreement. Annex 5 takes precedence over the Terms and Conditions. The Terms and Conditions take precedence over the Annexes other than Annex 5. Annex 2 takes precedence over Annex 1. ARTICLE 38 — CALCULATION OF PERIODS AND DEADLINES In accordance with Regulation No 1182/7119, periods expressed in days, months or years are calculated from the moment the triggering event occurs. The day during which that event occurs is not considered as falling within the period. ‘Days’ means calendar days, not working days. ARTICLE 39 — AMENDMENTS 39.1 Conditions The Agreement may be amended, unless the amendment entails changes to the Agreement which would call into question the decision awarding the grant or breach the principle of equal treatment of applicants. Amendments may be requested by any of the parties. 39.2 Procedure 19 Regulation (EEC, Euratom) No 1182/71 of the Council of 3 June 1971 determining the rules applicable to periods, dates and time-limits (OJ L 124, 8/6/1971, p. 1). 55 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 The party requesting an amendment must submit a request for amendment signed directly in the Portal Amendment tool. The coordinator submits and receives requests for amendment on behalf of the beneficiaries (see Annex 3). If a change of coordinator is requested without its agreement, the submission must be done by another beneficiary (acting on behalf of the other beneficiaries). The request for amendment must include: - the reasons why - the appropriate supporting documents and - for a change of coordinator without its agreement: the opinion of the coordinator (or proof that this opinion has been requested in writing). The granting authority may request additional information. If the party receiving the request agrees, it must sign the amendment in the tool within 45 days of receiving notification (or any additional information the granting authority has requested). If it does not agree, it must formally notify its disagreement within the same deadline. The deadline may be extended, if necessary for the assessment of the request. If no notification is received within the deadline, the request is considered to have been rejected. An amendment enters into force on the day of the signature of the receiving party. An amendment takes effect on the date of entry into force or other date specified in the amendment. ARTICLE 40 — ACCESSION AND ADDITION OF NEW BENEFICIARIES 40.1 Accession of the beneficiaries mentioned in the Preamble The beneficiaries which are not coordinator must accede to the grant by signing the accession form (see Annex 3) directly in the Portal Grant Preparation tool, within 30 days after the entry into force of the Agreement (see Article 44). They will assume the rights and obligations under the Agreement with effect from the date of its entry into force (see Article 44). If a beneficiary does not accede to the grant within the above deadline, the coordinator must — within 30 days — request an amendment (see Article 39) to terminate the beneficiary and make any changes necessary to ensure proper implementation of the action. This does not affect the granting authority’s right to terminate the grant (see Article 32). 40.2 Addition of new beneficiaries In justified cases, the beneficiaries may request the addition of a new beneficiary. For this purpose, the coordinator must submit a request for amendment in accordance with Article 39. It must include an accession form (see Annex 3) signed by the new beneficiary directly in the Portal Amendment tool. 56 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 New beneficiaries will assume the rights and obligations under the Agreement with effect from the date of their accession specified in the accession form (see Annex 3). Additions are also possible in mono-beneficiary grants. ARTICLE 41 — TRANSFER OF THE AGREEMENT In justified cases, the beneficiary of a mono-beneficiary grant may request the transfer of the grant to a new beneficiary, provided that this would not call into question the decision awarding the grant or breach the principle of equal treatment of applicants. The beneficiary must submit a request for amendment (see Article 39), with - the reasons why - the accession form (see Annex 3) signed by the new beneficiary directly in the Portal Amendment tool and - additional supporting documents (if required by the granting authority). The new beneficiary will assume the rights and obligations under the Agreement with effect from the date of accession specified in the accession form (see Annex 3). ARTICLE 42 — ASSIGNMENTS OF CLAIMS FOR PAYMENT AGAINST THE GRANTING AUTHORITY The beneficiaries may not assign any of their claims for payment against the granting authority to any third party, except if expressly approved in writing by the granting authority on the basis of a reasoned, written request by the coordinator (on behalf of the beneficiary concerned). If the granting authority has not accepted the assignment or if the terms of it are not observed, the assignment will have no effect on it. In no circumstances will an assignment release the beneficiaries from their obligations towards the granting authority. ARTICLE 43 — APPLICABLE LAW AND SETTLEMENT OF DISPUTES 43.1 Applicable law The Agreement is governed by the applicable EU law, supplemented if necessary by the law of Belgium. Special rules may apply for beneficiaries which are international organisations (if any; see Data Sheet, Point 5). 43.2 Dispute settlement If a dispute concerns the interpretation, application or validity of the Agreement, the parties must bring action before the EU General Court — or, on appeal, the EU Court of Justice — under Article 272 of the Treaty on the Functioning of the EU (TFEU). 57 Project: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 For non-EU beneficiaries (if any), such disputes must be brought before the courts of Brussels, Belgium — unless an international agreement provides for the enforceability of EU court judgements. For beneficiaries with arbitration as special dispute settlement forum (if any; see Data Sheet, Point 5), the dispute will — in the absence of an amicable settlement — be settled in accordance with the Rules for Arbitration published on the Portal. If a dispute concerns administrative sanctions, offsetting or an enforceable decision under Article 299 TFEU (see Articles 22 and 34), the beneficiaries must bring action before the General Court — or, on appeal, the Court of Justice — under Article 263 TFEU. For grants where the granting authority is an EU executive agency (see Preamble), actions against offsetting and enforceable decisions must be brought against the European Commission (not against the granting authority; see also Article 22). ARTICLE 44 — ENTRY INTO FORCE The Agreement will enter into force on the day of signature by the granting authority or the coordinator, depending on which is later. SIGNATURES For the coordinator For the granting authority [--TGSMark#signature-917673448_75_210--] [--TGSMark#signature-service_75_210--] Christian Veske with ECAS id n00erc43 signed in the Participant Signed by Maria TOMASSETTI with ECAS id tomasma as an Portal on 14/06/2024 at 08:28:00 (transaction id SigId-14996-ZlkzT authorised representative on 14-06-2024 18:36:07 (transaction id 1cQD4JBfUXPPK6WAR1kRAezHj1bDvAXYiqLbBiVjQ48Yp0KzKNfDEP SigId-25672-bQHJP6DMWum5Iic5dI9thU1eJelM3Fuzgw5EatxB0Bi1 OlcxGRBW4HdquOyZaAy83aofiIHG-yntOf97TTHqzVT78N4qoJh-Qa NQnyuAS4HH8FAokJAzWcSzkzzzJZWHNXZqLJvnkbYpcG- 43pHpBmve9hZbizGAzdxQzVx4vrArNL2lcBHlAhzIeQVTsypECMb1cj yntOf97TTHqzVT78N4qoJh-wZhntddlzIpnTlbT8EFzddkOqSYdGjY8A WGSipI8zoK4f5KAuf3w19szObqDzX0i). Timestamp by third party at swsKFj2kbJqXAr5YaYo5bQJx0eUUNTVpG2jOvP35N0X7MKjaZnaSy) 2024.06.14 08:28:10 CEST 2024.06.14 18:36:13 CEST 58  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 1 Citizens, Equality, Rights and Values Programme (CERV) Description of the action (DoA) Part A Part B Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 DESCRIPTION OF THE ACTION (PART A) COVER PAGE Part A of the Description of the Action (DoA) must be completed directly on the Portal Grant Preparation screens. PROJECT Grant Preparation (General Information screen) — Enter the info. Project number: 101144709 Project name: Improving response to risks of discrimination, bias and intolerance in automated decision-making systems to promote equality. Project acronym: EquiTech Call: CERV-2023-EQUAL Topic: CERV-2023-EQUAL Type of action: CERV-LS Service: JUST/H/03 Project starting date: fixed date: 1 May 2024 Project duration: 24 months TABLE OF CONTENTS Project summary ......................................................................................................................................................3 List of participants .................................................................................................................................................. 3 List of work packages .............................................................................................................................................4 Staff effort ............................................................................................................................................................. 11 List of deliverables ................................................................................................................................................12 List of milestones (outputs/outcomes) .................................................................................................................. 24 List of critical risks ............................................................................................................................................... 25 2 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 PROJECT SUMMARY Project summary Grant Preparation (General Information screen) — Provide an overall description of your project (including context and overall objectives, planned activities and main achievements, and expected results and impacts (on target groups, change procedures, capacities, innovation etc)). This summary should give readers a clear idea of what your project is about. Use the project summary from your proposal. Public authorities’ uptake of AI and other algorithmic tools in data collection, processing and services is a rapid and, most probably, a non-reversible new reality. Globally, many governments use automated decision support tools or even automated (algorithm-based) decision-making (ADM) systems in their everyday operations. ADM is the process of making a decision or creating other legally relevant output through automated means with limited or no human involvement. The ADM systems may process large amounts of information, identify patterns, make predictions, or take various types of actions based on predefined criteria. The general objective of the project is to enhance the capacity of public authorities and their co-operation partners in Estonia and Lithuania in addressing risks of discrimination and bias in ADM systems. This will be achieved by in-depth research, developing support materials, trainings and a media campaign. This project seeks to contribute to prevention of discrimination and gender inequality within the European Union context by ensuring that the ADM systems used in the public sector are fair and non-discriminatory. The key activities include: 1. Research on gaps in policies and legal frameworks on AI in Estonia and Lithuania together with mapping of potential risks of discrimination and bias in ADM systems deployed by Estonian public authorities; 2. Development of guidelines, impact assessment checklist, training materials and online course for public sector employees and system developers to address potential risks for discrimination and bias in design and deployment of ADM systems; 3. Capacity building among project stakeholders and target groups through trainings, study trips, international conference and an academic workshop; 4.Raising public awareness on the risks of discrimination and bias in public sector ADM systems. LIST OF PARTICIPANTS PARTICIPANTS Grant Preparation (Beneficiaries screen) — Enter the info. Number Role Short name Legal name Country PIC 1 COO SVV SOOLISE VORDOIGUSLIKKUSE JA VORDSE EE 917673448 KOHTLEMISE VOLINIKU KANTSELEI 2 BEN OEOO LYGIU GALIMYBIU KONTROLIERIAUS LT 913111732 TARNYBA 3 BEN TalTech TALLINNA TEHNIKAÜLIKOOL EE 999842536 4 BEN MKM MAJANDUS JA EE 963638450 KOMMUNIKATSIOONIMINISTEERIUM 5 BEN JUM MINISTRY OF JUSTICE EE 974582378 3 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 LIST OF WORK PACKAGES Work packages Grant Preparation (Work Packages screen) — Enter the info. Work Work Package name Lead Beneficiary Effort Start End Deliverables Package No (Person- Month Month Months) WP1 Management and coordination 1 - SVV 19.00 1 24 D1.1 – Progress report D1.2 – Partner meetings report WP2 Discrimination mapping and analysis 3 - TalTech 19.00 2 7 D2.1 – Audit on discrimination and bias within Estonian public sector ADM systems D2.2 – Legal analyses in Estonia and Lithuania WP3 Impact Assessment toolbox development 5 - JUM 19.00 7 17 D3.1 – Impact assessment checklist D3.2 – Online course for public sector D3.3 – Guidelines D3.4 – Training materials WP4 Testing the impact assessment toolbox 4 - MKM 6.00 12 17 D4.1 – Report of the testing the public ADM systems in Estonia WP5 Capacity building and training activities for 3 - TalTech 11.00 2 24 D5.1 – Report of the training for the national, regional and local authorities Estonian and Lithuanian employees of public sector organisations D5.2 – Report of the training for the data scientists and data stewards training D5.3 – The report on the study visits D5.4 – Report of the implementation of the online course for public sector organizations in Estonia and Lithuania. WP6 Dissemination and awareness raising 1 - SVV 21.00 15 24 D6.1 – Report of the international conference on discrimination in ADM systems D6.2 – High-category scholarly article 4 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Work packages Grant Preparation (Work Packages screen) — Enter the info. Work Work Package name Lead Beneficiary Effort Start End Deliverables Package No (Person- Month Month Months) D6.3 – Report on Public Awareness Raising Activities Concerning Potential Risks and Biases in Public ADM Systems D6.4 – Summary Report of the Academic Workshop D6.5 – Policy Development: Recommendations and Strategies for ADM Systems to Combat Discrimination D6.6 – Evaluation of the Value and Impact Created by the Activities in the Final Phase of the EquiTech Project 5 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Work package WP1 – Management and coordination Work Package Number WP1 Lead Beneficiary 1 - SVV Work Package Name Management and coordination Start Month 1 End Month 24 Objectives - Management and coordination of the EquiTech project - Evaluation of the project activities - Mid-time and final report Description 1.1. Management and coordination. Project coordination and management to follow the timeline of activities, coordinate the communication between the partners. Including the 4x partner meetings: 2x in Estonia, 1xLithuania, 1xOnline. Partner meetings are to coordinate the tasks and deliverables, to share the information and give the input for the ongoing actions and deliverables. At the partner meetings will be also coordinated the mid-term and final report. As the output of the he kick-off partner meeting will be created a workplan, map the possible partners and tools needed. Furthermore, also the Ethics Guideline will be elaborated at the first meeting of the project partners. The document will regulate various aspects related to the data collection, protection and preservation, ensuring confidentiality, prevention of plagiarism, preservation and exploitation of gender-sensitive data. 1.2. Reporting. The mid-term and final reports will also be coordinated during the partner meetings. Work package WP2 – Discrimination mapping and analysis Work Package Number WP2 Lead Beneficiary 3 - TalTech Work Package Name Discrimination mapping and analysis Start Month 2 End Month 7 Objectives ▪ Conducting a literature review on discrimination and bias in ADM systems. ▪ Legal analyses in Estonia and Lithuania. ▪ Auditing discrimination and bias within Estonian public sector ADM systems and current assessment practice. Description T2.1. Conducting a literature review on discrimination and bias in ADM systems. The EquiTech project will undertake a thorough literature review on discrimination and bias in ADM systems. This review aims to develop a systematic understanding of the extent of these risks and identify optimal solutions to address them. It will give a holistic view of the current situation of the discrimination topics related to the ADM systems. By conducting this comprehensive review, the EquiTech project aims to contribute to the advancement of knowledge and practices in addressing bias and discrimination in ADM systems, ultimately fostering more fair and equitable outcomes. Literature review will enable all project partners communicate on the concepts effectively. Systematic understanding of the extent of risks in context of Estonia and Lithuania is created to understand the best solutions to tackle them better. This task is a prerequisite for creating D2.1 and D2.2. T2.2. Legal analyses in Estonia and Lithuania. A legal analysis will be conducted in both countries – Estonia and Lithuania, focusing on the existing legal provisions and regulations. This analysis will help in identifying the legal framework surrounding discrimination and bias in ADM systems and provide valuable insights for future actions and improvements. 6 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 T2.3. Auditing discrimination and bias within Estonian public sector ADM systems and current assessment practice. The EquiTech project aims to establish an auditing process specifically designed to identify discrimination and bias risks within these systems and evaluate the current assessment practices in place in Estonia. By implementing the auditing process, the project seeks to enhance transparency and accountability, while also promoting the development of more inclusive and fair ADM practices. An audit on providing an overview of potential discriminatory and bias practices within the ADM systems used in the public sector, along with legal and organizational recommendations for improvement. Conducting a report – mapping the risks of bias and discrimination on ADM systems used in public sector, mapping the assessment practice and method(s) for a risk analysis. Work package WP3 – Impact Assessment toolbox development Work Package Number WP3 Lead Beneficiary 5 - JUM Work Package Name Impact Assessment toolbox development Start Month 7 End Month 17 Objectives ▪ Higher quality and more standardised practices across the public sector in the process of identifying, assessing, and mitigating potential biases and discriminatory practices in ADM systems. ▪ Increased awareness as well as the skills and capabilities among the key stakeholders in understanding and managing the above risks in ADM systems. ▪ Improved fairness of AI models and algorithms used in the public sector, increased transparency in administrative decision-making, and reduced risk of discriminatory outcomes. ▪ More inclusive and fair approach in the design, development, and implementation of ADM systems that consider the complexities, functionalities, and potential pitfalls of using ADM systems in the public sector. ▪ Reduced risk of legal or regulatory non-compliance of public sector ADM systems. ▪ Increased public trust in the public sector's use of ADM systems. Description T3.1. Impact assessment checklist development. Developing an impact assessment checklist for identifying and assessing potential risks of bias and discrimination within the ADM system. The checklist is structured to ask a series of questions or list considerations concerning these areas. It builds on and refers back to the guidelines in answering specific questions on how possible risks are being identified and mitigated. Impact assessment checklist will be available electronically in 3 languages (ET, LT, EN), content of 15 pages in each language. T3.2. Guidelines development for system developers, scientists, data stewards, and public service providers to detect bias and discrimination in the design, implementation, and development of ADM systems. Guidelines development for system developers, data stewards and public service providers to detect bias and discrimination in the design, implementation, and development of ADM systems. Guidelines is the document, which gives a step-by-step overview on how to use the impact assessment checklist, explaining assessment process and ways to avoid discrimination while developing and using AI solutions. The guidelines would be based on the latest research as well as findings from the situation mapping exercise done under WP2. The guidelines will be available electronically in 3 languages (ET, LT, EN), content of 60 pages in each language. T3.3. Creating an online course for Estonian and Lithuanian public sector representatives. The activity includes developing and making available an online course for the public sector officials. These courses would be designed to increase awareness and understanding of discrimination and bias in ADM systems. They would cover topics such as the potential sources of bias in AI, the impact of bias on decision-making, and basic suggestions for mitigating bias. It is a 60-minutes online course consisting of 30 pages of text content per language (ET, EN, LT) and up to 5 videos (3-5 min). The online course for the employees of the public sector will be available on two platforms: 1) For the employees of the public sector in Estonia in Estonian and in English on Estonian Digital Academy (digiriigiakadeemia.ee). 2) For the employees of the public sector in Lithuania in Lithuanian. T3.4. Developing training materials for data scientists, data stewards, and public service providers. This activity foresees developing a set of educational materials that assist in training public sector officials, system developers and other stakeholders about the complexities and potential risks concerning bias and discrimination in ADM 7 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 systems. Primarily, they are meant to support the training carried out under WP5 but can also be used separately. Content of of 25 pages for the data scientists, data stewards and 25 pages content for the employees of public service, both in 3 languages (ET, EN, LT). The Materials will be available electronically, but for the trainees is also a 32 pages printed version. Work package WP4 – Testing the impact assessment toolbox Work Package Number WP4 Lead Beneficiary 4 - MKM Work Package Name Testing the impact assessment toolbox Start Month 12 End Month 17 Objectives ▪ Testing the impact assessment checklist for potential biases and discriminatory practices within the ADM systems used in the public sector ▪ Evaluation of the discrimination risk profile of the ADM systems used in the public sector Description T4.1. Testing the Impact Assessment Checklist for Potential Biases and Discriminatory Practices within the ADM Systems Used in the Public Sector This activity involves applying the impact assessment checklist to three real solutions utilized in the Estonian public sector. This testing establishes a benchmark for future comprehensive testing of all systems. The impact assessment checklist is designed to identify potential biases and discriminatory practices within the ADM systems deployed in the public sector. This process will lead to an evaluation of the discrimination risk profile associated with these ADM systems. The methodology and process description for this task will be outlined in Deliverable 4.1. T4.2. Evaluation of the Discrimination Risk Profile of the ADM Systems Used in the Public Sector This involves evaluating the results of the assessment based on real AI use cases from Estonia, considering aspects that can be improved to enhance the assessment experience and the accuracy of the results. The evaluation of the Discrimination Risk Profile will be detailed in Deliverable 4.1. Based on the testing results, the impact assessment checklist (D3.1) and guidelines (D3.3) will be refined to support system developers, data scientists, data stewards, and public service employees in recognizing discrimination and bias during the design, implementation, and development stages of ADM systems. This testing sets the precedent for future comprehensive testing of all systems. Work package WP5 – Capacity building and training activities for national, regional and local authorities Work Package Number WP5 Lead Beneficiary 3 - TalTech Work Package Name Capacity building and training activities for national, regional and local authorities Start Month 2 End Month 24 Objectives The objective of the working package is to train and raise awareness among stakeholders and individuals working with public sector ADM systems on a daily basis about the potential risks of discrimination and bias. ▪ Training for Estonian and Lithuanian employees of national, regional and local authorities ▪ Study visits to European leaders in fighting discrimination through ADM implementation. ▪ Training in avoiding discrimination risks for data scientists and data stewards. ▪ Implementation of the online course for public sector organizations in Estonia and Lithuania. Description 8 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 T5.1. Training for Estonian and Lithuanian employees of public sector organisations. The face to face trainings for public authorities will be carried out to deepen the digital skills and critical thinking to identify the possible discrimination risks on the public sector ADM systems. Public sector organizations often handle vast amounts of sensitive data, making effective data stewardship essential for the successful implementation of AI- driven initiatives. In total 150 employees in public sector will be trained: 2x in Estonia in ET, 2x50 participants; 3x in Lithuania in LT, together 50 participants in 3 trainings. T5.2. Study visits to European leaders in fighting discrimination through ADM implementation. Study visits will be organized to European countries who have made significant progress in combating discrimination through the implementation of ADM systems in the public sector. Study visits offer an invaluable opportunity to observe and learn from the approaches taken by other European countries in addressing similar challenges and understanding discrimination cases. All consortium partners have a representatives at the study visit. T5.3. Training in avoiding discrimination risks for data scientists and data stewards. The face to face training teaches the criticality of effective data management in the successful deployment of AI solutions. Participants will gain a deep understanding of the role of data stewards and their responsibilities in ensuring data quality, integrity, and security throughout the AI development lifecycle with focus on risks of discrimination and bias in ADM systems. Face to face training for the data scientists and data stewards. The training program tailored for the public sector will concentrate on data management for AI deployment. In total 80 data scientists and data stewards will be trained: 2x Estonia in ET, 2x20 participants and 2x Lithuania in EN/ LT, 2x20 participants. T5.4. Implementation of the online course for public sector organisations in Estonia and Lithuania. The course is available on 2 platforms: Digital State Academy (digiriigiakadeemia.ee) in Estonian and English and in Lithuanian portal in Lithuanian. Work package WP6 – Dissemination and awareness raising Work Package Number WP6 Lead Beneficiary 1 - SVV Work Package Name Dissemination and awareness raising Start Month 15 End Month 24 Objectives The objective of the working package is to disseminate the knowledge gathered from the project's activities to professionals, the academic community, and the general public. Additionally, the aim is to share this knowledge with equality experts throughout Europe. ● General public awareness raising activities devising and implementing a media campaign to raise awareness about possible bias and discrimination in ADM systems. ● Organization of an international conference on discrimination in ADM systems. ● Academic and professional dissemination (High-category scholarly article, and workshop). ● Policy development: policy recommendations and strategies on ADM systems for the fight against discrimination. Description 6.1. Public awareness raising activities (Devising and implementing a media campaign to raise awareness about possible bias and discrimination in ADM systems). For the general public, we will focus on addressing the potential for discrimination when utilizing digital services. Currently, there is a lack of systematic and research-backed knowledge among professionals in Estonia and Lithuania regarding the discrimination and bias risks within the public sector ADM systems. The primary focus of our awareness campaign will be directed towards the vulnerable target groups through social media channels such as Facebook and Instagram. In addition to these efforts, we will also employ online and Google campaigns to further support our cause. Running in parallel with these activities, we plan to conduct interviews with experts from the EquiTech project on television and radio programmes. Awareness raising campaign in Lithuania will focus on general public with a goal to inform and raise awareness about the possible discrimination risks in AI. The campaign will include press releases, articles, interviews with experts, local TV and radio show appearances, social media content and ads. Reach 300000 citizens in Estonia and 500000 citizens in Lithuania. 6.2. Organization of an international conference on discrimination in ADM systems. An international conference on discrimination in ADM systems will be organized in Tallinn. The conference aims to bring 9 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 together experts on ADM systems in the public sector, ADM developers, and equality bodies and scientists from across European countries to share their expertise and insights. The primary goal is to facilitate mutual learning and explore ways to apply the combined knowledge of discrimination and AI systems.The conference is expected to attract a minimum of 75 participants, including representatives from various sectors. By gathering experts from different backgrounds, the conference seeks to create an environment conducive to collaborative learning, knowledge sharing, and the identification of common strategies to address discrimination in ADM systems. 6.3. Academic and professional dissemination (High-category scholarly article and workshop). The high-category scholarly article will raise awareness of discrimination in digital world among academics to discuss it further in different scope, facilitating also inserting it to the teaching and curricula. Academic workshop allows handle the topic in scientifically discussing the topic with other scientists and disseminating by this the topic of discrimination to the new research and scientific discussions. Number of participants 30, notably academics who are interested in developing the knowledge of non-discriminative AI related topics. 6.4. Policy development: policy recommendations and strategies on ADM systems for the fight against discrimination. Policy recommendations on ADM systems for fight against discrimination in Estonia and Lithuania: The project encompasses various activities – such as evaluating current practices for identifying and mitigating biases in public sector ADM systems, and the development and testing of guidelines – that can provide valuable input to shaping governance and policy decisions concerning ADM systems in the public sector. Therefore, in the final stage of the project, one of the tasks is to draft a paper featuring policy recommendations that reflect the insights from the preceding stages of the project. These recommendations could potentially include a wide range of actionable items, such as mandating the execution of an impact assessment checklist within the procurement process, establishing systematic bias and discrimination training for certain public office roles, advocating for periodic audits and reviews of the ADM systems, fostering multi-stakeholder dialogue. 6.5. Evaluation of project activities. Evaluation of the value and impact created by the activities in the final phase of the EquiTech project. 10 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 STAFF EFFORT Staff effort per participant Grant Preparation (Work packages - Effort screen) — Enter the info. Participant WP1 WP2 WP3 WP4 WP5 WP6 Total Person-Months 1 - SVV 11.00 1.00 1.00 4.00 9.00 26.00 2 - OEOO 4.00 2.00 3.00 2.00 5.00 16.00 3 - TalTech 2.00 11.00 5.00 1.00 4.00 5.00 28.00 4 - MKM 1.00 2.00 3.00 4.00 1.00 1.00 12.00 5 - JUM 1.00 3.00 7.00 1.00 1.00 13.00 Total Person-Months 19.00 19.00 19.00 6.00 11.00 21.00 95.00 11 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 LIST OF DELIVERABLES Deliverables Grant Preparation (Deliverables screen) — Enter the info. The labels used mean: Public — fully open ( automatically posted online) Sensitive — limited under the conditions of the Grant Agreement EU classified —RESTREINT-UE/EU-RESTRICTED, CONFIDENTIEL-UE/EU-CONFIDENTIAL, SECRET-UE/EU-SECRET under Decision 2015/444 Deliverable Deliverable Name Work Lead Beneficiary Type Dissemination Level Due Date No Package (month) No D1.1 Progress report WP1 1 - SVV R — Document, report SEN - Sensitive 13 D1.2 Partner meetings report WP1 1 - SVV R — Document, report SEN - Sensitive 24 D2.1 Audit on discrimination and bias within WP2 3 - TalTech R — Document, report PU - Public 7 Estonian public sector ADM systems D2.2 Legal analyses in Estonia and Lithuania WP2 5 - JUM R — Document, report PU - Public 7 D3.1 Impact assessment checklist WP3 5 - JUM DEM — Demonstrator, PU - Public 17 pilot, prototype D3.2 Online course for public sector WP3 5 - JUM DEC —Websites, patent PU - Public 17 filings, videos, etc D3.3 Guidelines WP3 5 - JUM R — Document, report PU - Public 17 D3.4 Training materials WP3 3 - TalTech R — Document, report PU - Public 17 D4.1 Report of the testing the public ADM WP4 4 - MKM DEM — Demonstrator, PU - Public 17 systems in Estonia pilot, prototype D5.1 Report of the training for the Estonian WP5 3 - TalTech OTHER SEN - Sensitive 23 and Lithuanian employees of public sector organisations 12 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Deliverables Grant Preparation (Deliverables screen) — Enter the info. The labels used mean: Public — fully open ( automatically posted online) Sensitive — limited under the conditions of the Grant Agreement EU classified —RESTREINT-UE/EU-RESTRICTED, CONFIDENTIEL-UE/EU-CONFIDENTIAL, SECRET-UE/EU-SECRET under Decision 2015/444 Deliverable Deliverable Name Work Lead Beneficiary Type Dissemination Level Due Date No Package (month) No D5.2 Report of the training for the data scientists WP5 4 - MKM OTHER SEN - Sensitive 23 and data stewards training D5.3 The report on the study visits WP5 3 - TalTech R — Document, report SEN - Sensitive 9 D5.4 Report of the implementation of the online WP5 3 - TalTech R — Document, report SEN - Sensitive 24 course for public sector organizations in Estonia and Lithuania. D6.1 Report of the international conference on WP6 1 - SVV OTHER SEN - Sensitive 23 discrimination in ADM systems D6.2 High-category scholarly article WP6 3 - TalTech R — Document, report PU - Public 24 D6.3 Report on Public Awareness Raising WP6 1 - SVV R — Document, report PU - Public 24 Activities Concerning Potential Risks and Biases in Public ADM Systems D6.4 Summary Report of the Academic Workshop WP6 3 - TalTech R — Document, report SEN - Sensitive 24 D6.5 Policy Development: Recommendations and WP6 3 - TalTech R — Document, report PU - Public 24 Strategies for ADM Systems to Combat Discrimination D6.6 Evaluation of the Value and Impact Created WP6 3 - TalTech R — Document, report PU - Public 24 by the Activities in the Final Phase of the EquiTech Project 13 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Deliverable D1.1 – Progress report Deliverable Number D1.1 Lead Beneficiary 1 - SVV Deliverable Name Progress report Type R — Document, report Dissemination Level SEN - Sensitive Due Date (month) 13 Work Package No WP1 Description EC standard template, in EN. Deliverable D1.2 – Partner meetings report Deliverable Number D1.2 Lead Beneficiary 1 - SVV Deliverable Name Partner meetings report Type R — Document, report Dissemination Level SEN - Sensitive Due Date (month) 24 Work Package No WP1 Description The summary report (around 2-4 pages in EN) will cover the 4x partner meetings: 2x in Estonia, 1x Lithuania, 1x Online. Partner meetings are to coordinate the tasks and deliverables, to share the information and give the input for the ongoing actions and deliverables. At the partner meetings will be also coordinated the mid-term and final report. The duration depends on the needs, between 4-8 hours per meeting. Approximately 10-15 people in total per meeting. The report will be supported by the agenda and signed presence sheets (presence sheets extracted from the platform used/screenshots for the online meeting). Deliverable D2.1 – Audit on discrimination and bias within Estonian public sector ADM systems Deliverable Number D2.1 Lead Beneficiary 3 - TalTech Deliverable Name Audit on discrimination and bias within Estonian public sector ADM systems Type R — Document, report Dissemination Level PU - Public Due Date (month) 7 Work Package No WP2 Description The EquiTech project will establish an auditing process to identify discrimination and bias risks within the AMD systems and evaluate the current assessment practices in place. The report (in EN, 30 pages) will be available electronically to the project partners (saved as a pdf-document) and will include description of the auditing process and methodology, the analysis and results of auditing discrimination and bias within the Estonian public sector ADM systems, evaluation of the current assessment practices as well as legal and organizational recommendations for improvement. The collected data will be utilized in subsequent Work Packages, for instance, in creating an online course, developing guidelines, preparing training materials, and during conference presentations and scholarly articles. 3 ADM systems will be audited. 14 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Deliverable D2.2 – Legal analyses in Estonia and Lithuania Deliverable Number D2.2 Lead Beneficiary 5 - JUM Deliverable Name Legal analyses in Estonia and Lithuania Type R — Document, report Dissemination Level PU - Public Due Date (month) 7 Work Package No WP2 Description A legal analysis will be conducted in both countries – Estonia and Lithuania, focusing on the existing legal provisions and regulations. The analysis are available electronically and in English. The legal analyses are available electronically to the project partners (saved as a pdf-document) and in English, comprising 30 pages per country. The collected data will be utilized in subsequent Work Packages, for instance, in creating an online course, developing guidelines, preparing training materials, and during conference presentations and scholarly articles. Deliverable D3.1 – Impact assessment checklist Deliverable Number D3.1 Lead Beneficiary 5 - JUM Deliverable Name Impact assessment checklist Type DEM — Demonstrator, Dissemination Level PU - Public pilot, prototype Due Date (month) 17 Work Package No WP3 Description Creating and providing an impact assessment checklist. Developing an impact assessment checklist for identifying and assessing potential risks of discrimination and bias within the ADM system. The checklist is structured to ask a series of questions or list considerations concerning these areas. It builds on and refers back to the guidelines in answering specific questions on how possible risks are being identified and mitigated. The goal of the assessment checklist is to proactively uncover potential issues that may arise when deploying an ADM system. The impact assessment checklist will be tested in WP4 and the report of the testing (D4.1) will provide final input to refine the developed impact assessment checklist which will be public. It allows public officials and developers to address these risks in the design and implementation stages, thereby minimising negative consequences. The assessment checklist will target both the public officials as well as system developers. The impact assessment checklist is targeted for the project team responsible for the development of the ADM system. Specifically, it is presumed to be used by the project lead together with the legal support team during and/or after the development of the ADM system. The checklist is in 3 languages (ET, LT, EN), 15 pages per language, in a digital document format (word-document) and made available for organizations through the Government website dedicated to responsible development and use of algorithms in public sector. The checklist will be adapted to each country's needs. Deliverable D3.2 – Online course for public sector Deliverable Number D3.2 Lead Beneficiary 5 - JUM Deliverable Name Online course for public sector Type DEC —Websites, patent Dissemination Level PU - Public filings, videos, etc 15 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Due Date (month) 17 Work Package No WP3 Description The deliverable is an online course, which consists of video as well as text materials and interactive tasks. The course will be designed to increase awareness and understanding of discrimination and bias in public sector ADM systems. It will describe risks related to biases in data in ADM systems and the possible discrimination it may lead to. It would cover topics such as the potential sources of bias in AI, the impact of bias on decision-making, and basic suggestions for mitigating bias. Among else, it will include an overview of the state-of-the-art mechanisms that are today adopted in assessing and testing for biases. This is meant as an introductory course, to create awareness and knowledge on the pertinent risks and solutions, rather than a deep-dive. The 60-minutes online course, consisting of 30 pages of text content per language and up to 5 videos (3-5 min). The video clips will include sign language interpretations and subtitles in ET (in Estonia). The online course for the employees of the public sector will be on two platforms: 1. For the employees of national, regional and local authorities in ET, in Estonian and English on Estonian Digital Academy (digiriigiakadeemia.ee). 2. For the employees of national, regional and local authorities in Lithuania in LT on the online learning platform for Lithuanian public sector: administrated by Lithuanian Equality Body. The online course will be adapted to each country's needs. Deliverable D3.3 – Guidelines Deliverable Number D3.3 Lead Beneficiary 5 - JUM Deliverable Name Guidelines Type R — Document, report Dissemination Level PU - Public Due Date (month) 17 Work Package No WP3 Description Guidelines x 2 for the target group: 1. data scientists, system developers and data stewards: available electronically in 3 languages (ET, LT, EN). 2. employees of the public sector: available electronically in 3 languages (ET, LT, EN). The guidelines will consist of at least 60 pages of substantial text per language = around 90 pages or more including the design, images, schematics, etc. (the total number of pages is for both guidelines combined). The guidelines assist the project team in detecting discrimination and bias in the design, implementation, and development of ADM systems. They give a step-by-step overview on how to use the impact assessment checklist, explaining assessment process and ways to avoid discrimination while developing and using AI solutions. The materials created will provide concrete technical and organisational recommendations through the whole lifecycle of the ADM system – including the initial data collection, algorithm design, model testing, output analysis, post-deployment risk management, etc. Among else, the guidelines will cover: - How to carry out inclusive system design. That is, to ensure that the ADM is designed with consideration for diverse groups with specific feature such as gender, race, disability and else other forms of discrimination. - How to identify and reduce biases in data and in the algorithmic model. - Recommendations on public consultation during the system development, including engaging with marginalised groups. - How to carry out testing of a new ADM system for identifying potential risks of discrimination. - How to measure the impact the AI system has on fairness post-deployment. The guidelines will be made available as a separate pdf-document or Word-document and will be made available on the Government websites dedicated to responsible development and use of algorithms in public sector. They will be further promoted by the responsible ministries through their communication channels and through other deliverables such as 16 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 the online course and impact assessment checklist. They are expected to reach all key public organisations developing or operating ADM systems. Est. number of people reached: 200. The guidelines will be adapted to each country's needs. Deliverable D3.4 – Training materials Deliverable Number D3.4 Lead Beneficiary 3 - TalTech Deliverable Name Training materials Type R — Document, report Dissemination Level PU - Public Due Date (month) 17 Work Package No WP3 Description This activity foresees developing a set of educational materials that assist in training public sector officials, system developers and other stakeholders about the complexities and potential risks concerning bias and discrimination in ADM systems. Primarily, they are meant to support the training carried out under WP5 but can also be used separately. Training material x 2 target groups: 1. Developing training materials in avoiding discrimination risks for the system developers, data scientists and data stewards. A set of educational materials available electronically in 3 languages (ET, LT, EN), content of 25 pages in each language. Participants will learn about data governance framework, data quality assurance, and the responsible use of data in the public sector context with specific focus on risks of discrimination and bias in ADM systems. 2. Developing training materials on avoiding discrimination risks for Estonian and Lithuanian employees of national, regional and local authorities. A set of educational materials available electronically in 3 languages (ET, LT, EN), content of 25 pages in each language. Training of public authorities will be carried out to deepen the digital skills and critical thinking to identify the possible discrimination risks in the public sector ADM systems. These will be complementary to the online courses and build on them. The specialised course for the public sector will focus on promoting trustworthiness and inclusion within governmental organisations. Participants will explore the specific considerations and strategies needed to build on one side the critical thinking and from the other side the trust among citizens and stakeholders, ensuring transparency, accountability, and ethical practices in deploying ADM systems. The course will delve into topics such as bias and discrimination, governance frameworks, stakeholder engagement with the specific focus to the risks of discrimination and bias in ADM systems equipping public sector professionals with the knowledge and tools necessary to foster trust and inclusion in their work. For the trainees of both target groups the materials are printed in 32 pages brochure (ET) for each trainee in Estonia, all materials are first available electronically. The material will be adapted to each country's needs. Deliverable D4.1 – Report of the testing the public ADM systems in Estonia Deliverable Number D4.1 Lead Beneficiary 4 - MKM Deliverable Name Report of the testing the public ADM systems in Estonia Type DEM — Demonstrator, Dissemination Level PU - Public pilot, prototype Due Date (month) 17 Work Package No WP4 17 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Description This report details the testing of the impact assessment checklist for potential biases and discriminatory practices within the ADM systems used in the public sector. It includes the application of the impact assessment checklist to three real solutions utilized in the Estonian public sector. The process aims to evaluate the discrimination risk profile associated with these ADM systems. Based on the testing results, the impact assessment checklist and guidelines will be refined to aid system developers, data scientists, data stewards, and public service employees in identifying and mitigating discrimination and bias during the design, implementation, and development stages of ADM systems. This testing will set the benchmark for comprehensive testing of all systems in the future. The report on testing the public ADM systems in Estonia encompasses the outcomes of two primary activities: 1. Methodology and Process Description: This section outlines the methodology and process of testing the impact assessment checklist for potential biases and discriminatory practices within the public sector's ADM systems. It includes the results of applying the checklist to three real solutions in the Estonian public sector, setting a precedent for future comprehensive system testing. 2. Evaluation of Discrimination Risk Profile: This part evaluates the discrimination risk profile of the ADM systems in use, assessing the results based on real AI use cases from Estonia. It considers areas for improvement to enhance the assessment experience and the accuracy of the results. The report is a minimum of 25 pages, written in English and Estonian, and presented in a digital document format (pdf document). It will be made available on the government website dedicated to the responsible development and use of algorithms in the public sector. The primary audience includes system developers, data scientists, data stewards, and project partners, who will provide final input to refine the developed impact assessment checklist in Deliverable 3.1. The final version of Deliverable 3.1 will be updated following the testing of the public ADM systems in Estonia. Deliverable D5.1 – Report of the training for the Estonian and Lithuanian employees of public sector organisations Deliverable Number D5.1 Lead Beneficiary 3 - TalTech Deliverable Name Report of the training for the Estonian and Lithuanian employees of public sector organisations Type OTHER Dissemination Level SEN - Sensitive Due Date (month) 23 Work Package No WP5 Description Face to Face trainings for the public authorities will be carried out to deepen the digital skills and critical thinking to identify the possible discrimination risks on the public sector ADM systems. Training will be carried out in Tallinn, Estonia and in Vilnius, Lithuania. Each training will be 4 hours. Report will be in English, length 10 pages (word-document), include invitation/agenda, summary report of the trainings, signed attendance sheets and photos. In total 150 employees in public sector will be trained: 2x in Estonia in ET, 2x50 participants 3x in Lithuania in LT, together 50 participants. Deliverable D5.2 – Report of the training for the data scientists and data stewards training Deliverable Number D5.2 Lead Beneficiary 4 - MKM Deliverable Name Report of the training for the data scientists and data stewards training 18 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Type OTHER Dissemination Level SEN - Sensitive Due Date (month) 23 Work Package No WP5 Description Face to face training for the data scientists and data stewards. The training program tailored for the public sector will concentrate on data management for AI deployment. Training will be carried out in Estonia in Tallinn, in Lithuania in Vilnius. Each training will be 4 hours in Estonia and 2 hours in Lithuania. Report will be in English, length 10 pages, include invitation/agenda, summary report of the trainings, signed attendance and photos. 2x Estonia in ET, 2x20 participants 2x Lithuania in EN/LT, 2x20 participants Deliverable D5.3 – The report on the study visits Deliverable Number D5.3 Lead Beneficiary 3 - TalTech Deliverable Name The report on the study visits Type R — Document, report Dissemination Level SEN - Sensitive Due Date (month) 9 Work Package No WP5 Description The report on the study visit will summarize the exchange of experiences and knowledge gained, contributing to the advancement of more inclusive and fair practices within the materials, guidelines, and training produced in the EquiTech project. The visits will be made to two other European countries that are also in the process of combating discrimination through the implementation of ADM systems in the public sector. The objective of the study visit is to observe and learn from the approaches taken by other European countries in addressing similar challenges and understanding discrimination cases. 10 individuals will visit one state, and another 10 will visit a different state. The duration of each study visit will be three days. The location will be chosen during the project; one of the states will be France. The report (around 5 pages) will be written in English and will provide an overview of the visits, including pictures and signed attendance sheets from each visit. Deliverable D5.4 – Report of the implementation of the online course for public sector organizations in Estonia and Lithuania. Deliverable Number D5.4 Lead Beneficiary 3 - TalTech Deliverable Name Report of the implementation of the online course for public sector organizations in Estonia and Lithuania. Type R — Document, report Dissemination Level SEN - Sensitive Due Date (month) 24 Work Package No WP5 Description Report of the implementation of the online course for public sector organizations in Estonia and Lithuania (the content of the online course - D3.2). The Deliverable is the result of the Task 5.4. The online course for the employees of the public sector will be on two platforms: 1. For the employees of national, regional and local authorities in ET: Approximately 100 employees will take the course. 19 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 2. For the employees of national, regional and local authorities in Lithuania in LT: Approximately 50 employees will take the course. The number of participants who completed the course during the project implementation will be extracted from the platform and provided in the report. Deliverable D6.1 – Report of the international conference on discrimination in ADM systems Deliverable Number D6.1 Lead Beneficiary 1 - SVV Deliverable Name Report of the international conference on discrimination in ADM systems Type OTHER Dissemination Level SEN - Sensitive Due Date (month) 23 Work Package No WP6 Description Report on the International Conference on Discrimination in ADM Systems An international conference on discrimination in ADM systems (1,5 days) will be organized in Tallinn. This face-to-face conference aims to assemble Equality Bodies, experts in ADM systems within the public sector, ADM developers, and scientists from across Europe to share their expertise and insights. It is anticipated that at least 75 participants (including around 20 foreign experts) will attend. The primary objective is to foster mutual learning and explore methods for applying collective knowledge to combat discrimination within AI systems. The conference will host keynote speeches in a large auditorium and facilitate round table discussions in smaller, more intimate groups. These sessions are designed to encourage experts from diverse fields to engage in in-depth conversations and exchange experiences. The event strives to foster an environment that promotes collaborative learning, knowledge dissemination, and the development of unified strategies to tackle discrimination in ADM systems. The English-language report (around 3 pages) will be supported by the agenda, materials such as invitations and presentations, signed attendance sheets, and multimedia content like pictures and videos. Deliverable D6.2 – High-category scholarly article Deliverable Number D6.2 Lead Beneficiary 3 - TalTech Deliverable Name High-category scholarly article Type R — Document, report Dissemination Level PU - Public Due Date (month) 24 Work Package No WP6 Description The high-category scholarly articles will aid in disseminating the knowledge created through the EquiTech project and raise awareness of methods to identify and mitigate bias and discrimination in public sector ADM systems. The above- referred academic workshop and international conference allow to further tackle the topic in a scientific community. The high-category scholarly article (6 000 words) in English will have open access and will be accessible to more than 3000 academics and readers. Published in the following academical journalt: TalTech Journal of European Studies https://sciendo.com/journal/BJES Scopus and WoS indexed Open access 20 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Deliverable D6.3 – Report on Public Awareness Raising Activities Concerning Potential Risks and Biases in Public ADM Systems Deliverable Number D6.3 Lead Beneficiary 1 - SVV Deliverable Name Report on Public Awareness Raising Activities Concerning Potential Risks and Biases in Public ADM Systems Type R — Document, report Dissemination Level PU - Public Due Date (month) 24 Work Package No WP6 Description Report on Public Awareness Raising Activities Concerning Potential Risks and Biases in Public ADM Systems This deliverable will consist of a comprehensive report on the public awareness-raising activities. The report is divided into two parts: 1. A report on the Public Awareness Raising campaign in Estonia. 2. A report on the Public Awareness Raising campaign in Lithuania. The report will include media plan for both states, separate statistical reports on the outreach across all utilized channels, screenshots of social media posts, and other campaign types such as ads, press releases, articles, interviews with experts, as well as appearances on local TV and radio shows. In addition, all the designs created for the awareness campaign will be compiled. The materials will be produced in two languages: (ET and LT). The planned channels to be utilized include: - National broadcasts: minumum 4 TV and 8 radio interviews, both in the morning and evening programmes - 10 articles in various newspapers and online publications, targeting different demographic groups - Paid (1) campaign on online media and websites through Google Ad offered channels for Estonian market - Paid (1) campaign on online Media in Estonia - Paid advertisements on Meta channels, LinkedIn, and YouTube. - minimum of 30 different postings, adding the variety of creation for different channels and target groups (LinkedIN - employers, Facebook - citizens, employees 35+, Instagram - younger citizens, employees). Paid boost for 10 postings. - 8 press releases, difficult to say how many articles will be created out of the press releases, target would be 20. - 10 interviews with experts in the targeted channels and blogs - Outdoor campaign in Lithuania (outdoor advertisements placed in bus stop shelters and other locations - 200 placements in the largest cities in Lithuania for 2 weeks) The Public awareness reach as 300000 in Estonia and 500000 in Lithuania are coming from the experience with the former projects and the media campaigns. Just as an example, giving an interview only in the evening show in Estonian public broadcast will give a reach around ca 100000 with the follow-up audience, adding an prime time interview in a private TV channel of 40000, the online media with 60000, and the target groups in Meta channels, Youtube and minor group in LinkedIN. In academia the target group or 30 has the reach over the face to face workshop, same time the high- category scholarly article in English will be accessible to more than 3000 academics and readers. Deliverable D6.4 – Summary Report of the Academic Workshop Deliverable Number D6.4 Lead Beneficiary 3 - TalTech Deliverable Name Summary Report of the Academic Workshop Type R — Document, report Dissemination Level SEN - Sensitive Due Date (month) 24 Work Package No WP6 Description The academic workshop will be designed to address the subject through scientific discourse, engaging other scientists in discussions and thereby disseminating knowledge on discrimination to new research areas and scientific debates. The in-person workshop aims to introduce the topic to the broader academic community, sparking interest among other 21 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 researchers in exploring issues of discrimination related to AI tools. The workshop will target 30 participants, primarily academics keen on advancing their understanding of topics related to non-discriminatory AI. The workshop (around 4 hours) will take place in Tallinn. The summary report (around 3 pages), written in English, will be supported by the agenda and invitations, presentations/ training materials, and signed attendance sheets, along with photos of the event. Deliverable D6.5 – Policy Development: Recommendations and Strategies for ADM Systems to Combat Discrimination Deliverable Number D6.5 Lead Beneficiary 3 - TalTech Deliverable Name Policy Development: Recommendations and Strategies for ADM Systems to Combat Discrimination Type R — Document, report Dissemination Level PU - Public Due Date (month) 24 Work Package No WP6 Description This section outlines policy recommendations and strategies regarding ADM systems aimed at combating discrimination. These recommendations will be provided by Estonia (in ET) and Lithuania (in LT) to the relevant policy institutions within each state. The primary target group is policymakers, specifically ministries responsible for ensuring equal treatment through technology. These recommendations and strategies will be delivered through official letters. The estimated length of the policy recommendations for both Estonia and Lithuania is between 5-10 pages. The recommendations will be available as professionally designed PDF documents on the websites of all project partners and the national equality bodies. Potential recommendations could encompass a broad spectrum of actionable measures. These may include, but are not limited to, requiring the execution of an impact assessment checklist during the procurement process, establishing systematic bias and discrimination training for certain public office roles, advocating for periodic audits and reviews of ADM systems, and encouraging multi-stakeholder dialogue. At the national level, this project will augment state policies and regulations related to AI-based discrimination. It will help prepare for the implementation of AI Act (AIA) regulations, assist in resolving discrimination cases stemming from algorithm-based decisions, and encourage collaboration between the public sector and developers. Furthermore, the project aims to enhance the Estonian National AI Strategy by assessing the current landscape and offering policy recommendations to promote a human-centric approach to AI implementation. This effort includes defining relevant strategic objectives, Key Performance Indicators (KPIs), and identifying necessary resources to advance the domain. Deliverable D6.6 – Evaluation of the Value and Impact Created by the Activities in the Final Phase of the EquiTech Project Deliverable Number D6.6 Lead Beneficiary 3 - TalTech Deliverable Name Evaluation of the Value and Impact Created by the Activities in the Final Phase of the EquiTech Project Type R — Document, report Dissemination Level PU - Public Due Date (month) 24 Work Package No WP6 Description This document outlines the framework for an internal evaluation aimed at assessing the impact generated by the EquiTech project's activities during its final phase, specifically within Estonia (ET) and Lithuania (LT). The purpose of this evaluation is to systematically examine the outcomes and overall value brought about by the project's initiatives in these 22 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 states. This analysis will serve as a critical tool for internal stakeholders to understand the project's effectiveness, areas of success, and opportunities for improvement. The evaluation will be meticulously prepared and structured to encompass a comprehensive short review of five pages in EN. It will include quantitative and qualitative assessments of the project's contributions to advancing knowledge, policy development, and practical implementations in the realms of equality and technology. Through this evaluation, the project team aims to capture and articulate the tangible and intangible benefits realized in Estonia and Lithuania as a direct result of the EquiTech project's endeavors. This internal document will not only reflect on the achievements and challenges faced but also provide insights into future directions and strategies for sustaining and expanding the project's impact. The evaluation's findings are intended for internal use, to inform both strategic planning and operational improvements for ongoing and future initiatives. 23 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 LIST OF MILESTONES Milestones Grant Preparation (Milestones screen) — Enter the info. Milestone Milestone Name Work Package No Lead Beneficiary Means of Verification Due Date No (month) 1 Kick-off partner meeting WP1 1 - SVV Meeting agenda and attendance list 2 2 The Ethics Guideline WP1 1 - SVV The Ethics Guideline approved by the Project 6 Management & Quality Board 3 Legal analyses WP2 3 - TalTech Draft of the report of the legal analysis in 6 Estonia and Lithuania approved by the Project Management & Quality Board. 4 Report of the auditing WP2 3 - TalTech Drafted report of the audit approved by the 6 Project Management & Quality Board 5 Finding the subcontrating partners WP3 5 - JUM Signed subcontracts with all partners. 8 6 Guidelines WP3 5 - JUM Draft of the guidelines 17 7 Impact assessment checklist WP3 4 - MKM Draft version of approved by the Project 17 Management & Quality Board. 8 Online course WP3 5 - JUM First “offline” versions ready on Estonian and 17 Lithuanian portal. 9 Training materials WP3 5 - JUM The training materials are approved by the Project 17 Management & Quality Board. 10 Mapping the ADM systems to be tested WP4 4 - MKM The plan of the testings is approved by the Project 14 Management & Quality Board. 11 Testing WP4 4 - MKM The report is approved by the Project 17 Management & Quality Board. 12 Study trips WP5 1 - SVV 2 study trips have successfully taken place. 6 13 Trainings in Estonia WP5 3 - TalTech 4 trainings have successfully taken place. 22 24 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Milestones Grant Preparation (Milestones screen) — Enter the info. Milestone Milestone Name Work Package No Lead Beneficiary Means of Verification Due Date No (month) 14 Trainings in Lithuania WP5 2 - OEOO 5 trainings have successfully taken place. 22 15 Media plan WP6 1 - SVV Media plan approved by Project Management & 20 Quality Board. 16 Planning of the conference WP6 1 - SVV The conference has successfully taken place. 23 17 Evaluation WP6 3 - TalTech Report based on the project aims and 24 requirements. LIST OF CRITICAL RISKS Critical risks & risk management strategy Grant Preparation (Critical Risks screen) — Enter the info. Risk Description Work Package Proposed Mitigation Measures number No(s) 1 Potential obstacles to secondments of partners, WP3, WP1, WP6, Where possible, the time plan will be rearranged to reflect the unexpected changes. If not such as experts, academic staff, participation in the WP5, WP2, WP4 possible, alternative representatives of the partners will be appointed. conference, such as health, personal issues, etc. 2 Problems in the division of tasks WP3, WP1, WP6, The coordinator with each representative of each partner will have online meetings to divide WP5, WP2, WP4 the work and establish clear patterns of collaboration avoiding conflicts on the division of tasks. 3 Gender equality WP3, WP1, WP6, EquiTech includes in all its activities a gender-based analysis, a project design considering WP5, WP2, WP4 gender issues, project implementation includes gender-based analysis, project staff is trained in gender equality issues. 25 Project: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Critical risks & risk management strategy Grant Preparation (Critical Risks screen) — Enter the info. Risk Description Work Package Proposed Mitigation Measures number No(s) 4 Low interest in cooperation from the entities using WP4 Beforehand communication and clarifying the importance of the assessment tool. AI solutions to assess them 5 A partner is not aware of expectations WP3, WP1, WP6, As the current proposal was written as a collaborative effort of all partners, they are all WP5, WP2, WP4 familiar with the text and know what kind of commitment is expected of them. Expectations of all partners are sufficiently managed. 6 Insufficient coordination between activities/partners WP3, WP1, WP6, The project’s progress will be constantly monitored by the Coordinator based on a work plan WP5, WP2, WP4 and communication plan. These will be prepared before the first meeting and revisited at every further meeting. 7 The deliverables (publications, workshop etc) show WP3, WP1, WP6, The Coordinator will be engaged in a constant quality monitoring process. If deliverables low or insufficient quality WP5, WP2, WP4 do not meet quality standards, discussion will be organized in order to find possibilities for suitable corrective action. Issues related to the quality of deliverables will be discussed in the first meeting and standards agreed. All deliverables will be cross-checked by Coordinator for quality before being officially delivered. 8 Insufficient commitment WP3, WP1, WP6, Partners have been screened for selection during the preparation of this proposal. WP5, WP2, WP4 Expectations and responsibilities have been clarified from all sides. Risk associated with a member of the teams changing affiliation are also seen as low. 26  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 1 Citizens, Equality, Rights and Values Programme (CERV) Description of the action (DoA) Part B Version 1.0 April 2024  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 IMPORTANT NOTICE What is the Description of the Action (DoA)? The Description of the Action (DoA) is the Annex of the Grant Agreement which contains the details of how the project will be carried out. For EU framework partnerships for grants (FPAs) this Annex is called Action Plan. It consists of 2 parts, which must be generated from the submitted proposal: • Part A contains structured tables with project information • Part B is a narrative description on the work to be carried out. Part A is generated by the IT system. It is based on the information which you enter into the Portal Grant Preparation screens. Part B (+ annexes) must be uploaded on the Grant Preparation Documents screen. Make sure that Part B is synchronised with the information entered into the screens. Make sure that any changes are agreed with us. Page 2 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 PROJECT Grant Preparation (General Information screen) — Enter the info. Project number: 101144709 Improving response to risks of discrimination, Project name: bias and intolerance in automated decision- making systems to promote equality. Project acronym: EquiTech Call: CERV-2023-EQUAL Topic: CERV-2023-EQUAL Type of action: CERV-LS Service: JUST Project starting date: May 1, 2024 Project duration: 24 Page 3 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 TABLE OF CONTENTS VERSION 1.0....................................................................................................................................................................... 1 TECHNICAL DESCRIPTION (PART B) ............................................................................................................................. 5 1. RELEVANCE ............................................................................................................................................................... 5 1.1 Background and general objectives ....................................................................................................................... 5 1.2 Needs analysis and specific objectives ................................................................................................................ 12 1.3 Complementarity with other actions and innovation — European added value ................................................... 15 2. QUALITY ................................................................................................................................................................... 16 2.1 Concept and methodology .................................................................................................................................... 16 2.2 Consortium set-up ................................................................................................................................................ 17 2.3 Project teams, staff and experts ........................................................................................................................... 18 2.4 Consortium management and decision-making ................................................................................................... 21 2.5 Project management, quality assurance and monitoring and evaluation strategy ............................................... 21 2.6 Cost effectiveness and financial management ..................................................................................................... 22 3. IMPACT ..................................................................................................................................................................... 22 3.1 Impact and ambition ............................................................................................................................................. 22 3.2 Communication, dissemination and visibility ........................................................................................................ 24 3.3 Sustainability and continuation ............................................................................................................................. 26 4. WORKPLAN, WORK PACKAGES, ACTIVITIES, RESOURCES AND TIMING ...................................................... 27 4.1 Work plan ............................................................................................................................................................. 27 Subcontracting (n/a for prefixed Lump Sum Grants) ........................................................................................... 30 Timetable ............................................................................................................................................................. 32 5. OTHER ....................................................................................................................................................................... 35 5.1 Ethics and EU values ........................................................................................................................................... 35 5.2 Security ................................................................................................................................................................. 35 6. DECLARATIONS ....................................................................................................................................................... 35 Page 4 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 TECHNICAL DESCRIPTION (PART B) 1. RELEVANCE 1.1 Background and general objectives Background and general objectives Describe the background and rationale of the project. How is the project relevant to the scope of the call? How does the project address the general objectives of the call? What is the project’s contribution to the priorities of the call? Which target groups will be supported/assisted by/in the project. Why have you chosen to focus on them? Background and rationale This project proposal stems from the need to understand and address the bias and risks that automated decision-making systems (ADM) may pose for fundamental rights and gender equality. Both Estonia and Lithuania face increasing use of ADM- s in the public sector, yet knowledge about these risks is scarce. Estonia is an advanced digital state and has a an ambition to remain a leader of innovation in e-governance1. Most of its public services can be accessed online and there is a fast-growing number of algorithmic solutions integrated into various public procedures and administrative decisions. Similarly, Lithuania has been developing digital solutions in the public sector. However, an urgent and rapidly evolving challenge lies in ensuring that the algorithmic solutions deployed by the public sector would adhere to the established anti-discrimination legal framework. While the ever-increasing adoption of IT solutions (including AI systems)2 the public sector has numerous positive benefits, such as better access to public services (including for vulnerable groups) as well as more efficient and high-quality decision-making. Digitalisation and automation in the public sector inevitably also leads to numerous new challenges, including concerning potential discrimination and biases. For example, in the SyRI case in the Netherlands, it was revealed that the algorithms used by the Dutch tax authorities incorrectly classified approximately 26,000 parents as fraudulent in their child care benefit applications, many of whom had immigration background.3 It emerged that the data used in the system was discriminatory, leading to welfare applications by immigrants being more likely to have been classified as fraudulent. Similarly, the use of the COMPAS recidivism assessment system by the US courts led to black people being more likely to be assessed as recidivist, impacting their eventual sentence by the court.4 While these are perhaps the most famous cases of discrimination by public sector ADM systems, there are numerous other discriminatory practices within AI and other algorithmic systems reported daily.5 These incidents provide a valuable reminder how the public sector decision-making systems can bring about extensive and systematic discrimination of vulnerable groups if biases in the data and the algorithms are not correctly identified and removed. Until now, no cases regarding discrimination in ADM systems have reached the Estonian or Lithuanian Equality Bodies. However, as mentioned above, bias in the ADM systems is a difficult and ever-increasing problem. Many of these issues are already evident, even if not reported. For example, AI-based predictive policing systems are increasingly used by law enforcement to predict and/or prevent fraud or other forms of crime. Such systems have a high potential for discrimination – as they rely on historical data, they could inadvertently amplify and perpetuate existing patterns of over-policing in specific communities or regions.6 However, this does not get reported as the discriminatory patterns are not transparent or recognized. It's also often very difficult to definitively prove that an algorithm is discriminatory. Thus, immediate steps must be taken to make sure that there is sufficient awareness and know-how among public officials, system developers but also other actors that more or less directly can help to supervise these systems, such as Equality Bodies, academia and the general public. The continuous digitalisation of public administration requires an increasingly robust approach7 to identify and mitigate these risks8. While Estonia9 and Lithuania have an extensive legal framework to tackle discrimination, there is an evident lack of means and competence to implement the principle of non-discrimination into practical solutions during the digitalisation process. 1 This amibition is also set forward in the Estonian Government’s Digital Agenda 2030, accessible (also in English) here: According to the Estonian Digital Agenda 2030, the aim of using of ICT in daily life and business is to improve economic competitiveness, the well-being of the people, and the efficiency of public administration, thus serving the society at large. Kerikmäe, T., & Pärn-Lee, E. (2021). Legal dilemmas of Estonian artificial intelligence strategy: in between of e-society and global race. Ai & Society, 36, 561-572. 2 Classified by the European Union as one of the “most strategic technologies in the 21st century”, Antonov, A., & Kerikmäe, T. (2020). Trustworthy AI as a Future Driver for Competitiveness and Social Change in the EU. The EU in the 21st Century: Challenges and Opportunities for the European Integration Process, 135-154. 3 Rechtbank Den Haag, C/09/550982, 05.02.2020, ECLI:NL:RBDHA:2020:1878. See also Toh, A. (2020), ‘Dutch Ruling a Victory for Rights of the Poor’ (Human Rights Watch, and Shumilo, O., & Kerikmae, T. (2021). The European approach to building AI policy and governance: a haven for bureaucrats or innovators?. IDP: revista de Internet, derecho y política= revista d'Internet, dret i política, (34), 4. 4 See Angwin, J, Larson, J, Mattu, S and Kirchner, L, ’Machine Bias‘, Propublica, 5 See, for example, the AI Incident Database - https://incidentdatabase.ai. 6 See Alikhademi Kiana et al. (2022), A review of predictive policing from the perspective of fairness, Artificial Intelligence and Law, Vol. 30, pp. 1–17 and Bacchini Fabio/Lorussi Ludovica (2019), Race, again: how face recognition technology reinforces racial discrimination, Journal of Information, Communication and Ethics in Society, Vol. 17 No. 3, pp. 321–335. 7 Likewise taking precautions via privacy design approach under GDPR. Hoffmann, T. (2020). The impact of digital autonomous tools on private autonomy. Available at SSRN 3771306. 8 The increasing possibility to undertake various transactions without direct human contact, via different electronic means also affects the protection of fundamental rights and may influence the rule of law. Kerikmäe, T., & Metcalf, K. N. (2020). The rule of law and the protection of fundamental human rights in an era of automation. In Smart Technologies and Fundamental Rights (pp. 221-239). Brill. 9 Kerikmäe, T., Metcalf, K., Hoffmann, T., Minn, M., Liiv, I., Taveter, K., Shumilo, O. Solarte Vasquez, M.C. & Antonov, A. (2019). 1st Report on Legal Framework and Analysis Related to Autonomous Intelligent Technologies.(1− 11). Page 5 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 Estonian and Lithuanian legal acts and policies follow the principles provided by the Racial Equality Directive10 and of the Framework Decision on combating certain forms and expressions of racism and xenophobia11, also Commission Recommendation C (2018) 3850 on standards for equality bodies. Estonian gender equality and equal treatment policies are governed by the Equal Treatment Act and guided by the Estonian Welfare Development Plan (2023-2030)12. The Plan includes objectives and actions to reduce inequalities based on grounds set in the Equal Treatment Act. The policy has two distinct parts: gender equality and equal treatment. The European Institute for Gender Equality’ Gender Equality Index scores Estonia with 61 points (2022), while the Welfare Development the target level for 2030 at 67.5. Percentage of people who consider themselves as belonging to a group discriminated against in Estonia is currently 7,3% (2021) and should be decreased by 2030. The policy has two distinct parts: gender equality and equal treatment. Annex to the Plan13 also stipulates that ”In an increasingly digital society, attention must also be paid to the fact that gender stereotypes do not only affect human decision-making – they can also be transmitted, either consciously or subconsciously, into ADM, including through the use of AI- and algorithm-based technologies that not only reproduce but also reinforce stereotypes and biases.“ The Lithuanian Office of the Equal Opportunities Ombudsperson (Lithuanian Equality Body), as the guardian of equal treatment and prevention of discrimination, has identified an urgent need in building the knowledge on the risks AI may pose. This would support the body to efficiently solve discrimination complaints linked to the use of algorithms. Meeting the new challenges to equality and non-discrimination from increased digitalisation and the use of AI is a new role for Equality Bodies. Collaborating with other institutions responsible for state digitalisation will help enhance the knowledge and skills of Equality Bodies. General Objective and Scope of the Project The general objective of the EquiTech project is to enhance the capacity of public authorities and their co-operation partners in Estonia and Lithuania in addressing risks of discrimination and bias in public sector ADM systems. Discrimination is a negative result of bias. For this reason, the term “bias” (as a cause for algorithmic discrimination) has been used in the project alongside the term "discrimination”. The term “ADM systems'' throughout the application refers to all forms of algorithms – including machine learning and other AI models as well as logic-based systems, expert systems and other more conventional algorithms, which are used in the public sector. These algorithms can be used to fully automate a certain public procedure or a public task, or they can be used as a fact-finding tool or as systems to provide suggestions for the public officials in undertaking administrative decisions that impact the rights or obligations of people.14 The EquiTech project’s main focus will be to develop a comprehensive set of support tools and materials that will assist civil officials and their co-operation partners to develop and deploy AI and other IT systems in a manner that ensures fairness and compliance with the legal framework on discrimination. This also includes mapping of risks and solutions in existing systems, developing guidelines as well as carrying out training to increase awareness and competence among public authorities, developers and other relevant parties to support the fight against discrimination. The EquiTech project tackles all forms of possible discrimination within the ADM systems. For example, one of the notable forms of discrimination evident in the ADM systems may stem from gender bias, which can be embedded in the AI or other algorithmic model or data by design, reflecting and perpetuating existing societal norms or the personal biases of those who design the systems. The desk research will reflect the academic literature that has addressed gender-bias in the ADM systems. During mapping and auditing of existing ADM system, we will assess whether there is any sex-disaggregated data, gender-sensitive variables or other similar approaches used in developing the models, which may indicate a gender bias. We will seek to understand whether in the course of developing the existing systems any specific measures were taken to identify and minimise possible biases. The subsequent guidelines and other materials and trainings will build on these findings, and propose specific measures for public sector officials and system developers in mitigating the possible risks. In addition to gender bias, the project will also address other forms of discrimination such as racial or ethnic bias, socio-economic bias, and age bias that may be present in the ADM systems. These biases can similarly be embedded in the algorithms or data, potentially leading to unfair outcomes for certain groups. Finally, it is important to outline that one of the specific goals of the EquiTech project is sharing the experience about developing solutions in identifying and mitigating risks concerning discrimination and bias in ADM systems gained in Estonia and Lithuania with the EU community – e.g. all key deliverables will be (at minimum) translated to English and there will be various dissemination activities with an international perspective. The EquiTech project has three supportive organizations: EQUINET, the European Network of Equality Bodies, the Non-Discrimination Ombudsman in Finland and the Federal Anti-Discrimination Agency in Germany (FADA) (see the Annexes). All of these organizations are interested in the research and deliverables that contribute to combating the acute problem in Europe. Relevance to call scope 4 10 European Union. (2000). Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin. Official Journal of the European Communities, L 180, 22-26. 11 European Council. (2008). Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law. Official Journal of the European Union, L 328, 55-58. 12 Welfare Development Plan 2023-2030. Available: https://www.sm.ee/heaolu-arengukava-2023-2030#heaolu-arengukava-20 13 Annex to the Welfare Development Plan. Available: https://www.sm.ee/heaolu-arengukava-2023-2030#heaolu-arengukava-20 14 For example, a machine vision tool SATIKAS by the Estonian Agriculture and Registration Agency uses satellite photos from the European Space Agency to detect whether beneficiaries have maintained the fields (cut crass) before the deadline, which determines whether they have a right to a benefit. See European Association of Remote Sensing Companies (2021), A Case Study Grassland Monitoring in Estonia Page 6 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 CALL ACTIVITIES PROJECT ACTIVITIES Coalition building, The consortium is proposing a series of coalition-building events with the partner organisations capacity building in Estonia and Lithuania. The purpose of these activities is to develop a common understanding and training for across the life cycle of ADM systems development and implementation in Estonian and professionals and Lithuanian and engage state institutions and ADM developers for system improvement related victims of to the risk of discrimination and bias. (intersectional) discrimination; The EquiTech project includes tasks focused on capacity building: AND ● Creating and providing an impact assessment checklist in 3 languages (ET, LT, EN). 15 pages per language. Developing an impact assessment checklist for identifying and assessing potential Promotion of risks of discrimination and bias within the ADM system. The checklist is structured to ask a series of digital skills and questions or list considerations concerning these areas. It builds on and refers back to the guidelines in critical thinking; answering specific questions on how possible risks are being identified and mitigated. The goal of the assessment checklist is to proactively uncover potential issues that may arise when deploying an ADM system. It allows public officials and developers to address these risks in the design and implementation stages, thereby minimising negative consequences. The assessment checklist will target both the public officials as well as system developers. It will be made available in 3 languages. ● Guidelines development for system developers, scientists, data stewards, and public service providers to detect discrimination and bias in the design, implementation, and development of ADM systems. Approximately 60 pages per language (ET, EN, LT). Guidelines is the document, which gives a step-by-step overview on how to use the impact assessment checklist, explaining assessment process and ways to avoid discrimination while developing and using AI solutions. The materials created under this activity will provide concrete technical and organisational recommendations through the whole lifecycle of the ADM system – including the initial data collection, algorithm design, model testing, output analysis, post-deployment risk management, etc. Among else, the guidelines will cover: - How to carry out inclusive system design. That is, to ensure that the ADM is designed with consideration for diverse groups with specific feature such as gender, race, disability and else other forms of discrimination. - How to identify and reduce biases in data and in the algorithmic model. - Recommendations on public consultation during the system development, including engaging with marginalised groups. - How to carry out testing of a new ADM system for identifying potential risks of discrimination. - How to measure the impact the AI system has on fairness post-deployment. ● Creating and implementing an online course for Estonian and Lithuanian employees of national, regional and local authorities. These courses would be designed to increase awareness and understanding of discrimination and bias in public sector ADM systems. They would cover topics such as the potential sources of bias in AI, the impact of bias on decision-making, and basic suggestions for mitigating bias to avoid discrimination. The online course for the employees of the public sector will be on two platforms: 1. For the employees of national, regional and local authorities in ET 60-minutes online course, consisting of 30 pages of text content per language and up to 5 videos (3-5 min) in Estonian and in English on Estonian Digital Academy (digiriigiakadeemia.ee). 2. For the employees of national, regional and local authorities in Lithuania in LT Online learning platform for Lithuanian public sector: administrated by Lithuanian Equality Body, is a learning space where public sector employees have the possibility to gain knowledge on equality topics. The platform has 4 online courses (topics: gender mainstreaming through strategic planning and service provision; equal opportunities and the creation of an inclusive working environment) at the moment and is intended for both public and private sector. ● Developing training materials and organising training on avoiding discrimination risks for Estonian and Lithuanian employees of national, regional and local authorities. Materials 25 pages per language. Trainings for the public authorities (2x ET, 3x LT, altogether 150 trainees) Training of public authorities will be carried out to deepen the digital skills and critical thinking to identify the possible discrimination risks in the public sector ADM systems. These will be complementary to the online courses and build on them. The specialised course for the public sector will focus on promoting trustworthiness and inclusion within governmental organisations. Participants will explore the specific considerations and strategies needed to build on one side the critical thinking and from the other side the trust among citizens and stakeholders, ensuring transparency, accountability, and ethical practices in deploying ADM systems. The course will delve into topics such as bias and discrimination, governance Page 7 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 frameworks, stakeholder engagement with the specific focus to the risks of discrimination and bias in ADM systems equipping public sector professionals with the knowledge and tools necessary to foster trust and inclusion in their work. ● Developing training materials and organising the training in avoiding discrimination risks for the system developers, data scientists and data stewards (2x Estonia (40 trainees), 2x Lithuania (40 trainees)) Materials of 25 pages in ET and EN. The training materials will assist in the training system developers, data scientists and data stewards about the complexities and potential risks concerning discrimination and bias in public sector ADM systems. Participants will learn about data governance framework, data quality assurance, and the responsible use of data in the public sector context with specific focus on risks of discrimination and bias in ADM systems. Mutual learning, Engaging relevant stakeholders for mutual learning and experience sharing between the two exchange of good participating countries. practices, Collaboration between Estonian and Lithuanian Equality Bodies gives an opportunity for mutual cooperation, learning and exchange of good practices. This is valuable for promoting equality and combating including discrimination, especially in emerging areas like AI systems. Estonia and Lithuania can benefit from identifying best each other's experiences and practices in addressing discrimination issues related to AI. Through practices which collaboration, Estonian and Lithuanian Equality Bodies can develop a stronger understanding of the may be challenges and opportunities presented by AI systems in relation to discrimination. By exchanging ideas, transferable to experiences, and good practices, they can work together to solve cases related to discrimination in ADM other participating systems and work out preventive actions for discrimination. countries; Collaboration between Estonian ministries, Tallinn University of Technology (TalTech) and Equality Bodies in Estonia and Lithuania will not only enhance knowledge and skills regarding discrimination in ADM systems but also facilitate future cooperation related to discrimination matters. The outcomes of EquiTech project can be beneficial not only to other EU member states but also to a broader audience.By working together, the Estonian ministries, TalTech, and Equality Bodies will contribute to a deeper understanding of the intersection between discrimination and AI in digital systems. This collaboration will facilitate the exchange of expertise and best practices, leading to improved approaches in addressing todays and increasing discrimination challenges in AI systems. The project's results will have a significant impact, as they can serve as a valuable resource for other EU member states and beyond to learn about risks of discrimination and bias in ADM systems and contribute to the development of more inclusive and fair digital systems in various jurisdictions. Study visits to other EU Equality bodies with practice of discrimination on ADM systems will offer an opportunity to observe and learn from the approaches taken by other European countries in tackling similar challenges and learning from the discrimination cases. Given that discrimination risks in ADM systems are becoming a growing concern worldwide, it is crucial to learn from and exchange experiences with other countries. As the number of discrimination cases continues to rise, it is important to stay informed about the latest developments in order to effectively respond and mitigate risks in the future. The aim is to gain valuable insights and exchange experiences related to ADM implementation, regulation, cooperation among various stakeholders in the field, especially among public authorities, and its impact on discrimination. Each partner organisation will send two representatives to participate in the study visits. The visits will take place in two different European countries, allowing for a diverse range of perspectives and practices to be explored. The exchange of experiences and knowledge will contribute to the advancement of more inclusive and fair practices within the materials, guidelines and trainings produced in the EquiTech project. As part of the EquiTech project, study visits will be organised to two other European countries who are also in the process of combating discrimination through the implementation of ADM systems in the public sector. These study visits offer an invaluable opportunity to observe and learn from the approaches taken by other European countries in addressing similar challenges and understanding discrimination cases. The knowledge and experiences gained from these visits will be incorporated into the project deliverables, including the impact assessment checklist, guidelines, and training materials. An international conference on discrimination in ADM systems will be organised in Tallinn. The conference aims to bring together experts on ADM systems in the public sector, ADM developers, and Equality Bodies and scientists from across European countries to share their expertise and insights. The primary goal is to facilitate mutual learning and explore ways to apply the combined knowledge of discrimination and AI systems. The conference will feature keynote speeches in a large auditorium, as well as round table sessions in small groups. These sessions will provide an opportunity for experts from both sides to engage in tailored conversations and exchange experiences. The conference is expected to attract approximately 75 participants from diverse sectors. By gathering Page 8 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 experts from different backgrounds, the conference seeks to create an environment conducive to collaborative learning, knowledge sharing, and the identification of common strategies to address discrimination in ADM systems. Academic workshop organised by TalTech will introduce the topic to the wider academia to raise the interest of other researchers to deal with the discrimination topic regarding AI tools. Number of participants 30, notably academics who are interested in developing the knowledge of non-discriminative AI related topics. Dissemination and The EquiTech project will focus on academic dissemination and both professional and public awareness awareness-raising, raising about potential discrimination in ADM procedures including general knowledge about including social discrimination as such. media or press Awareness raising among public sector employees, software developers and data scientists. campaigns; The project will help to understand problems and risks concerning discrimination in public sector ADM systems and provide understanding on how to address these risks. It’ll help the responsible parties to ensure compliance of ADM systems with laws and state policies concerning discrimination. It will also help to prepare for the implementation of the EU AI Act. Improving Estonian and Lithuanian National AI Strategy. The EquiTech project will map as-is situation and provide policy recommendations to advance the human-centric implementation of AI within the country, together with relevant strategic objectives, Key Performance Indicators (KPI) and resources needed to advance the field. For the general public the EquiTech project will raise awareness through a media campaign on the possible risk of discrimination when using digital services. Specifically vulnerable groups can be more vulnerable to discrimination by the ADM systems. Higher awareness will enable these vulnerable groups to navigate the ADM landscape more effectively, recognize potential biases, and actively advocate for fair and equitable treatment. By acknowledging and addressing the risks of discrimination within ADM systems, stakeholders can work towards fostering a more inclusive and just society in all European Union. Academic dissemination. The high-category scholarly articles will aid in disseminating the knowledge created through the EquiTech project and raise awareness of methods to identify and mitigate bias and discrimination in public sector ADM systems. The above-referred academic workshop and international conference allow to further tackle the topic in a scientific community. Data recording, The Estonian government currently possesses an expansive digital database, boasting a commendable data collection, level of security in handling its data. However, there remains an ongoing imperative to comprehend the surveys, appropriate procedures for managing this data in the context of ADM. Specifically, it is crucial to establish monitoring and a framework for secure and transparent data processing that refrains from discriminatory practices reporting of against citizens. This entails addressing three key aspects: incidents of 1. the identification of discrimination cases through data analysis; discrimination; 2. the implementation of effective support mechanisms; 3. the establishment of robust protocols for reporting such incidents. By conscientiously addressing these concerns, the Estonian state can ensure a fair and just approach to data processing that upholds the principles of security, transparency, and non-discrimination. Conducting a literature review on discrimination and bias in ADM systems. The EquiTech project will undertake a thorough literature review on discrimination and bias in ADM systems. This review aims to develop a systematic understanding of the extent of these risks and identify optimal solutions to address them. It will give a holistic view of the current situation of the discrimination topics related to the ADM systems. By conducting this comprehensive review, the EquiTech project aims to contribute to the advancement of knowledge and practices in addressing bias and discrimination in ADM systems, ultimately fostering more fair and equitable outcomes. Legal analyses in Estonia and Lithuania. Equinet has opened a discussion regarding AI based discrimination. One task they have proposed to the Equality Bodies is to conduct a legal “gap analyses” regarding the risk of discrimination in AI systems. The idea is to analyse the legislation to find such legal caps which can lead to the discrimination. By analysing the legal regulations related to AI the legal gaps will be found if there are any and the suggestions for law-drafting will be given to improve them. A legal analysis will be conducted in both countries – Estonia and Lithuania, focusing on the existing legal provisions and regulations. This analysis will help in identifying the legal framework surrounding bias and discrimination in ADM systems and provide valuable insights for future actions and improvements. Auditing discrimination and bias within Estonian public sector ADM systems and current assessment practice. The EquiTech project aims to establish an auditing process specifically designed to identify discrimination and bias risks within these systems and evaluate the current assessment practices in place in Estonia. By implementing the auditing process, the project seeks to enhance transparency and Page 9 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 accountability, while also promoting the development of more inclusive and fair ADM practices. Testing the impact assessment checklist for potential biases and discriminatory practices within the ADM systems used in the public sector. The impact assessment checklist will be used to test potential biases and discriminatory practices within the ADM systems employed in the public sector. This process will generate an evaluation of the discrimination risk profile associated with the ADM systems used in the public sector. Based on the results of the testing, the impact assessment checklist and guidelines will be enhanced to assist system developers, data scientists, data stewards, and public service employees in identifying discrimination and bias during the design, implementation, and development stages of ADM systems. The activity includes testing out impact assessment checklist on at least 3 real solutions used in the Estonian public sector. The testing is giving the scale for real testing for all the systems in the future. Gender and age While Estonian and Lithuanian law provides protection against discrimination based on gender, race, sensitive victim age and other grounds, it is essential to acknowledge that these forms of discrimination persist in various empowerment and contexts. Public authorities have a vital role in addressing these issues effectively. By enhancing their support; capacity to identify and address discrimination, public authorities can contribute to a more equitable and inclusive society. As already outlined above, many of the project activities intend to increase the capacity of public sector employees and their co-operation partners in ensuring that the ADM systems enhance empowerment and involvement of marginalized groups. For example, the guidelines, the impact assessment and the training materials will, among else, focus on engaging with all relevant stakeholders during the ADM system design phase, to account for the possible concerns. Among other things, this includes ensuring that people of all ages and regardless of technical knowledge would have good accessibility to the algorithmic service and understanding its impacts. Many of the other procedural measures covered by the guidelines in tackling bias in the algorithms will also focus on empowering and supporting the people subject to the ADM systems. Furthermore, raising awareness among the general public on the possible risk of discrimination when using digital services is a measure to increase empowerment and control over the quality of these systems. These automated systems can inadvertently perpetuate biases and discriminatory outcomes. By promoting awareness and understanding of the risks associated with gender and age discrimination in ADM systems, stakeholders can take proactive measures to prevent and rectify such issues. Thus, ensuring the capacity of public authorities and fostering awareness regarding potential discrimination in ADM systems are vital steps in combating gender and age discrimination effectively. Designing and One of the EquiTech project activities is developing policy recommendations that reflect the insights implementing from the preceding stages of the project. The recommendations can provide valuable input to shaping strategies or plans the future governance and policy decisions concerning ADM systems in the public sector, impacting the of action; future national policies on digital state or AI strategies. Meanwhile, the project itself helps to advance several existing policies. The Estonian Digital Agenda 203015 establishes human-centric digital government as one of its four key objectives. This means ensuring that the digital public services would ensure high level protection of fundamental rights, including as regards to discrimination. Among other things, it foresees that “when developing and offering public services, it is guaranteed that everyone has an equal opportunity to participate in digital society and use digital services”.16 To achieve that, the document foresees developing measures to manage risks to human rights and raising the awareness of the general public about the digital public services.17 EquiTech project directly fulfils several aims regarding equal treatment of Estonian Welfare Development Plan, inter alia, improve the knowledge and skills of policy makers, decrease the equality index in gender and minority groups and supporting Estonian Equality Body in preventing the discrimination caused by the development of technology. Project activities support Lithuanian Strategy on AI (developed in 2019) by building capacities of both public and private sector representatives in order to create ethical and lawful ADM systems. In addition, one of the project activities is mapping the discrimination and bias risk profile of several ADM systems used in the public sector. This evaluation helps the responsible public authorities in understanding the level of discrimination risk associated with these systems and can be valuable input for their decisions on using or further developing the particular ADM systems. Capacity building Capacity building among the employees of national, regional and local authorities is a key instrument and training through which the EquiTech project seeks to achieve the goals of mitigating discrimination and bias in activities for public sector ADM systems. As explained previously, this will be mainly achieved through developing a national, regional toolbox to identify and mitigate discrimination in ADM systems and raising know-how among the 15 See here: https://www.mkm.ee/digiriik-ja-uhenduvus/digiuhiskonna-arengukava-2030 16 Ibid, p 17. 17 Ibid, p 18. Page 10 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 and local responsible actors. The main activities include: authorities. 1. Development of guidelines, impact assessment checklist, training materials and free online courses for public sector employees that provide concrete support in understanding the potential risks for discrimination and bias in ADM systems as well as various technical and organization measures that can be taken to mitigate these biases. 2. Carrying out specialized training for the employees of national, regional and local authorities are involved with designing, developing and deploying ADM systems. 3. Carrying out capacity building among the project team – which involves public sector employees - through study trips and collaboration throughout this project, including sharing of knowledge between Estonian and Lithuanian colleagues. 4. Carrying out capacity building through international conference and academic workshop. Contribution to call objectives The EquiTech project aims to strengthen Estonia’s and Lithuania's institutional capacity to address inequality within ADM systems. The rapid development of technology and the widespread use of ADM systems can inadvertently lead to discriminatory outcomes. Currently, there is limited research on how AI systems effectively address bias and ensure non-discrimination. Through the project, in-depth research and mapping will be conducted on the existing ADM systems in the public sector, encompassing all forms of discrimination. The research and subsequent activities of the EquiTech project will strive to increase competence and awareness among relevant stakeholders to facilitate effective identification and mitigation of bias and discrimination in public sector ADM systems. By undertaking these actions, the project aims to prevent and combat discrimination in all its protected forms. The goal is to promote the development and implementation of ADM systems that uphold principles of fairness, equality, and non- discrimination. By enhancing the understanding and capabilities of Estonia and Lithuania in this regard, the EquiTech project contributes to fostering more equitable and inclusive ADM systems. Furthermore, the knowledge and practices gained through this project have the potential to be shared and adopted by other European countries, promoting widespread adoption of learnable and adaptable approaches to combating all forms of discrimination in ADM systems. While Estonian and Lithuanian law and policy follow the EU policies in equal treatment promoting equality and supporting the fight against racism, xenophobia and discrimination, the result of this project should provide competence and know-how in implementing these rules into practical solutions, in the context of ADM systems. As a result, we will minimize the probability of discrimination and bias in Estonian and Lithuanian ADM systems and hope to set an example to other EU member states, contributing to the national action plans against racism and strategies against antisemitism, anti-Muslim hatred and xenophobia, LGBTIQ-phobia and all other forms of intolerance. Contribution to call priorities The EquiTech project strengthens the authority of Equality Bodies of Estonia and Lithuania by enhancing their capacity in investigating and fighting against discrimination. Collaboration between these two institutions gives an opportunity for mutual learning and exchange of good practices. This is indeed valuable for promoting equality and combating discrimination, especially in emerging areas like AI. Collaboration of Estonian ministries, Tallinn University of Technology (TalTech) and Equality Bodies in Estonia and Lithuania will not only enhance knowledge and skills regarding discrimination in ADM systems but also facilitate future cooperation related to discrimination matters. Creating teaching materials and trainings will provide knowledge and tools to teach also local governments who do not participate the provided trainings, later after the project about discrimination with focus on AI. The outcomes of this project can be beneficial not only to other EU member states but also to a broader audience. As mentioned above the scope of the project is directly related to the Estonian policy regarding discrimination which inter alia cover all the following policies and legal acts: Racial Equality Directive and of the Framework Decision on combating certain forms and expressions of racism and xenophobia, Commission Recommendation C(2018) 3850 on standards for equality bodies as well as develop and implement national action plans to combat racism, xenophobia, LGBTIQ-phobia and all other forms of intolerance, and to develop and implement national strategies against antisemitism and the Council Conclusions on Combating Racism and Antisemitism of 4 March 2022, Member States have committed to developing national action plans against racism and strategies against antisemitism by the end of 2022. Target groups As the EquiTech project is focused on mitigating discrimination and bias in the course of developing and deploying ADM systems in the public sector, the key target groups of the project are public officials, including equality experts from equality bodies, and system developers, data scientists, data stewards developing and operating the ADM systems. These target groups play a central role in the design, development, deployment, and ongoing management of these ADM systems. As such, they are uniquely positioned to influence the fairness, non-discrimination, and impartiality of the systems. Public officials have a responsibility to uphold the public trust, which includes ensuring that ADM systems are employed in a manner that is fair and does not perpetuate discriminatory practices. Support documents such as the guidelines, impact assessment checklist and training materials will provide specific suggestions and insights to the domain experts (officials who have domain knowledge regarding the algorithm’s scope of application), data protection officers, legal advisers, project leader but also to management of the public agency deploying the ADM systems. Another aspect of the measures focuses on the system developers, data scientists and data stewards used in designing, developing, operating and maintaining the public sector ADM systems. They have the technical expertise necessary to understand and tackle the intricate problems of bias in machine learning models and algorithms that drive the systems. Page 11 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 As both groups are most likely to be directly involved in the decision-making processes around the deployment of the systems and have the greatest influence over their design and quality, then focusing the measures on these groups can have the most immediate and most significant impact on the fair use of ADM systems in the public sector. However, there are also some activities that target the general public and academia. The media awareness campaign is directed at the general public, in order to educate the wider public about the potential biases and discriminatory practices that can arise from ADM systems. Particular attention will be given to vulnerable groups who face potential risks of discrimination associated with ADM systems. By raising awareness among vulnerable populations about the risks associated with these systems, the project can empower them to understand their rights, recognize discrimination, and take appropriate actions. The measures directed at the academic community include workshop and publishing a high-category scholarly article. Increasing collaboration with academia helps to further understand the risks concerning these systems, developing innovative methods to reduce bias, and providing insights that inform policy and practice. By sharing knowledge with academics researching the field, the project can contribute to the ongoing scholarly discourse on addressing discrimination and bias in ADM systems. 1.2 Needs analysis and specific objectives Needs analysis and specific objectives (n/a for Programme Contact Points) Provide a needs assessment. A need is a gap between what is and what should/ would be helpful or useful. The needs assessment should be your starting point. Specify what needs will be addressed and how they have been identified. It should be specific and focus on the actual needs of the target group. It should include relevant, reliable data and, a robust analysis clearly demonstrating the need for the action (therefore, avoid references to generic statements and information about the problems and needs of the target group). The needs assessment should incorporate gender equality issues and non-discrimination considerations that identify the differences between and among women and men, girls and boys, in terms of their relative position in society and the distribution of resources, opportunities, constraints and power in a given context. The data supporting the needs assessment should be dissaggregated by sex, as well as age or disability, whenever possible. You can refer to existing research, studies and previous projects that already demonstrate the need for action. If your project is supported by a public authority, annex the Letter of support. Needs analysis According to the Estonian Digital State strategy18 nearly all direct public services have been digitalised one way or another by now (p 6). Many of the algorithms used in such services have an important role within administrative procedure, such as identifying facts or data, providing decision-support or communicating with people, thereby directly affecting various administrative decisions and the rights, freedoms and obligations of people. While these systems must adhere to the rules on fairness and non-discrimination, in practice there is generally a lack of knowledge among officials, developers and project leaders on how to assess and mitigate biases. In the Estonian context, a similar concern has been observed by the ministries responsible for ensuring that the algorithmic solutions in the public sector would remain human centric and in compliance with the fundamental rights (i.e. the Estonian Ministry of Justice and the Estonian Ministry of Communication and Economic Affairs).19 The limitations in knowledge are in part due to the lack of proper training of the officials, developers and project leaders responsible for these systems. Additionally, there is a lack of practical support materials that could be relied on, in assessing and mitigating the aforementioned risks. These needs are not specific only to Estonia and Lithuania but are rather common in all EU countries and elsewhere, since there is globally an increased focus on deploying public sector ADM systems. As the project undertakes to advance the best practices in developing solutions in identifying and mitigating risks, and as one of the goals is sharing the experience gained in Estonia and Lithuania with the EU community as well as translating the key deliverables (at minimum) to English, the project can help to tackle the need also in other EU countries. Knowledge and skills granted by the project are, among else, valuable in preparing for the application of EU AI Act. Furthermore, there are several digital policies of EU and Council of Europe, which the activities of this project support. For example, CoE “Preventing discrimination caused using AI: Resolution 2343” (2020), Recommendations 2183 “Preventing discrimination caused by the use of artificial intelligence” (2020), Digital Agenda 2022-2025, CAI Towards regulation of AI systems (2020), EU Ethics Guidelines for Trustworthy AI, AI for Sustainability, and EU Sustainable Development Goals. General objective The general objective of the EquiTech project is to enhance the capacity of public authorities and their co-operation partners in Estonia and Lithuania in addressing risks of discrimination and bias in the ADM system. This will be achieved by developing research, support materials, training and a media campaign. This project seeks to contribute to the overall reduction of inequality within the European Union context by ensuring that the ADM systems used in the public sector are fair and non-discriminatory. ADM continues to grow in scale. It is widely known that algorithm-based decisions have a risk for biases, but it is less known how to avoid this. Some of these biases can lead to discrimination, which often may be systematic and wide-scale. There is, thus, a need to consider more thoroughly the possible implications of the ADM systems on discrimination. It is evident that institutions using algorithmic decisions in practice do not know clearly how to consider, measure and mitigate unfairness. This includes both, the employees of the public agencies but also the software developers or other co-operation partners. A clear guidance is needed to raise the capacity and capability of public institutions to investigate algorithmic discrimination. In addition to public institutions, there are also other key stakeholders, such as the academic community and citizens, who play 18 Estonian Digital Agenda 2030. Available: https://www.mkm.ee/digiriik-ja-uhenduvus/digiuhiskonna-arengukava-2030 19 These observations have notably been made as part of early experimental government support services such as the Estonian data panel and sandbox that provide legal guidance to public sector algorithmic projects (several members of these experimental support services are also part of the team of this project). Page 12 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 an important role in helping to identify and supervise discrimination in ADM systems. Thus, the EquiTech project also engages these stakeholders in order to increase their awareness of the issue. Specific objectives As noted above, the underlying goal is to develop awareness and competence among public sector officials and their project partners and provide guidance in identifying and mitigating discrimination and bias. Pursuing the below specific objectives of the project will enable reaching the general objective: 1. Development of a systematic understanding of the extent and risk of potential discrimination and bias in ADM systems in Estonia and identification of optimal solutions to address them. (WP2) Currently, there is a lack of understanding on how to effectively monitor and identify potential biases in ADM systems to mitigate the risks of discrimination and bias. To address this issue, the EquiTech project will undertake a thorough literature review on discrimination and bias in ADM systems. This review aims to develop a systematic understanding of the extent of these risks within the Estonian context and identify optimal solutions to address them. By conducting this comprehensive review, the EquiTech project aims to contribute to the advancement of knowledge and practices in addressing discrimination and bias in ADM systems, ultimately fostering more fair and equitable outcomes. 2. Conducting a legal analysis of Estonia and Lithuania to detail the legal framework surrounding discrimination and bias in ADM. (WP2) There is a need to gain an overview and understanding of the legal framework pertaining to discrimination and bias in ADM systems, specifically within the contexts of Estonia and Lithuania. To achieve this, a legal analysis will be conducted in both countries, focusing on the existing legal provisions and regulations. This analysis will help in identifying the legal framework surrounding discrimination and bias in ADM systems and provide valuable insights for future actions and improvements. 3. Establishment of an audit process and methodology designed to identify discrimination and bias risks within ADM systems and evaluation of the current assessment practices in general and in Estonia. (WP2) Currently, there is no existing practice of conducting audits for discrimination and bias within Estonian public sector ADM systems. The EquiTech project aims to establish an auditing process specifically designed to identify discrimination and bias risks within these systems and evaluate the current assessment practices in place. The project recognizes the importance of addressing potential discrimination and biases in ADM systems used within the Estonian public sector. By implementing the auditing process, the project seeks to enhance transparency and accountability, while also promoting the development of more inclusive and fair ADM practices. Today, there is no established methodology for checking discrimination and bias risks within the Estonian public sector ADM system. To address this gap, the EquiTech project aims to develop an impact assessment checklist specifically designed to identify and assess potential risks of discrimination and bias within the ADM system. The impact assessment checklist will be structured to include a series of questions and considerations related to these areas. It will be based on and refer to existing guidelines, providing specific guidance on how to identify and mitigate potential risks. By utilising this checklist, public officials and developers can proactively address these risks during the design and implementation stages, thereby minimising negative consequences. 4. Development of guidelines to detect discrimination and bias during the design, implementation, and development of the Estonian public sector ADM systems. (WP3) Testing the practical value of the materials (WP4). Currently, there are no existing guidelines to effectively detect discrimination and during the design, implementation, and development of the Estonian public sector ADM systems. This lack of guidelines makes it challenging to minimize discrimination risks in practice. However, as part of the EquiTech project, comprehensive guidelines will be developed to address this gap. The EquiTech project will create guidelines for two key stakeholders: 1) system developers, data scientists, and data stewards, and 2) public service employees. These guidelines will provide essential guidance and support documentation to help organisations to understand how to assess discrimination and bias during the system's design phase and throughout its lifecycle. Moreover, the guidelines will propose measures to mitigate potential risks. The support materials created for the project will be tested on actual ADM systems to enhance their practicality and ensure their effective utilisation. This iterative process enables the project to increase the practicality and real-world applicability of the materials, ultimately contributing to the overall effectiveness in identifying and mitigating bias and discrimination risks within the Estonian public sector ADM systems. 5. Development of training materials and modules and training of Estonian and Lithuanian employees for national, regional and local authorities, stakeholders and professionals working with ADM systems on the risks of discrimination and bias and how to avoid or mitigate them. (WP3, WP5) To date, no practical training materials have been produced to educate stakeholders in the Estonian and Lithuanian public sectors or stakeholders about possible discrimination and bias risks in ADM systems. However, employees who work daily with ADM systems play a crucial role in detecting discrimination when it occurs. As part of EquiTech project, one of the deliverables is the creation and implementation of an online course specifically designed for public sector representatives in Estonia and Lithuania. It will be made available to public sector organisations in both countries. Additionally, the EquiTech project includes the development of training materials and the organisation of training sessions. The objective is to train employees of public sector organisations in Estonia and Lithuania, as well as data scientists and data stewards involved in public sector ADM systems, on how to avoid discrimination risks effectively. By developing tailored training materials and organising relevant training programs, the EquiTech project aims to enhance the understanding and capabilities of employees working with ADM systems in the public sector. These training materials are adaptable and can be freely translated for use in all European countries. Page 13 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 Currently, employees of national, regional and local authorities in Estonia and Lithuania who work with ADM systems do not receive any training on avoiding discrimination risks. To address this gap, the EquiTech project aims to provide training for public authorities in both countries. A total of two trainings will be conducted in Estonia, and three trainings will be conducted in Lithuania. These trainings will complement the online courses and further build upon the knowledge gained from them. The trainings for public authorities will focus on enhancing digital skills and critical thinking to effectively identify potential discrimination risks within the public sector ADM systems. During the trainings, participants will delve into specific considerations and strategies necessary for fostering critical thinking. They will also explore methods to build trust among citizens and stakeholders, ensuring transparency, accountability, and ethical practices in the implementation of ADM systems. By offering these trainings, the EquiTech project aims to deepen participants' understanding of the complexities surrounding discrimination risks in ADM systems. The trainings seek to empower public authorities with the necessary skills and knowledge to navigate these challenges effectively, promoting fairness, equity, and ethical decision-making within the public sector. Currently, also ADM developers in the Estonian and Lithuanian public sector do not receive any training on avoiding discrimination and bias risks in the ADM systems. To address this gap, the EquiTech project will organise training programs specifically designed for data scientists and data stewards. These training programs will focus on mitigating discrimination risks in the public sector. The training program tailored for the public sector will primarily concentrate on effective data management for AI deployment. Given that public sector organisations often handle vast amounts of sensitive data, ensuring proper data stewardship is crucial for the successful implementation of AI-driven initiatives. During the training sessions, participants will learn about important aspects such as data governance frameworks, data quality assurance, privacy regulations, and the responsible use of data within the context of the public sector. The aim is to equip participants with the knowledge and skills necessary to manage data in a way that minimises discrimination risks and promotes fairness and transparency. The training programs will be conducted four times, with two sessions taking place in Estonia and two sessions in Lithuania. To maximise participation and ensure a broader impact, the training will be conducted in Estonian and English, allowing a wider range of individuals to benefit from the program. By offering these trainings, the EquiTech project aims to enhance the understanding and capabilities of ADM developers, data scientists, and data stewards in identifying and addressing discrimination risks in ADM systems within the public sector. Ultimately, this will contribute to the responsible and ethical use of AI technologies and promote fairness and non-discrimination in decision-making processes. 6. Development of an impact assessment checklist for identifying and assessing potential risks of bias and discrimination within the Estonian public sector ADM system (WP4) One of the objectives of the EquiTech project is to develop an impact assessment checklist for identifying and assessing potential risks of bias and discrimination within the Estonian public sector ADM system. This will provide the first opportunity to test the developed method on public sector ADM systems. A minimum of five different public sector ADM systems will be subjected to testing. The goal of the assessment checklist is to proactively uncover potential issues that may arise during the deployment of an ADM system. It empowers public officials and developers to address these risks during the design and implementation stages, thereby minimising negative consequences. Additionally, the support materials created for the project will be tested to enhance their practicality and ensure their effective utilisation. By conducting testing on the public sector ADM systems and assessing the impact of the support materials, the EquiTech project aims to refine and improve the checklist and associated resources. This iterative process enables the project to increase the practicality and real-world applicability of the materials, ultimately contributing to the overall effectiveness in identifying and mitigating bias and discrimination risks within the Estonian public sector ADM system. 7. Academic dissemination, general communication and exploitation of results (WP6) There is currently a significant knowledge gap in every sector, but also in the academic world regarding both theoretical and practical understanding of discrimination and bias risks in ADM systems. To address this issue, the EquiTech project aims to enhance knowledge and promote awareness among academics and professionals through the dissemination of a high-quality scholarly article and a workshop. These activities will serve to bridge the knowledge gap and foster a deeper understanding of the complexities surrounding bias and discrimination in ADM systems. By engaging with academia and professionals, the EquiTech project seeks to generate valuable insights, promote research collaboration, and encourage the exchange of expertise. This concerted effort will contribute to the development of robust theoretical frameworks and practical strategies for addressing bias and discrimination risks in public sector ADM systems. Overall, the project strives to promote the advancement of knowledge and scholarship in this field, ultimately leading to more informed decision-making, improved practices, and a more equitable and inclusive use of ADM systems in the public sector services. Currently, there is a lack of systematic and research-backed knowledge among professionals in Estonia and Lithuania regarding the discrimination and bias risks within the public sector ADM systems. Furthermore, the general public is generally unaware of the potential discrimination risks associated with these systems. To address this issue, the EquiTech project aims to raise awareness among the general public about the possibilities of discrimination when using digital services. It is important to recognize that the general public, especially vulnerable groups, can become victims of discrimination facilitated by ADM systems. By increasing awareness, these vulnerable populations can navigate the ADM landscape more effectively, identify potential biases, and actively advocate for fair and equitable treatment. Through this process, stakeholders can acknowledge and address the risks of discrimination within ADM systems, working towards fostering a more inclusive and just society throughout the European Union. The EquiTech project seeks to empower both professionals and the general public with the knowledge and understanding needed to recognize and mitigate discrimination risks in the use of ADM systems. By promoting awareness and advocating for fairness, the project aims to create a more equitable and inclusive digital environment for all. #@COM-PLE-CP@# Page 14 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 1.3 Complementarity with other actions and innovation — European added value Complementarity with other actions and innovation (n/a for Programme Contact Points) Explain how the project builds on the results of past activities carried out in the field and describe its innovative aspects. Explain how the activities are complementary to other activities carried out by other organisations. Illustrate the European dimension of the activities: trans-national dimension of the project; impact/interest for a number of EU countries; possibility to use the results in other countries, potential to develop mutual trust/cross-border cooperation among EU countries, etc. Which countries will benefit from the project (directly and indirectly and why have you chosen them)? Where will the activities take place? Clarify to what extent the project builds on synergies with other EU projects. If applicable, explain to what extent your project builds on previous project results in this field (state of play, relation to existing/recent developments, approaches, achievements, other EU programmes). Note: The project should also complement or add benefits to the EU Member States' interventions in the area of gender equality and non- discrimination mainstreaming. Although the auditing, development of support materials and testing and improvement of those materials take specifically place in Estonia and Lithuania, the project has a clear intention of pushing forward the best practices across the EU in developing effective solution to tackle discrimination in public sector algorithms. Firstly, the concerns with discriminatory AI systems are similar around the world and there is a lot of research and reports outlining the issue.20 Meanwhile, there is globally an increased focus on deploying public sector ADM systems, which are often accompanied by various policies on EU and national level to fight possible bias and discrimination that may emerge within these systems. However, it can be observed that the policies and recommendations developed so far in the EU are oriented to outlining the risks and issues that need to be solved, rather than develop concrete solutions. For example, the AI HLEG guidelines on trustworthy AI and the other deliverables based upon it21 help to understand problems concerning bias and discrimination risks in AI, but do not provide actual solutions or suggestions on how to test for bias, how to improve the understanding of system’s data, which measure to introduce to identify risk, etc. Our aim is to be more specific and produce as a result a concrete toolbox for fighting against discrimination regarding the ADM system. Estonia as a digital state with ambitious public sector IT solutions is very well suitable for the research concerning such human-centric digital governance tools. We argue that the deliverables developed can be easily transferred to every other EU country. Results of the project – specifically impact assessment checklist, guidelines, online course and training materials – will provide valuable contributions to other countries in the EU by giving them tools and knowledge to tackle discrimination and bias in their systems. Sharing the experience and solutions developed in Estonia and Lithuania with the EU community is specifically established as one of the goals of the project. To achieve that, the project will translate the key deliverables (at minimum) to English and includes an extensive dissemination campaign, in the form of drafting a final report to map the effects achieved through the project, developing academic literature as well as by organising an international conference at the end of the project. The existing guidelines that have been developed in several other EU countries (e.g. NL22 and FR23), by the EU24 and elsewhere (e.g. UK25) will also be taken as input for this project. The intention of the project is to build on them and thereby advance the solutions in fighting against the discrimination and bias. The project also has a strong link to other activities in the EU. One of the most important developments in this domain is the EU AI Act26 (AIA) which, upon its adoption, is going to be the fully harmonising regulation concerning AI in the EU Member States. While the AIA is still in the trialogue negotiation phase, it can be expected that numerous ADM system used in public sector, will fall under the definition of ‘artificial intelligence’ under the AIA and will be used in a context that will bring it under the scope of the AIA (e.g. used by law enforcement, within administration of justice, etc. – see notably Annex III sections 2, 3, 6, 7 and 827 . The AIA will be an important development also in light of tackling discrimination and bias in the AI systems, as it foresees a requirement for AI providers (and, in some cases, users) to take measures to assess and mitigate these risks. However, the AIA will not establish what specifically these measures are and, rather, relies on best practices and development of standards over time. The practical guidelines and solutions developed and deployed under our project can provide valuable empirical feedback to accelerate and improve the uptake of effective measures in identifying biases and discrimination that the AIA requires. Furthermore, the project intends to support cooperation and development of competence among European Equality Bodies in this matter. It is common that new topics regarding discrimination are discussed either by individual contacts or in the meetings organised by the Equinet. As cooperation can always be better – especially in the domain of the emerging technologies 20 ERA 2022. Bias on Algorithms – Artificial Intelligence and Discrimination. Doi: 10.2811/536044; Orwat C. 2019. Risks of Discrimination through the Use of Algorithms. A study compiled with a grant from the Federal Anti-Discrimination Agency. Available here; Centre fo Data Ethics and Innovation. 2020. Review into bias in algorithmic decision-making. Available here. Algorithmic discrimination in Europe Including summaries in English, French and German Challenges and opportunities for gender equality and non-discrimination law; European Commission; Algorythms and human rights, Council of Europe; #BigData: Discrimination in data-supported decision making, FRA 21 The AI HLEG’s Ethics Guidelines for trustworthy AI, Assessment List for Trustworthy AI and the Policy and Investment Recommendations 22 See the guidelines developed in Netherland on the non-discrimination by design (available only in Dutch) 23 In France, the Defender of Rights and the National Commission on Informatics and Liberty (CNIL) have issued recommendations to prevent algorithmic discrimination 24 Notably the AI HLEG’s Ethics Guidelines for trustworthy AI, Assessment List for Trustworthy AI and the Policy and Investment Recommendations Another key research in EU is the European Union Agency for Fundamental Rights study “Bias in Algorithms – Artificial Intelligence and Discrimination” (2022) 25 For example, see the UK Centre for Data Ethics and Innovation review into bias in algorithmic decision-making. Furthermore, the Alan Turing Institute has published a guide for the responsible design and implementation of AI systems in the public sector, which includes a chapter on fairness and non-discrimination. 26 European Commission (2021), Proposal for a Regulation of the European Parliament and of the Council laying down harmonized rules on artificial intelligence (Artificial Intelligence Act) and amending certain Union legislative acts, COM(2021) 206 final, Brussels, 21 April 2021. 27 The Commission proposal Page 15 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 such as AI and the increasing impact of data economy. The project promotes such cross-border and cross-institutional co- operation as it brings together Equality Bodies of two countries as well as relevant Ministries and academia. The intent is to also share them with the other Equality Bodies in the EU. Furthermore, the Equinet provides its support to the EquiTech project through a formal letter of support. The co-operation among equality bodies will further be supported by some of the specific project tasks, such as the study visits. Some of the other synergies with the key EU projects include: - AI HLEG (the high-level expert group on artificial intelligence) has developed Ethics Guidelines of Trustworthy AI as well as several other deliverables, such as policy and investment recommendations on trustworthy AI.28 One of its seven key requirements for Trustworthy AI is ensuring diversity, non-discrimination and fairness. To achieve that the guidelines suggest, among other, developing processes to analyse and address the system’s purpose, constraints, requirements and decisions; promote accessibility and user-centric design as well as ensure stakeholder participation in developing the system.29 All of these aspects will be covered by this project. - Work on algorithmic bias by the European Union Agency for Fundamental Rights (FRA). Their recent study outlines that the problem of bias in algorithms “still lacks a tangible evidence base that employs technical assessments of algorithms in practice and their outcomes. How exactly AI can lead to fundamental rights violations is not yet fully known. More evidence- based assessments are urgently needed to fill this gap.”30 The current project will have specific focus on remaining empirical and evidence focused. Furthermore, FRA has argued that “To increase the application of trustworthy AI, compliant with fundamental rights, more EU and national funding for fundamental rights assessments of existing software and algorithms is needed to support studies of available general purpose algorithms.”31 - The European Parliament Centre for AI (C4AI) produces studies, organises public events, and acts as a platform for dialogue and information exchange on AI-relevant topics.32 This project can provide empirical insights to the discussions of C4AI on mitigation bias and discrimination in cases where AI or other algorithms are used to automate public proceedings. - TEU anti-racism action plan 2020-202533 provides that “the use of algorithms can perpetuate or even stimulate racial bias if data to train algorithms does not reflect the diversity of EU society.” (p 9) Between hate and equality: the EU as a guard of human rights and non-discrimination researchesrisk of discrimination and violence motivated by discrimination women and LGBT+ persons. - “Together Against Discrimination” focuses on how to overcome prejudices against minorities, other nations or different sex and age group.34 #§COM-PLE-CP§# #§PRJ-OBJ-PO§# #§REL-EVA-RE§# #@QUA-LIT-QL@# #@CON-MET-CM@# 2. QUALITY 2.1 Concept and methodology Concept and methodology Outline the approach and methodology behind the project. Explain why they are the most suitable for achieving the project’s objectives. Include ethical and safety considerations to ensure that target groups are not subjected to harm in any way. Note: Methodology is not a list of activities but are instruments, approaches that will be used, applied and created. To fulfil our main project objective, the EquiTech project employs a five-phase model which includes, research, content creation, testing, training and dissemination components in order to reach to project outcomes in a comprehensive five phase model. Five phase model In Phase 1 (WP 2), we aim to conduct a thorough literature review, practical research and mapping in Estonia, as well as legal In Phase 1 (WP 2), we aim to conduct a thorough literature review and mapping the situation in Estonia, as well as legal analysis in Lithuania and Estonia regarding discrimination and bias risks in ADM systems. In this phase a mixed method approach will be carried out in order to gather information from different resources that will be the basis for Phase 2. In addition to this, as part of the practical research in Estonia, audits will be carried out within Estonian public sector ADM systems. Starting with the literature review, audits and analyses will enable all project partners to communicate on the concepts effectively. The objective of this phase is to create a systematic understanding of the extent of risks in the context of Estonia and Lithuania and understanding the best solutions to tackle them better. In Phase 2 (WP 3), we will create an Impact assessment checklist, guidelines and online course and develop special training materials. The objective of this phase is to develop guidance and other support documentation and training materials, which help organisations to understand how to assess discrimination and bias risks in system’s design phase and throughout its lifecycle and suggest measures to mitigate possible risks. In Phase 3 (WP 4), testing activities will be carried out. The objective of this phase is to test the support material that has been created for the project, which allows to increase their practicality and actual uptake. In Phase 4 (WP 5), we will carry out trainings, realise the set of three online courses. All material will be improved in accordance with the results of the testing phase before being used. Having different kinds of training to accommodate different needs of target groups will enable the project team to have a user-friendly approach. The objective of this phase is to ensure that the 28 The ethics guidelines and other deliverables developed by AI HLEG 29 Ibid, p 18-19. 30 European Union Agency for Fundamental Rights study “Bias in Algorithms – Artificial Intelligence and Discrimination” (2022), p 7. 31 Ibid, p 15. 32 https://www.europarl.europa.eu/stoa/en/centre-for-AI 33 A Union of equality: EU anti-racism action plan 2020-2025. 34 Together against discrimination, Project ID 101054046 Page 16 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 materials created are systematically disseminated and sufficient training is provided to support the awareness of the identified risks and uptake of the guidance documents. In Phase 5 (WP 6), dissemination activities will be carried out on national, academic and public levels. We will have a systematic approach in this phase, where dissemination activities will be specialised for the target groups to ensure to communicate the results effectively. The objective of this phase is to increase general awareness level of the individuals and organisations in the risks of discrimination and bias risks in ADM systems used. Integration of methods and disciplines to pursue the objectives: Our strategy for involving a variety of knowledge transfers and knowledge gathering is centred on both interdisciplinary and intersectoral approaches. Partners to the project are selected to ensure multidisciplinarity and enhance diversity and complementarity. Gender dimension and other diversity aspects: Our gender policy is fully consistent with the European Institute for Gender Equality’s Gender Mainstreaming Toolkit. Equal empowerment of women and men in all their diversity will be taken into account essentially in all project activities. We will also seek to reduce levels of discrimination suffered by particular groups (as well as those at risk of multiple discrimination) and improve equality outcomes for individuals. Research data management and management of other research outputs: In data using the project team will promote data transparency, quality, and visibility. Project team will use only collected anonymous data. All data collected and shared in this network will comply with relevant EU and non-EU legal regulations regarding personal data protection, as well as regarding the issues of research data generation, storage, and international transfer. #§CON-MET-CM§# #@CON-SOR-CS@# 2.2 Consortium set-up Consortium cooperation and division of roles (if applicable) (n/a for Town Twinning and Programme Contact Points) Describe the participants (Beneficiaries, Affiliated Entities and Associated Partners, if any) and explain how they will work together to implement the project. How will they bring together the necessary expertise? How will they complement each other? In what way does each of the participants contribute to the project? Show that each has a valid role and adequate resources to fulfil that role. Note: When building your consortium you should think of organisations that can help you reach objectives and solve problems. The quality of the consortium is expected to be considerable because: • It will leverage the experience of the numerous EU projects developed and implemented by the key staff members. • It includes state Equality Bodies of two EU countries. • It includes two ministries with the highest level of experts in their field. • It includes high-level academics from TalTech who have been carefully selected and have connections with other European universities. • The consortium has strong expertise in all areas of the project and promotes a wide range of activities thanks to the variety of its internal composition. The coordinator of the consortium is The Gender Equality and Equal Treatment Commissioner in Estonia (Estonian Equality Body), an independent and impartial expert who acts independently, monitors compliance with the requirements of the Gender Equality Act, and performs other functions imposed by law. The Commissioner's mandate of investigating complaints and conducting independent research on discrimination provides a solid foundation for understanding the discriminatory risks associated with AI systems. Collaboration with various institutions, agencies, and international organisations enhances its ability to engage in meaningful dialogue and exchange knowledge about discrimination and equality, including the risks of discrimination in ADM systems. The other Equality Body in the consortium is the Office of the Equal Opportunities Ombudsman of Lithuania (Lithuanian Equality Body), which is highly relevant to the project's goal of preventing various types of discrimination through AI awareness raising in Lithuania. The Lithuanian Equality Body’s mandate of investigating complaints and conducting independent research on discrimination provides a solid foundation for understanding the discriminatory risks associated with AI systems. By leveraging their expertise in identifying and analysing different forms of discrimination, the office can effectively assess how AI algorithms and systems may perpetuate bias and unequal treatment. Its role in providing conclusions and recommendations on discrimination-related issues is invaluable when formulating strategies and guidelines for mitigating bias and promoting fairness in AI development and deployment. Furthermore, both Equality Bodies in the consortium engage in preventive and educational activities that closely align with the objectives of the project call. Through communication, training, and consultation, they have established effective channels for raising awareness about discrimination and promoting equal opportunities. This experience can be leveraged to design and implement an AI awareness raising campaign that targets the general public and educates them about the discrimination risks associated with AI technology. Moreover, the collaboration with various institutions, agencies, and international organizations enhances their ability to engage in meaningful dialogue and exchange knowledge about discrimination and equality. This network can be tapped to gather insights, share best practices of the EquiTech project, and seek partnerships in the context of addressing risks of discrimination and bias in public sector ADM systems. The academic partner of the consortium is TalTech, one of the most innovative universities in the Baltic States. TalTech is the sole technological university in Estonia, offering Bachelor's, Master's, and Doctorate degrees in technology, applied science, IT, business, and maritime studies. TalTech places a very high priority on international cooperation and participates in numerous academic networks and cross-border research projects. The Department of Law at TalTech is an international research Page 17 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 institution based in Tallinn, focusing on a broad spectrum of issues involving law, technology, and EU law, including fundamental rights. Under the EquiTech project, they provide academic expertise in the field of law and academic literature. Furthermore, the EquiTech project team at TalTech includes an independent IT expert, who adds valuable neutral viewpoints to the analyses and research. The EquiTech project consortium is complemented by two ministries in Estonia: the Ministry of Economic Affairs and Communications and the Ministry of Justice. The Ministry of Economic Affairs and Communications is one of the main stakeholders responsible for the digitalization and innovation of the public sector in Estonia. The ministry supports the creation of capacity for the deployment of AI solutions in public sector bodies, raises awareness of the potential of AI to add value to businesses and the public sector, develops strong competencies and skills in AI-related topics, and has a general overview of the AI field in the Estonian public sector, as well as involvement in the creation of many AI solutions currently in use. Their expertise and access contribute significantly to the research and deliverables of the EquiTech project, and they have the legal right and a high interest in using the EquiTech deliverables to combat all forms of discrimination. The Ministry of Justice is the main stakeholder responsible for planning and carrying out the judicial and criminal policy of the state, ensuring an open and secure society where people can be assured of the use and protection of their rights. The ministry strengthens the principles of a democratic state based on the rule of law by guaranteeing the fundamental rights of the people and structuring state management. As such, they are also responsible for the legislative policy concerning non-discrimination, and the applicability of this rights in the context of public sector ADM systems. The expert team of the Ministry of justice have a high competence on fundamental rights at EU level and human rights at European and UN level. The governability of the consortium is ensured by selecting a coordinating partner with comprehensive experience in managing EU and innovative projects of similar size and complexity, as well as leveraging appropriate management structures and procedures. The efficient operations of the project are also facilitated by the fact that the consortium is built on existing successful collaborations between individual partners. Additionally, the input in the evaluation of test results would be beneficial for the project's main objectives. In terms of the mix of competences required for the project, our consortium has been selected primarily based on the expertise each partner brings. However, we have also paid attention to how several overlapping considerations make the consortium optimal, including the subject matter, the discipline of activity, and the geographical angle. The first profile to be explained is the Call for proposals to promote equality and fight against racism, xenophobia, and discrimination (CERV-2023-EQUAL) and TOPIC ID: CERV-2023- EQUAL - CERV-LS CERV Lump Sum Grants. 2.3 Project teams, staff and experts Project teams and staff Describe the project teams and how they will work together to implement the project. List the staff included in the project budget (budget category A) by function/profile (e.g. project manager, senior expert/advisor/researcher, junior expert/advisor/researcher, trainers/teachers, technical personnel, administrative personnel etc. and describe briefly their tasks. Provide CVs of all key actors (if required). Note: Please ensure a gender-balanced representation in the composition of project teams and staff performing the action. Name and function Organis Role/tasks/professional profile and expertise ation Christian Veske, Gender SVV Christian Veske is the Gender Equality and Equal Treatment Commissioner in Equality and Equal Estonia. In addition to his other tasks, he acts as a spokesperson for equality-related Treatment Commissioner in matters, identifies gaps in equality policies in Estonia, and cooperates with Estonia institutional partners. In his former position as EIGE's liaison officer, he also managed strategic working relations with various DGs of the European Commission. In the EquiTech project, he provides expertise input for all project deliverables to minimise the risks of discrimination and bias in public sector ADM systems. Merli Kaus, legal adviser in SVV Merli Kaus is a legal adviser in the office of The Gender Equality and Equal the office of The Gender Treatment Commissioner. An expert in labour law, family law, and equal treatment. Equality and Equal Her expertise lies in handling discrimination cases and conducting investigations Treatment Commissioner into complaints. Provides input on discrimination matters for all the deliverables produced by EquiTech. Teele Holmberg, adviser in SVV Teele Holmberg is an advisor in the office of The Gender Equality and Equal the office of The Gender Treatment Commissioner, with strong competencies in project management and Equality and Equal budgeting. She also has over 15 years of experience in communication and event Treatment Commissioner management. In the EquiTech project, she is responsible for project management and dissemination. Miglė Kolinytė, awareness OEOO Miglė Kolinytė is a communication expert. 8 years of experience in awareness raising expert raising, communication and public relations activities. Role/tasks: development of guidelines and impact assessment checklist, Page 18 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 organisation of a study visit, expert work in developing training materials and organisation of trainings, coordination and implementation of communication activities in WP6, participation at international conference. Rūta Juodelytė, equality OEOO Rūta Juodelytė is an equality mainstreaming expert. 7 years of experience in equal mainstreaming expert opportunities mainstreaming, strong project management competencies. Role/tasks: coordination of WP1 activities in Lithuania, development of guidelines, impact assessment checklist and online course material including equality and diversity aspects, expert work in developing training materials and organisation of trainings, lecturer, participation and presentation at international conference. Karolis Čepas, lawyer senior OEOO Karolis Čepas expertise in non-discrimination law, specialization in disability and legal adviser social status aspects: investigating complaints, preforming independent researches, related to discrimination and equal opportunities. Role/tasks: legal analysis of Lithuanian judicial acts, development of impact assessment checklist and online course, participation at international conference. Audronė Kazakevičienė, OEOO Audronė Kazakevičienė is a project financer for Lithuanian team. More than 30 years financer of experience in accounting and finance management. Tanel Kerikmäe, PhD, TalTech Tanel Kerikmäe is a professor of European Legal policy and Law & Technology, an Professor, Department of expert related to legislative drafting, legal analysis and institutional reforms for public Law, School of Business and and private institutions, international organizations (UNESCO, WHO, EU Governance, TalTech institutions, ICF, Deloitte, Valdani Vicari & Associati, Sparks Legal Networks, Bolt. He has been an EU key expert in Central Asia as a head of the component of constitutional law and has been involved in several law&tech related initiatives and development projects. In the project he gives an expertise on AI law, EU and international law and digital governance, manages the organisation of a workshop and writes the academic article. Kristi Joamets, PhD, Senior TalTech Kristi Joamets is a researcher and senior lecturer on technology law, labour law, Lecturer, Study Programme family law and equal treatment. She is also a Study Programme director for the law Director Department of Law, programmes with specialisation on EU and International Law, and Technology law School of Business and and Business Law. In the project she gives expertise on equality law and Governance, TalTech technology law, in creating the teaching materials and courses. Artem Boyarchuk, PhD TalTech Artem Boyarchuk is an IT researcher. He has a solid mathematical and engineering Postdoctoral Researcher, background and extensive experience in developing dependability evaluation and Department of Law, School assessment techniques, particularly, in probability and stochastic processes of Business and theories, experimental data analysis (Matlab). Governance, TalTech Ebru Metin, Doctoral TalTech Ebru Metin is a junior researcher. Her research areas are digitalisation, digital Student, Department of Law, transformation, automation, accountability, dispute resolution, ethics. In the project School of Business and she gives expertise on technology law, in creating the teaching materials and Governance, TalTech courses. Ott Velsberg, Chief Data MKM Ott Velsberg is an Estonian IT specialist and government official known for expertise Officer MoEAC of Estonia in data governance and data science. As the current Chief Data Officer (CDO) of Estonia, he is responsible for driving the country's data policy and initiatives related to the use of data in the public sector. Ott has a wealth of experience in coordinating data governance and data science, focusing on domains such as artificial intelligence, open data, citizen-centric data governance, data management, and digital transformation. Under his leadership, Estonia has gained a reputation as a trendsetter in data governance, open data, and artificial intelligence. With his passion for technology and data, he supports Estonia's digital transformation and shapes the country's future in data governance. Sofia Paes, Data Policy MKM Sofia Paes has experience in AI and open data from managing and executing Advisor MoEAC of Estonia numerous public sector projects aimed at optimising governance. Sofia is passionate about educational and development programs, so she devotes part of her work to improving the data literacy of public employees through workshops, e- courses, and coordinating events. She has extensive knowledge of the Estonian public sector's organisational structure as a result of her work both in local and Page 19 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 national levels. Sofia uses this knowledge to speed up the digital transformation of various public sector organisations through promotion of AI solutions and data governance practices. Henrik Trasberg, Legal JUM Henrik Trasberg is an expert on AI, IT/IP law and digital policy. He is involved in the Advisor on New development of the regulatory policy concerning AI and human-centred digital state. Technologies and Among other, he participates in the negotiations of the EU AIA and other legal acts Digitalisation, Legislative regulating the digital space and represents the Ministry of Justice in the taskforce Policy Department, Ministry for creating and implementing an AI action plan of the Estonian Government to of Justice of Estonia facilitate the uptake of trustworthy and human-centred AI. Siiri Aulik, Legal Adviser, JUM Siiri Aulik is an expert in fundamental rights at EU level and human rights at Legislative Policy European and UN level, including in the right to equality and non-discrimination. She Department, Ministry of has been working in the area of ethical and trustworthy AI since 2018. She has Justice of Estonia served as the Head of Delegation for Estonia for the negotiations of the UNESCO Recommendation on the Ethics of AI and in the Council of Europe’s ad hoc Committee on AI (CAHAI) and its successor the Committee on AI (CAI) for the elaboration and negotiations of a Council of Europe convention to ensure that the design, development and use of AI systems meets the Council of Europe’s human rights, democracy and rule of law standards. She is also involved in the negotiations of the AIA through the Telecom working party, and is responsible for the fundamental rights aspects of the file. Outside resources (subcontracting, seconded staff, etc) If you do not have all skills/resources in-house, describe how you intend to get them (contributions of members, partner organisations, subcontracting, etc). If there is subcontracting, please also complete the table in section 4. Moreover, ensure that subcontractors are aware of gender mainstreaming and non-discrimination mainstreaming. Impact assessment checklist: Partially subcontracted up to the sum of 12 000 €. What is outsourced and why: The checklist is outsourced for the same reasons as the guidelines. As they focus on proactive identification and mitigation of potential ADM system deployment risks, they will require the integration of specific technical and organizational considerations. Outsourcing this task is logical as it calls for deep understanding and expertise in IT and data science, with a specialization in algorithmic bias - an expertise that is more available within specialized entities. Outsourcing these technical components will therefore enhance the quality of the developed checklist. What is not outsourced: The legal aspects and procedural questions and explanations within the checklist. Outsourcing procedure and the possible service provider: to guarantee quality and best-value-for-money in subcontracting, there will be a public procurement carried out to identify the most suitable candidate. The procurement will consider the cost of services but also the quality, capacity, and reliability of the prospective subcontractors. Additional outsourcing will be allocated for translating the document (up to 800 €). Guidelines: Partially subcontracted up to the sum of 40 000 €. What is outsourced and why: As the document have a solution-oriented focus, they should describe specific technical mechanisms and organizational approaches concerning initial data collection, algorithm design, model testing, output analysis, post-deployment risk management, etc. that can be integrated into the development and deployment of the system. This part will be outsourced as it requires unique knowledge from IT experts and data scientists that are specialized on algorithmic bias to identify and communicate these solutions in an effective manner. Such expertise is limited in the project team but available at much higher level within specialized organizations. Thus, outsourcing the technical aspects in these documents will ensure a significantly higher quality for the materials developed. What is not outsourced: In the guidelines, the legal aspects of these materials will be created by the project team, which includes description of the problem and the regulatory framework as well as some specific procedural recommendations, such as concerning engaging with marginalized groups during the system design, etc. Outsourcing procedure and the possible service provider: To guarantee quality and best-value-for-money in subcontracting, there will be a public procurement carried out to identify the most suitable candidate among specialized academic teams or organizations focused on algorithmic risk management. The procurement will consider the cost of services but also the quality, capacity, and reliability of the prospective subcontractors. Additional outsourcing will be done for translating the documents (up to 2000 €). Training Materials: Partially subcontracted up to the sum of 15 000 €. What is outsourced and why: The training materials are outsourced for the same reasons as the guidelines. The creation of comprehensive training materials addressing bias and discrimination complexities in ADM systems requires specialised technical expertise. These materials need to effectively communicate intricate concepts and potential risks to a varied audience, including public sector officials and system developers. The task will be outsourced as it necessitates a blend of educational theory, technical understanding, and design skills, which can be found at a higher level within specialised organisations. Page 20 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 Furthermore, subcontracting will cover the translation of these materials. What is not outsourced: The project team will be responsible for crafting the legal and procedural content within the training materials. This includes among else identifying the problem, outlining and explaining the legal framework. Outsourcing procedure and the possible service provider: to guarantee quality and best-value-for-money in subcontracting, there will be a public procurement carried out to identify the most suitable candidate. The procurement will consider the cost of services but also the quality, capacity, and reliability of the prospective subcontractors. Additional outsourcing will be allocated for translating the document (up to 1500 €). An online course: Partially subcontracted up to the sum of 15 000 €. Its content will be developed by the project team. However, the course will include short video clips and interactive tasks integrated into the course. The course should also have a dynamic user-friendly content design. To ensure the professional nature of the video clips, the interactive tasks, and the design, it is reasonable to involve a service provider who has extensive experience in creating online courses and the capacity to shoot high-quality video clips. The subcontractor will be selected through a competitive bidding process, evaluating both cost and quality. 2.4 Consortium management and decision-making Consortium management and decision-making (if applicable) (n/a for Town Twinning and Programme Contact Points) Explain the management structures and decision-making mechanisms within the consortium. Describe how decisions will be taken and how regular and effective communication will be ensured. Describe methods to ensure planning and control. Note: The concept (including organisational structure and decision-making mechanisms) must be adapted to the complexity and scale of the project. Project management is structured as a separate activity in the WP1, led by the coordinator of the EquiTech project – Estonian Equality Body. Project Management & Quality Board will be set up at the beginning of the project, consisting of one representative of each project partner. All major decisions will be made within this board, ensuring all partners’ needs and observations are considered in the best possible way. The participants will work closely together and throughout the project and meet according to the need online or in person. Each partner is responsible for specific tasks (as exposed in the WPs). The organisations of the activities are the responsibility of the Project Management & Quality Board, taking decisions in a collegiate way. The implementation is the responsibility of each WP, including staff from all the partners under the guidance of the selected leader. There will be online or in person meetings on each WP and on each activity to clarify task distribution. A dedicated ICT environment for communication between partners will be set up immediately on project inception. This environment will be used for uploading project results, easy communication between partners and for sharing relevant information. At the beginning of the project work, when the agreements are made, with all partners, practical matters on the decision-making and daily management will be agreed. Partners will be responsible for designating a lead person from each partner as main contact and representative in the Project Management & Quality Board, but the environment will be open so that all persons involved will have access to the information. In an unlikely event when some information should not be public, it will be possible to restrict access to a selected individual or group. Decisions to be made by the consortium during the project can be initiated by any partner, will then be discussed within the Project Management & Quality Board, making sure all project partners are informed, being given adequate possibility to raise questions, concerns, experience, purpose solution, as well as agree with the matter. #§CON-SOR-CS§# #@PRJ-MGT-PM@# 2.5 Project management, quality assurance and monitoring and evaluation strategy Project management, quality assurance and monitoring and evaluation strategy Describe the measures planned to ensure that the project implementation is of high quality and completed in time. Describe the methods to ensure good quality, monitoring, planning and control. Describe the evaluation methods and indicators (quantitative and qualitative) to monitor and verify the outreach and coverage of the activities and results (including unit of measurement, baseline and target values). The indicators proposed to measure progress should be relevant, realistic and measurable. Note: The monitoring and evaluation strategy should also incorporate gender and non-discrimination considerations in order to measure changes and assess impact on gender equality issues. The indicators should be gender responsive so that they can measure gender equality changes over time. For instance, a gender responsive indicator can measure the increase in women’s rate of employment or changes in social attitudes towards gender roles in work-life balance. The evaluation should be participatory and inclusive to all stakeholders, ensuring that women’s and men’s voices are prevalent throughout the entire evaluation process. Project Management: The EquiTech project coordinator is experienced in project activities, and the consortium is well-established as partners have worked together. Everyone involved has experience in project management, and the participant is a university with additional project management support resources. The coordinator will lead the project and be responsible for the Management and coordination WP and communication with the EU/project officer. The project will have a Project Management & Quality Board consisting of one representative from each partner. The board will regularly meet to set general guidelines/directions and solve issues. WP leaders are responsible for the given WPs and activities within those WPs. The project team will follow the guidelines set in the Grant Agreement, and further details will be described in the Consortium Agreement. For project coordination, regular partner meetings will take place four times in 24 months: two in Estonia, one in Lithuania, and Page 21 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 one online. In addition to these meetings, smaller group meetings will be held regularly and more frequently based on the needs of the project activities to address urgent questions arising from the activities. Project Monitoring: The Project Management & Quality Board will be responsible for monitoring and evaluating the project activities according to the criteria and guidelines set in the Grant Agreement. In qualitative evaluation, the deliverables will be reviewed by the Project Management Board and commented on for further editing, if necessary, to ensure the quality aligns with the set goals before submission to the European Commission. Furthermore, feedback from workshop, conferences, reviews of reports, etc., will be collected and taken into account. For quantitative evaluation, the board will ensure that all the set numbers are reached in terms of activities, workshops, conferences, reports, papers, etc. The project manager will work closely with the academics and ADM professionals to identify risks at the earliest possible stage, enabling efficient countermeasures to be taken in a timely manner. The quality of the project is expected to be considerable because the consortium has strong expertise in all areas covered by the project. At the initial stage of the project an Ethics Guideline will be elaborated to guide the project team on academic integrity and ethical principles of social research throughout the project lifetime. The Ethics Guideline will be elaborated at the first meeting of the project partners. The document will regulate various aspects related to the data collection, protection and preservation, ensuring confidentiality, plagiarism, preservation and exploitation of gender-sensitive data. Evaluation Strategy: The quality of the project will be assessed based on three main aspects: 1. The quantity and quality of the program activities delivered. 2. The quantity and quality of the outcomes achieved by the program. 3. The monitoring and evaluation strategy should incorporate gender and non-discrimination considerations to measure changes and assess the impact on gender equality issues. Quality Assurance: The quality control and monitoring associated with the project are divided into: 1) General aspects: The project has a solid basis, and communication between different experts and topics, with the proper support, will enable progress and resolution of difficulties. 2) Recruitment issues: This project will provide opportunities for several experts in the field of non-discrimination and ADM systems. However, recruitment can be time-consuming and may result in a low recruitment of subjects. 3) Data management: From the beginning, the data will be optimized by targeting a low number of critical obligatory variables but ensuring high quality. #§PRJ-MGT-PM§# #@FIN-MGT-FM@# 2.6 Cost effectiveness and financial management Cost effectiveness and financial management (n/a for prefixed Lump Sum Grants) Describe the measures adopted to ensure that the proposed results and objectives will be achieved in the most cost-effective way. Indicate the arrangements adopted for the financial management of the project and, in particular, how the financial resources will be allocated and managed within the consortium. Do NOT compare and justify the costs of each work package, but summarize briefly why your budget is cost effective. The coordinator is responsible for managing the finances of the EquiTech project, adhering to the relevant public sector laws in Estonia and Lithuania. In accordance with the financial management regulations, the procurement process will be conducted in compliance with the applicable laws and procedures. For the procurement of services, the coordinator will seek three comparative offers to ensure transparency and fairness. In Estonia, the Public Procurement Act establishes a threshold of 30,000 euros for simple procurement contracts involving goods or services. Therefore, when selecting a project caterer or any other service provider, a price request will be sent to several potential providers to ensure a competitive process. In cases where the estimated cost of services exceeds 30,000 euros, an international tender will be organised to solicit offers from a broader range of providers. This ensures that the project's procurement process aligns with legal requirements and follows a transparent and fair approach. Throughout the project, the finances will be managed in strict adherence to the applicable laws and regulations, emphasising transparency and accountability. By following these procedures, the EquiTech project demonstrates its commitment to lawful and transparent financial management practices. #§FIN-MGT-FM§# #@RSK-MGT-RM@# 3. IMPACT 3.1 Impact and ambition Impact and ambition Define the short, medium and long-term effects of the project. Who are the target groups? How will the target groups benefit concretely from the project and what would change for them? In what way will the gap identified be reduced? How will the activities contribute to improve the situation (difference between starting point/state of play and the situation after the end of the project?) How will the activities contribute to the promotion and advancement of gender equality and non- discrimination mainstreaming? Does the project aim to trigger change/innovation? If so, describe them and the degree of ambition (progress beyond the status quo). Note: Results/outcomes are immediate changes that materialise for the target groups after the end of the project (e.g. improved knowledge, Page 22 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 increased awareness). Results/outcomes are different to deliverables. Deliverables are activities undertaken and outputs produced with the resources allocated to the project, e.g. training courses, conferences, manuals, video etc. When defining expected results/outcomes and deliverables please consider if and how they will reduce, maintain, or increase inequalities between women and men, boys and girls, in all their diversity. What gender, age and disability differentiated results can be expected? How expected results will affect women and men, boys and girls from a range of diverse social groups, differently? The final beneficiaries are intended to be all vulnerable groups who face a risk of whichever form of discrimination associated with public sector ADM systems. The ambition of the EquiTech project is an effective adoption of state-of-the-art solutions and procedures in mitigating risk of discrimination and bias in the public sector ADM systems, ensuring a maximum level of fairness of these systems. Short-term impacts To achieve the above, the project delivers a set of support materials and tools for the public sector officials and system developers such as guidelines, impact assessment checklist, online courses and training materials about the risks of discrimination and bias in ADM system. These materials are based on the initial mapping of specific concerns identified in the existing ADM systems and building on top of practices from other countries (see WP2). For example, these deliverables intend to assist public sector officials and system developers in tackling questions such as whether to utilize automated digital solutions in the first place (considering the possible risks), which models and data should be used, how to organize human supervision, which solutions can be deployed to identify possible biases, how to ensure effective involvement of stakeholders prior to system design, etc. The goal of these deliverables is to ensure that the actors that determine the qualitative and procedural choices in developing and deploying the ADM systems are thoroughly informed on the possible risks concerning possible bias and discrimination in the ADM systems and are provided a set of tools and resources that help in mitigating the biases. More specifically, these deliverables should achieve the following: - Higher quality and more standardized practices across the public sector in the process of identifying, assessing, and mitigating potential biases and discriminatory practices in ADM systems. - Increased awareness as well as the skills and capabilities among the key stakeholders in understanding and managing the risks of discrimination and bias in ADM systems. - More inclusive and fair approach in the design, development, and implementation of ADM systems that consider the complexities, functionalities, and potential pitfalls of using ADM systems in public sector. - Reduced risk of regulatory non-compliance of public sector ADM systems. - Increased public trust in the public sector's use of ADM systems. Simultaneously, the project will steer cooperation between Equality Bodies, state institutions (e.g. Ministries) and academia through workshop, study visits and international conference. Increasing awareness and competence as well as creating collaboration among these actors will have an important role in ensuring that the governance over the deployment of ADM systems in public sector would be competent. Medium and long-term impacts The above measures are expected to also have a medium and long-term impact. The deliverables themselves remain available and can be easily updated as the state-of-the-art progresses. One of the deliverables to support the long-term impact of the support tools and resources will be the policy recommendations drafted based on the learnings throughout the project. It is likely that these recommendations will result in some of the project deliverables becoming a compulsory part of ADM system development in the public sector (e.g. obligation to carry out bias and discrimination impact assessment or regulator audits of systems developed as a result of a public procurement). Furthermore, the project will have created a holistic basis to continue with various efforts in terms of ensuring further cooperation between key stakeholders, carrying out continuous training (based on the created training material) and developing new technical or procedural solutions for tackling discrimination in ADM systems. Notably, the project will raise the capacity of Equality Bodies and other public authorities in dealing with (intersectional) discrimination, racism, antisemitism and xenophobia, LGBTIQ-phobia and all other forms of intolerance that could be hidden in the underlying data of the ADM systems. This knowledge will be transferrable to different level of governance, incl municipal and regional. The project also intends to raise general awareness of discrimination and bias risks in ADM systems among academics and general public. One of the project deliverables will be media awareness campaign, which is directed at the general public, in order to also educate wider public about the potential biases and discriminatory practices that can arise from ADM systems. In this process, particular attention will be given to vulnerable groups who face potential risks of discrimination associated with ADM systems. These groups include individuals who may be subjected to discrimination based on factors such as gender, age, nationality (including racism, xenophobia, and antisemitism), skin colour, religion, LGBTIQ status, beliefs, disability, family obligations etc. By raising awareness among vulnerable populations about the risks associated with these systems, the project can empower them to understand their rights, recognize discrimination, and take appropriate actions. Providing support and knowledge to these groups is essential in ensuring that they are equipped to challenge discriminatory practices and advocate for fair treatment. Through this, the project aims to create a more inclusive and equitable society, where the impacts of discrimination and bias in ADM systems are minimised. The measures directed at the academic community (notably they include workshop and publishing high-category scholarly article) help to further understand the risks concerning these systems, developing innovative methods to reduce bias, and providing insights that inform policy and practice. As one of the project partners is also an academic institution, the intention is also to foster collaboration between academia and the public sector, ensuring that future research findings are better integrated into real-world ADM system design and deployment. Furthermore, the various project outcomes will be disseminated through high category academic article, ensuring that the academic community is aware of the latest developments and insights. By Page 23 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 sharing knowledge with academics researching the field, the project can contribute to the ongoing scholarly discourse on addressing discrimination and bias in ADM systems. Impact on the EU level: The various deliverables of the project, including the impact assessment checklist, guidelines, and educational resources, can also assist EU member states in addressing systemic discrimination and bias in their ADM systems. Sharing the experience and solutions developed in Estonia and Lithuania with the EU community is specifically established as one of the goals of the project. To facilitate this, the key deliverables will be translated into English, and a comprehensive dissemination plan will be carried out, including drafting a final impact report, academic literature, and a concluding international conference. Impact on the private sector organizations: Although this project doesn't specifically target private sector organizations, the standards and quality set within public sector IT systems can often serve as an example for the private sector. Hence, there is a potential ripple effect: − Developers and product managers, after implementing bias mitigation procedures in public IT systems, may carry these standards into the private sector. − Public sector efforts can heighten societal expectations in terms of how discrimination in the ADM systems should be mitigated. − Upon this project's completion, it's plausible that subsequent activities may focus on raising awareness and competence in private sector companies as part of a follow-up project. #§IMP-ACT-IA§# #@COM-DIS-VIS-CDV@# 3.2 Communication, dissemination and visibility Communication, dissemination and visibility of funding Describe the communication and dissemination activities which are planned in order to promote the activities/results and maximise the impact (to whom, which format, how many, etc.). Clarify how you will reach the target groups, relevant stakeholders, policymakers and the general public and explain the choice of the dissemination channels. Describe how the visibility of EU funding will be ensured. Communication and dissemination activities should also contribute to the promotion of gender equality and non-discrimination. Communication materials should use gender inclusive language and positive visual representations. E.g.: when developing videos or leaflets ensure that women and men, in all their diversity, are equally represented in a non-stereotypical fashion and portrayed in active empowered roles. In addition, consider using communication channels that they are accessible to general audience, in particular to persons with disabilities, or people from marginalised groups. The EquiTech project will primarily focus on raising awareness among professionals (Equality Bodies, public service employees, data scientists, and data stewards) and the general public regarding potential discrimination in public sector ADM procedures, including imparting general knowledge about discrimination itself. For communication and visuality, the EquiTech project is using different channels, choosing what is best for the target group. Professional awareness raising. At the national level, this project will complement state policies and regulations concerning AI-based discrimination. It will aid in preparing for the implementation of AIA regulations, assisting in the resolution of discrimination cases related to algorithm-based decisions, and fostering collaboration between the public sector and developers. Additionally, it aims to enhance the Estonian National AI Strategy by assessing the current situation and providing policy recommendations to promote the human-centric implementation of AI within the country. This includes defining relevant strategic objectives, Key Performance Indicators (KPIs), and the necessary resources to advance the field. At the European Union level, the project will provide practical insights into the application of regulations regarding discrimination in AI systems, complementing the planned activities and research conducted by the EU in this field. All the deliverables produced in the EquiTech project are given a free license to be adopted by all European countries. The research report generated by the project will be disseminated through the Equinet annual meeting for Equality bodies across the EU. This dissemination aims to introduce the project's findings and spark discussions on new discrimination-related topics that EquiTech is addressing. Public awareness-raising activities. For the general public, we will focus on addressing the potential for discrimination when utilizing digital services. Currently, there is a lack of systematic and research-backed knowledge among professionals in Estonia and Lithuania regarding the discrimination and bias risks within the public sector ADM systems. Furthermore, the general public is generally unaware of the potential discrimination risks associated with these systems. To address this issue, the EquiTech project aims to raise awareness among the general public about the possibilities of discrimination when using digital services. It is important to recognize that the general public, especially vulnerable groups, can become victims of discrimination facilitated by ADM systems. By increasing awareness, these vulnerable populations can navigate the ADM landscape more effectively, identify potential biases, and actively advocate for fair and equitable treatment. Through this process, stakeholders can acknowledge and address the risks of discrimination within ADM systems, working towards fostering a more inclusive and just society throughout the European Union. The EquiTech project seeks to empower both professionals and the general public with the knowledge and understanding needed to recognize and mitigate discrimination risks in the use of ADM systems. By promoting awareness and advocating for fairness, the project aims to create a more equitable and inclusive digital environment for all. The primary focus of our awareness campaign will be directed towards the vulnerable target groups through social media channels such as Facebook and Instagram. In addition to these efforts, we will also employ online and Google campaigns to further support our cause. Running in parallel with these activities, we plan to conduct interviews with experts from the EquiTech project on television and radio programmes. Awareness raising campaign in Lithuania will focus on general public with a goal to inform and raise awareness about Page 24 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 the possible discrimination risks in AI. The campaign will include press releases, articles, interviews with experts, local TV and radio show appearances, social media content and ads. Academic and professional dissemination (High-category scholarly article and workshop). Within the academic sphere, this project aims to raise awareness about discrimination in the digital world through high-quality scholarly article. This article will serve as a foundation for further discussion in various academic contexts and facilitate their incorporation into teaching and curricula. Additionally, academic workshop will provide a platform for scientific discussions on the topic of discrimination, allowing researchers to disseminate their findings and stimulate new research and scientific debates. There is currently a significant knowledge gap in every sector, but also in the academic world regarding both theoretical and practical understanding of discrimination and bias risks in ADM systems. To address this issue, the EquiTech project aims to enhance knowledge and promote awareness among academics and professionals through the dissemination of high-quality scholarly article and workshop. These activities will serve to bridge the knowledge gap and foster a deeper understanding of the complexities surrounding bias and discrimination in ADM systems. By engaging with academia and professionals, the EquiTech project seeks to generate valuable insights, promote research collaboration, and encourage the exchange of expertise. This concerted effort will contribute to the development of robust theoretical frameworks and practical strategies for addressing bias and discrimination risks in public sector ADM systems. Overall, the project strives to promote the advancement of knowledge and scholarship in this field, ultimately leading to more informed decision-making, improved practices, and a more equitable and inclusive use of ADM systems in the public sector services. Organization of an international conference on discrimination in ADM systems. In the Baltics, there has never been a dedicated conference focused on discrimination in ADM systems, which has limited the opportunity for professionals in the equality and ADM fields to engage in mutual learning. To address this gap, the EquiTech project will organise an international conference on discrimination in ADM systems as one of its deliverables. The conference aims to bring together experts in ADM systems within the public sector, ADM developers, and Equality Bodies from across European countries to share their expertise and insights. The primary objective is to facilitate mutual learning and explore the application of combined knowledge in addressing discrimination and AI systems. The conference will include keynote speeches in a large hall, as well as round table sessions in small groups. These sessions will provide an opportunity for experts from both sides to engage in focused conversations and exchange experiences. The conference is expected to attract 75 participants, including representatives from various sectors and from all European countries. This would spare the knowledge wider than only the EquiTech project countries. Equinet, as one of the supporting organisations of the EquiTech project, is interested in sharing knowledge. All Equinet member organisations are invited to the conference. Partnership with Equinet - the European Network of Equality Bodies The EquiTech project has garnered attention and recognition, aligning with the goals of Equinet. Equinet provides its support to the EquiTech project through a formal letter of support (see the Annex). This initiative acknowledges the importance of discussing new topics related to discrimination, either through direct communication with individual countries or within the Equinet network. In these rapidly evolving times, it is crucial for Equality Bodies to collaborate and exchange knowledge in order to minimise discrimination and bias in emerging technologies like AI and the expanding influence of the data economy. In fact, FRA has concluded in its recent study on bias in algorithms that “Equality Bodies should step up their efforts to address discrimination complaints and cases linked to the use of algorithms. In order to do this effectively, they should employ specialised staff and cooperate with data protection authorities and other relevant oversight bodies.”35. The EquiTech project strives to foster cross-border and cross-institutional cooperation by bringing together Equality Bodies from two countries, as well as relevant Ministries and academia. Moreover, the intention is to share the insights gained from this collaboration with other Equality Bodies throughout the European Union. By doing so, the project aims to promote a comprehensive approach to addressing discrimination and bias in the digital era. Partnership with the Federal Anti-Discrimination Agency (FADA) in Germany The EquiTech project has garnered attention from an additional party, the Federal Anti-Discrimination Agency (FADA) in Germany. FADA, highly interested in the subject of discrimination risks associated with ADM systems, is keen on finding solutions to overcome these challenges. Consequently, they are delighted to extend their support to the EquiTech project by providing a much-needed letter of endorsement. In Germany, the issue of discrimination in ADM systems has been gaining prominence, making it a pressing concern. FADA recognizes the urgency of addressing this issue and acknowledges the potential of EquiTech to contribute significantly to the field. FADA is particularly interested in the tangible outcomes and deliverables generated by the project. By sharing experiences and best practices from one country, FADA believes it can enhance its understanding of research on discrimination, which has thus far been predominantly limited to a theoretical level. Moreover, sharing the experiences helps to combat the risk of discrimination and bias in ADM systems. The support and collaboration from FADA will undoubtedly bolster EquiTech's efforts to explore and tackle the challenges posed by discrimination in ADM systems. With the shared goal of promoting fairness and equality, both EquiTech and FADA aim to drive positive change and ensure that ADM systems can be free from biases and discriminatory practices. Partnership with the Non-Discrimination Ombudsman of Finland The EquiTech project will establish a valuable partnership with the Non-Discrimination Ombudsman of Finland, which serves as a significant source of support and collaboration. As Finland also finds the topic acute, requiring extensive resources for research and learning, we strive to achieve our project objectives. We recognize together the importance of knowledge dissemination and fostering cooperation across borders. In line with this commitment, the deliverables produced by the EquiTech project will be shared with the esteemed Finnish Equality Body. By doing so, we aim to facilitate the sharing of valuable insights and expertise with public institutions responsible for the 35 European Union Agency for Fundamental Rights study “Bias in Algorithms – Artificial Intelligence and Discrimination” (2022), p 13. Page 25 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 administration of the public sector ADM system in Finland. Our intention is to provide the mutual learning process with the necessary knowledge and resources to enhance their understanding and implementation of equitable practices. The Non-Discrimination Ombudsman of Finland plays a pivotal role in our partnership, actively monitoring and engaging with our project activities. Their invaluable involvement ensures that the knowledge gained through the EquiTech project is disseminated in parallel with our ongoing efforts. This parallel sharing of knowledge serves to create a powerful synergy between our project's outcomes and the initiatives spearheaded by the Non-Discrimination Ombudsman of Finland. Together, we are committed to promoting inclusivity, non-discrimination, and equality, both within Finland and beyond. Through this partnership, we endeavor to drive positive change, influence policy decisions, and foster a more equitable society. #§COM-DIS-VIS-CDV§# #@SUS-CON-SC@# 3.3 Sustainability and continuation Sustainability, long-term impact and continuation Describe the follow-up of the project after the EU funding ends. How will the project impact be ensured and sustained? What will need to be done? Which parts of the project should be continued or maintained? How will this be achieved? Which resources will be necessary to continue the project? How will the results be used? Are there any possible synergies/complementarities with other (EU funded) activities that can build on the project results? The EquiTech project is planned to be a continuous process, constantly moving forward. A desk review on discrimination and bias in ADM systems, legal analyses, and audits of discrimination and bias within Estonian public sector ADM systems, along with current assessment practices, will provide sufficient material for discussing the topic of discrimination from various perspectives, including policy-making and the academic field. The results of the EquiTech project will improve public sector ADM systems in Lithuania and also have an influence on Finland and Germany as supportive countries. In this regard, the project will serve as a starting point for using the knowledge gained in future developments regarding the risks of discrimination. The materials created for training, as well as the training sessions themselves, will be supplemented and adapted to future needs and specific requirements. They can also serve as a foundation for developing other teaching materials and can be translated into multiple languages. The impact assessment checklist and guidelines will serve as base documents that can be improved in accordance with future changes in society. The online course can be translated into several languages and delivered in other countries, extending its use to a wider learning group in the future. Study visits will facilitate new cooperation opportunities and potential collaborations in other new projects. Conference, workshop, and other dissemination activities will shed light on the topic of discrimination from a fresh perspective, generating new discussions, collaborations, and advancements in this field, ultimately promoting a general awareness of equal treatment. The knowledge and skills acquired from the project will also provide a foundation for policy and law development in the participating states and will be presented to Equinet members at the Equinet Annual Meeting, contributing to a broader context. The online course is designed to be adaptable and translatable, making it suitable for other countries across Europe. This allows for wider dissemination and adoption of the training material to enhance awareness and address discrimination and bias in ADM systems on a broader scale. By providing this training opportunity, the EquiTech project seeks to empower public sector officials with the knowledge and tools needed to identify and combat bias and discrimination effectively. Policy recommendations on ADM systems for fight against discrimination in Estonia and Lithuania: The project encompasses various activities – such as evaluating current practices for identifying and mitigating biases in public sector ADM systems, and the development and testing of guidelines – that can provide valuable input to shaping governance and policy decisions concerning ADM systems in the public sector. Therefore, in the final stage of the project, one of the tasks is to draft a paper featuring policy recommendations that reflect the insights from the preceding stages of the project. These recommendations could potentially include a wide range of actionable items, such as mandating the execution of an impact assessment checklist within the procurement process, establishing systematic discrimination and bias training for certain public office roles, advocating for periodic audits and reviews of the ADM systems, fostering multi-stakeholder dialogue. Another aspect of the project involves developing comprehensive procurement conditions that facilitate the trustworthy development and deployment of AI solutions. With the increasing pervasiveness of AI technologies, ensuring trustworthiness in AI systems becomes crucial in mitigating risks and safeguarding the interests of organisations and end-users. Thus, the project intends to address this challenge by creating a set of procurement conditions that can guide organisations in acquiring AI solutions that are reliable, transparent, ethical, and compliant with relevant regulations and standards. An international conference on discrimination in public ADM systems will be organised in Tallinn. The conference aims to bring together Equality Bodies, experts on ADM systems in the public sector, ADM developers, and scientists from across European countries to share their expertise and insights. The primary goal is to facilitate mutual learning and explore ways to apply the combined knowledge of discrimination and AI systems. The conference will feature keynote speeches in a large auditorium, as well as round table sessions in small groups. These sessions will provide an opportunity for experts from both sides to engage in tailored conversations and exchange experiences. By gathering experts from different backgrounds, the conference seeks to create an environment conducive to collaborative learning, knowledge sharing, and the identification of common strategies to address discrimination in ADM systems. Workshop organised by TalTech will introduce the topic to wider academia to raise the interest of other researchers to deal with the discrimination topic regarding AI tools. #§SUS-CON-SC§# Page 26 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 #@WRK-PLA-WP@# 4. WORKPLAN, WORK PACKAGES, ACTIVITIES, RESOURCES AND TIMING 4.1 Work plan Work plan Provide a brief description of the overall structure of the work plan (list of work packages or graphical presentation (Pert chart or similar)). Page 27 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 WP 1: Management and coordination (SVV) 1.1. Management and coordination. Project coordination and management to follow the timeline of activities, coordinate the communication between the partners. Including the 4x partner meetings: 2x in Estonia, 1xLithuania, 1xOnline. Partner meetings serve to coordinate tasks and deliverables, share information, and provide input for ongoing actions and deliverables. 1.2. Reporting. The mid-term and final reports will also be coordinated during the partner meetings. WP 2: Discrimination mapping and analysis (TalTech) 2.1. Conducting a literature review on discrimination and bias in ADM systems. Literature review will enable all project partners communicate on the concepts effectively. Systematic understanding of the extent of risks in context of Estonia and Lithuania is created to understand the best solutions to tackle them better. 2.2. Legal analyses in Estonia and Lithuania. A legal analysis will be conducted in both countries – Estonia and Lithuania, focusing on the existing legal provisions and regulations. This analysis will help in identifying the legal framework surrounding discrimination and bias in ADM systems and provide valuable insights for future actions and improvements. A legal analysis will be conducted in both countries – Estonia and Lithuania, focusing on the existing legal provisions and regulations. This analysis will help in identifying the legal framework surrounding discrimination and bias in ADM systems and provide valuable insights for future actions and improvements. 2.3. Auditing discrimination and bias within Estonian public sector ADM systems and current assessment practice. An audit on providing an overview of potential discriminatory and bias practices within the ADM systems used in the public sector, along with legal and organizational recommendations for improvement. Conducting a report – mapping the risks of bias and discrimination on ADM systems used in public sector, mapping the assessment practice and method(s) for a risk analysis WP 3: Impact Assessment toolbox development (JUM) 3.1. Impact assessment checklist development. Chart 1: PERT chart of the EquiTech project Developing an impact assessment checklist for identifying and assessing potential risks of bias and discrimination within the ADM system. The checklist is structured to ask a series of questions or list considerations concerning these areas. It builds on and refers back to the guidelines in answering specific questions on how possible risks are being identified and mitigated. Impact assessment checklist will be available electronically in 3 languages (ET, LT, EN), content of 15 pages in each language. 3.2. Guidelines development for system developers, scientists, data stewards, and public service providers to detect bias and discrimination in the design, implementation, and development of ADM systems. Guidelines development for system developers, data stewards and public service providers to detect bias and discrimination in the design, implementation, and development of ADM Page 28 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 systems. Guidelines is the document, which gives a step-by-step overview on how to use the impact assessment checklist, explaining assessment process and ways to avoid discrimination while developing and using AI solutions. The guidelines would be based on the latest research as well as findings from the situation mapping exercise done under WP2. The guidelines will be available electronically in 3 languages (ET, LT, EN), content of 60 pages in each language. 3.3. Creating an online course for Estonian and Lithuanian public sector representatives. The activity includes developing and making available an online course for the public sector officials. These courses would be designed to increase awareness and understanding of discrimination and bias in ADM systems. They would cover topics such as the potential sources of bias in AI, the impact of bias on decision-making, and basic suggestions for mitigating bias. It is a 60-minutes online course consisting of 30 pages of text content per language (ET, EN, LT) and up to 5 videos (3-5 min). The online course for the employees of the public sector will be available on two platforms: 1) For the employees of the public sector in Estonia in Estonian and in English on Estonian Digital Academy (digiriigiakadeemia.ee). 2) For the employees of the public sector in Lithuania in Lithuanian. 3.4. Developing training materials for data scientists, data stewards, and public service providers. This activity foresees developing a set of educational materials that assist in training public sector officials, system developers and other stakeholders about the complexities and potential risks concerning bias and discrimination in ADM systems. Primarily, they are meant to support the training carried out under WP5 but can also be used separately. Content of of 25 pages for the data scientists, data stewards and 25 pages content for the employees of public service, both in 3 languages (ET, EN, LT). The Materials will be available electronically, but for the trainees is also a 32 pages printed version. WP 4: Testing the impact assessment toolbox (MKM) 4.1. Testing the impact assessment checklist for potential biases and discriminatory practices within the ADM systems used in the public sector The activity includes testing out impact assessment checklist on 3 of the real solutions used in the Estonian public sector. The testing is giving the scale for real testing for all the systems in the future. 4.2. Evaluation of the discrimination risk profile of the ADM systems used in the public sector Evaluating the results of the assessment on the real AI use-cases from Estonia, taking in consideration the things, which can be improved to achieve better assessment experience and accuracy of results. WP 5: Capacity building and training activities for national, regional and local authorities (TalTech) 5.1. Training for Estonian and Lithuanian employees of public sector organisations. The face to face trainings for public authorities will be carried out to deepen the digital skills and critical thinking to identify the possible discrimination risks on the public sector ADM systems. Public sector organizations often handle vast amounts of sensitive data, making effective data stewardship essential for the successful implementation of AI-driven initiatives. In total 150 employees in public sector will be trained: 2x in Estonia in ET, 2x50 participants; 3x in Lithuania in LT, together 50 participants in 3 trainings. 5.2. Study visits to European leaders in fighting discrimination through ADM implementation. Study visits will be organized to European countries who have made significant progress in combating discrimination through the implementation of ADM systems in the public sector. Study visits offer an invaluable opportunity to observe and learn from the approaches taken by other European countries in addressing similar challenges and understanding discrimination cases. All consortium partners have a representatives at the study visit. 5.3. Training in avoiding discrimination risks for data scientists and data stewards. The face to face training teaches the criticality of effective data management in the successful deployment of AI solutions. Participants will gain a deep understanding of the role of data stewards and their responsibilities in ensuring data quality, integrity, and security throughout the AI development lifecycle with focus on risks of discrimination and bias in ADM systems. Face to face training for the data scientists and data stewards. The training program tailored for the public sector will concentrate on data management for AI deployment. In total 80 data scientists and data stewards will be trained: 2x Estonia in ET, 2x20 participants and 2x Lithuania in EN/LT, 2x20 participants. 5.4. Implementation of the online course for public sector organisations in Estonia and Lithuania. The course is available on 2 platforms: Digital State Academy (digiriigiakadeemia.ee) in Estonian and English and in Lithuanian portal in Lithuanian. WP 6: Dissemination and awareness raising (SVV) 6.1. Public awareness raising activities (Devising and implementing a media campaign to raise awareness about possible bias and discrimination in ADM systems). Page 29 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 For the general public, we will focus on addressing the potential for discrimination when utilizing digital services. Currently, there is a lack of systematic and research-backed knowledge among professionals in Estonia and Lithuania regarding the discrimination and bias risks within the public sector ADM systems. The primary focus of our awareness campaign will be directed towards the vulnerable target groups through social media channels such as Facebook and Instagram. In addition to these efforts, we will also employ online and Google campaigns to further support our cause. Running in parallel with these activities, we plan to conduct interviews with experts from the EquiTech project on television and radio programmes. Awareness raising campaign in Lithuania will focus on general public with a goal to inform and raise awareness about the possible discrimination risks in AI. The campaign will include press releases, articles, interviews with experts, local TV and radio show appearances, social media content and ads. Reach 300000 citizens in Estonia and 500000 citizens in Lithuania. 6.2. Organization of an international conference on discrimination in ADM systems. An international conference on discrimination in ADM systems will be organized in Tallinn. The conference aims to bring together experts on ADM systems in the public sector, ADM developers, and equality bodies and scientists from across European countries to share their expertise and insights. The primary goal is to facilitate mutual learning and explore ways to apply the combined knowledge of discrimination and AI systems.The conference is expected to attract a minimum of 75 participants, including representatives from various sectors. By gathering experts from different backgrounds, the conference seeks to create an environment conducive to collaborative learning, knowledge sharing, and the identification of common strategies to address discrimination in ADM systems. 6.3. Academic and professional dissemination (High-category scholarly article and workshop). The high-category scholarly article will raise awareness of discrimination in digital world among academics to discuss it further in different scope, facilitating also inserting it to the teaching and curricula. Academic workshop allows handle the topic in scientifically discussing the topic with other scientists and disseminating by this the topic of discrimination to the new research and scientific discussions. Number of participants 30, notably academics who are interested in developing the knowledge of non-discriminative AI related topics. 6.4. Policy development: policy recommendations and strategies on ADM systems for the fight against discrimination. Policy recommendations on ADM systems for fight against discrimination in Estonia and Lithuania: The project encompasses various activities – such as evaluating current practices for identifying and mitigating biases in public sector ADM systems, and the development and testing of guidelines – that can provide valuable input to shaping governance and policy decisions concerning ADM systems in the public sector. Therefore, in the final stage of the project, one of the tasks is to draft a paper featuring policy recommendations that reflect the insights from the preceding stages of the project. These recommendations could potentially include a wide range of actionable items, such as mandating the execution of an impact assessment checklist within the procurement process, establishing systematic bias and discrimination training for certain public office roles, advocating for periodic audits and reviews of the ADM systems, fostering multi-stakeholder dialogue. 6.5. Evaluation of project activities. Evaluation of the value and impact created by the activities in the final phase of the EquiTech project. Subcontracting (n/a for prefixed Lump Sum Grants) Subcontracting Give details on subcontracted project tasks (if any) and explain the reasons why (as opposed to direct implementation by the Beneficiaries/Affiliated Entities). Subcontracting — Subcontracting means the implementation of ‘action tasks’, i.e. specific tasks which are part of the EU grant and are described in Annex 1 of the Grant Agreement. Note: Subcontracting concerns the outsourcing of a part of the project to a party outside the consortium. It is not simply about purchasing goods or services. We normally expect that the participants have sufficient operational capacity to implement the project activities themselves. Subcontracting should therefore be exceptional. Include only subcontracts that comply with the rules (i.e. best value for money and no conflict of interest; no subcontracting of coordinator tasks). Make sure that subcontractors are aware of the principles of gender mainstreaming and non-discrimination mainstreaming. Work Subcontra Subcontract Description Estimate Justification Best-Value-for-Money Pack ct No Name (including task d Costs (why is subcontracting necessary?) (how do you intend to ensure it?) age (continuou (subcontract number and (EUR) No s ed action BEN/AE to which it Page 30 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 numbering tasks) is linked) linked to WP) 3 S1.1 Impact 3.1. Impact 12000€ The checklist is outsourced for the same reasons as the guidelines. As they To guarantee quality and best- BEN05 assessment Assessment focus on proactive identification and mitigation of potential ADM system value-for-money in (JUM) checklist checklist in 3 deployment risks, they will require the integration of specific technical and subcontracting, there will be a languages organizational considerations. Outsourcing this task is logical as it calls for public procurement carried out to deep understanding and expertise in IT and data science, with a specialization identify the most suitable in algorithmic bias - an expertise that is more available within specialized candidate. The procurement will entities. Outsourcing these technical components will therefore enhance the consider the cost of services but quality of the developed checklist. also the quality, capacity, and What is not outsourced: The legal aspects and procedural questions and reliability of the prospective explanations within the checklist. subcontractors. 3 S1.2 Guidelines T.3.2 Guidelines 40000€ As the document have a solution-oriented focus, they should describe specific To guarantee quality and best- BEN05 development for technical mechanisms and organizational approaches concerning initial data value-for-money in (JUM) system developers, collection, algorithm design, model testing, output analysis, post-deployment subcontracting, there will be a scientists, data risk management, etc. that can be integrated into the development and public procurement carried out to stewards, and deployment of the system. This part will be outsourced as it requires unique identify the most suitable public service knowledge from IT experts and data scientists that are specialized on candidate among specialized providers to detect algorithmic bias to identify and communicate these solutions in an effective academic teams or organizations bias and manner. Such expertise is limited in the project team but available at much focused on algorithmic risk discrimination in the higher level within specialized organizations. Thus, outsourcing the technical management. The procurement design, aspects in these documents will ensure a significantly higher quality for the will consider the cost of services implementation, materials developed. but also the quality, capacity, and and development of What is not outsourced: In the guidelines, the legal aspects of these reliability of the prospective ADM systems. materials will be created by the project team, which includes description of the subcontractors. problem and the regulatory framework as well as some specific procedural recommendations, such as concerning engaging with marginalized groups during the system design, etc. 3 S1.3 Online T3.3 Creating an 15000€ Its content will be developed by the project team. However, the course will The subcontractor will be selected BEN05 course online course for include short video clips and interactive tasks integrated into the course. through a competitive bidding (JUM) Estonian and The course should also have a dynamic user-friendly content design. To process, evaluating both cost and Lithuanian public ensure the professional nature of the video clips, the interactive tasks, and the quality. sector design, it is reasonable to involve service provider who has extensive representatives. experience in creating online courses and the capacity to shoot high-quality video clips. Page 31 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 3 S1.4 Training T3.4 Developing 15000€ What is outsourced and why: The training materials are outsourced for the To guarantee quality and best- BEN05 materials training materials same reasons as the guidelines. The creation of comprehensive training value-for-money in (JUM) for data scientists, materials addressing bias and discrimination complexities in ADM systems subcontracting, there will be a data stewards, and requires specialized technical expertise. These materials need to effectively public procurement carried out to public service communicate intricate concepts and potential risks to a varied audience, identify the most suitable providers. including public sector officials and system developers. The task will be candidate. The procurement will outsourced as it necessitates a blend of educational theory, technical consider the cost of services but understanding, and design skills, which can be found at a higher level within also the quality, capacity, and specialized organizations. Furthermore, subcontracting will cover the reliability of the prospective translation of these materials. subcontractors. What is not outsourced: The project team will be responsible for crafting the legal and procedural content within the training materials. This includes among else identifying the problem, outlining and explaining the legal framework. Other issues: If subcontracting for the project goes beyond 30% of the n/a total eligible costs, give specific reasons. Timetable Timetable (projects up to 2 years) Fill in cells in beige to show the duration of activities. Repeat lines/columns as necessary. Note: Use the project month numbers instead of calendar months. Month 1 marks always the start of the project. In the timeline you should indicate the timing of each activity per WP. ACTIVITY MONTHS M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 M16 M17 M18 M19 M20 M21 M22 M23 M24 WP 1: Management and coordination 1.1. Management and coordination 1.3. Reporting WP 2: Discrimination mapping and analysis 2.1. Conducting a literature review on discrimination and bias in ADM systems. Page 32 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 2.2. Legal analyses in Estonia and Lithuania. 2.3. Auditing discrimination and bias within Estonian public sector automated decision- making (ADM) systems and current assessment practice. WP 3: Impact Assessment toolbox development 3.1. Impact assessment checklist in 3 languages 3.2. Guidelines development for system developers, scientists, data stewards, and public service providers to detect bias and discrimination in the design, implementation, and development of ADM systems. 3.3. Creating an online course for Estonian and Lithuanian public sector representatives. 3.4. Developing training materials for data scientists, data stewards, and public service providers and public organizations. WP 4: Testing the impact assessment toolbox 4.1. Testing the impact assessment checklist for potential biases and discriminatory practices within the ADM systems used in the public sector 4.2. Evaluation of the discrimination risk profile of the ADM systems used in the public sector WP 5: Capacity building and training activities for national, regional and local authorities 5.1. Training for Estonian and Lithuanian employees of public sector organizations. Page 33 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 5.2. Study visits to European leaders in fighting discrimination through ADM implementation. 5.3. Training in avoiding discrimination risks for data scientists and data stewards. 5.4. Implementation of the online course for public sector organizations in Estonia and Lithuania. WP 6: Dissemination and awareness raising 6.1. Public awareness-raising activities (Devising and implementing a media campaign to raise awareness about possible bias and discrimination in ADM systems). 6.2. Organization of an international conference on discrimination in ADM systems. 6.3. Academic and professional dissemination (High-category scholarly articles and workshop) 6.4. Policy development: policy recommendations and strategies on ADM systems for the fight against discrimination. 6.5. Evaluation of the value/impact created by the activities in the final phase of the CERV project Page 34 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 #@ETH-ICS-EI@# 5. OTHER 5.1 Ethics and EU values Ethics and EU values Describe ethics issues that may arise during the project implementation and the measures you intend to take to solve/avoid them. Describe how you will ensure gender and non-discrimination mainstreaming in the project cycle. This means integrating gender equality and non-discrimination considerations in the design, implementation, monitoring and evaluation of project activities. Projects activities should be pro-active and contribute to the equal empowerment of women and men, girls and boys, in all their diversity, and ensure that they achieve their full potential, enjoy the same rights and opportunities. Gender and non-discrimination mainstreaming are a key mechanism for achieving gender equality and combating multiple and intersecting discrimination. In the delivery of project activities gender mainstreaming shall be ensured by systematically monitoring access, participation, and benefits among different genders, and by incorporating remedial action that redresses any gender inequalities and discriminatory effects in implementation of planned activities. The activities shall also seek to reduce levels of discrimination suffered by particular groups (as well as those at risk of multiple discrimination) and to improve equality outcomes for individuals. If your project has a direct or indirect impact on children and their rights, indicate it clearly here. Make sure that your project is based on a child rights approach, i.e. that all the rights of the EU Charter of Fundamental Rights and the United Nations Convention on the Rights of the Child (UNCRC) and the Optional protocols, are promoted, respected, protected and fulfilled. The project should address children as rights holders and should ensure their participation in the design and implementation of the project. If you will have direct contacts with children you will have to provide a child protection policies in line with the Keeping Children Safe Child Safeguarding Standards. Explain how you intend to address privacy/data protection issues related to data collection, analysis and dissemination. Outline measures to be taken and the policies in place to guarantee full compliance with the EU values mentioned in Article 2 of the Treaty on the European Union and Article 21 of the EU Charter of Fundamental Rights. The project consortium does not expect the project to encounter any ethical issues during the implementation of the project. All partners of the project are aware of ethical issues, including personal data handling. The project team will fully comply with social research ethical standards, which will be discussed during first project meeting, to ensure all partners follow the guidelines. At TalTech, the project will be reviewed by the university’s Academic Ethics Committee for ethical considerations after the data collection, procession and preservation. Equality Bodies and ministries are required to follow the data protection law based on their institutional status in the government. At the initial stage of the project an Ethics Guideline will be elaborated to guide the project team on academic integrity and ethical principles of social research throughout the project lifetime. The Ethics Guideline will be elaborated at the first meeting of the project partners. The document will regulate various aspects related to the data collection, protection and preservation, ensuring confidentiality, plagiarism, preservation and exploitation of gender- sensitive data. The document will also regulate the non-discriminatory policy in respect to research and will monitor that the research outputs do not contain any hate speech, abusive information or cause stigmatization of any social group. EU standards in relation to data protection and privacy will be thoroughly observed. An ethical review of all proposals will be conducted by the participants own institution following their local procedures, all participants must fulfil EU ethics framework. Estonian Equality Body as the coordinator of the project will lead the elaboration of Ethics Guidelines mentioned above and control the following them. Any mistreatment of the data obtained within the project or the violation of ethical norms of social research, especially in respect to discriminating/abusing any social (or vulnerable) groups will be strictly monitored and respective measures taken by consortium of the project. The project will not involve interviews, the personal data will be collected from the databases. Project directly impacts the gender and non-discrimination mainstreaming in the project cycle as all the activities in the project are meant to develop the non-discrimination based on all the grounds of discrimination. Mapping, analysing and offering solutions how to avoid discrimination when using ADM, special attention will be given to the multiple and intersecting discrimination. As state institutions, specifically Equality Bodies, as participants of the project are established to protect general European values, including human dignity, democracy, rule of law, equality etc, provided by the Art 2 of the Treaty of the European Union and non-discrimination provided by the EU Charter of Fundamental Rights (Art 21). TalTech has longstanding practice in teaching and research of those EU values. #§ETH-ICS-EI§# #@SEC-URI-SU@# 5.2 Security Security Not applicable. #§SEC-URI-SU§# #@DEC-LAR-DL@# 6. DECLARATIONS Double funding Information concerning other EU grants for this project Please note that there is a strict prohibition of double funding from the EU budget YES/NO Page 35 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 (except under EU Synergies actions). We confirm that to our best knowledge neither the project as a whole nor any parts of it have YES benefitted from any other EU grant (including EU funding managed by authorities in EU Member States or other funding bodies, e.g. EU Regional Funds, EU Agricultural Funds, etc). If NO, explain and provide details. We confirm that to our best knowledge neither the project as a whole nor any parts of it are YES (nor will be) submitted for any other EU grant (including EU funding managed by authorities in EU Member States or other funding bodies, e.g. EU Regional Funds, EU Agricultural Funds, etc). If NO, explain and provide details. Financial support to third parties (if applicable) If in your project the maximum amount per third party will be more than the threshold amount set in the Call document, justify and explain why the higher amount is necessary in order to fulfil your project’s objectives. N/A #§DEC-LAR-DL§# Special Other annexes (annex 5 to Part B) — mandatory, if required in the Call document HISTORY OF CHANGES PUBLICATION VERSION CHANGE DATE 1.0 01.04.2021 Initial version (new MFF). 2.0 01.06.2022 Consolidation, formatting and layout changes. Tags added. 3.0 15.01.2024 New Deliverable was added into working package 1. As there was an accidental omission of a deliverable in Work Package 1, it has now been added: D1.1 Progress Report using the EC standard template, in English. 4.0 13.03.2024 Milestone 3 "Mid-term and final report" deleted 4.0 13.03.2024 Deliverable 2.1. supplemented with additional information. 4.0 13.03.2024 Deliverable 2.2. supplemented with additional information. 4.0 13.03.2024 The tasks of Working Package 2 supplemented with additional detailed information. 4.0 13.03.2024 Deliverable 3.1. supplemented with additional information. 4.0 13.03.2024 Deliverable 3.2. supplemented with additional information. 4.0 13.03.2024 Deliverable 3.3. supplemented with additional and more detailed information. 4.0 13.03.2024 Deliverable 3.4. supplemented with additional and more detailed information. 4.0 13.03.2024 Deliverable 4.1. supplemented with additional and more detailed information 4.0 13.03.2024 Tasks 4.1 and 4.2. are supplemented with additional information. 4.0 13.03.2024 The dissemination level of Deliverable 5.1. has been changed to sensitive and has been supplemented with additional information. 4.0 13.03.2024 The dissemination level of Deliverable 5.1. has been changed to sensitive and has been supplemented with additional information. 4.0 13.03.2024 A new deliverable is created - D5.3 The report on the study visit that will summarize the exchange of experiences and knowledge gained, contributing to the advancement of more inclusive and fair practices within the materials, guidelines, and training produced in the EquiTech project. 4.0 13.03.2024 A new deliverable is created - D5.4 Report of the implementation of the online course for public sector organizations in Estonia and Lithuania. The Deliverable is the result of the Task 5.4. 4.0 13.03.2024 Deliverable 6.2. supplemented with additional information. Page 36 of 37  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 Project: 101144709 - EquiTech - CERV-2023-EQUAL EU Grants: Description of the action (DoA) — Annex 1 (CERV): V1.0 – 01.04.2021 4.0 13.03.2024 A new Deliverable is added into WP6. D 6.3 Report on Public Awareness Raising Activities Concerning Potential Risks and Biases in Public ADM Systems 4.0 13.03.2024 A new Deliverable is added into WP6. D 6.4. Summary Report of the Academic Workshop 4.0 13.03.2024 A new Deliverable is added into WP6. D 6.5. Policy Development: Recommendations and Strategies for ADM Systems to Combat Discrimination 4.0 13.03.2024 A new Deliverable is added into WP6. D 6.6. Evaluation of the Value and Impact Created by the Activities in the Final Phase of the EquiTech Project 4.0 13.03.2024 Deliverable 6.1. supplemented with additional information 4.0 13.03.2024 The duration of the tasks in the timeline are updated, WP4 starts M12 (was M13) and the tasks of WP 5 start M2 (was M4). 5.0 14.03.2024 The value of the WP 3 and WP has been changed, 0.5 human months (€2250) has been from WP3 to WP5 for BEN05. The new values for these two WPs are: WP3: €111,865 and WP5: €5,717. 6.0 24.04.2024 EU contribution for WP5 increased by 158 EUR, as with the previous transfer, we did not take into consideration the corresponding indirect costs that should have been transferred. 158 EUR equals to 7% of 2500 EUR, multiplied by 90% (EU contribution). The modified values are correct for BEN5: WP3: 111707 EUR, WP5: 5875 EUR. Page 37 of 37  Grant Agreement number: 101144709 — EquiTech — CERV-2023-EQUAL CERV Associated Lump with Sum Ref. document MGAAres(2024)4260729 — Multi & Mono:-v1.0 12/06/2024 ANNEX 2 ESTIMATED BUDGET (LUMP SUM BREAKDOWN) FOR THE ACTION Estimated EU contribution Estimated eligible lump sum contributions (per work package) WP5 Capacity building and WP2 Discrimination WP3 Impact Assessment WP4 Testing the impact WP6 Dissemination WP1 Management and coordination training activities for national, Maximum grant amount1 mapping and analysis toolbox development assessment toolbox and awareness raising regional and local authorities Forms of funding Lump sum contribution Lump sum contribution Lump sum contribution Lump sum contribution Lump sum contribution Lump sum contribution a b c d e f g=a+b+c+d+e+f 1 - SVV 44 780.00 3 371.00 5 297.00 0.00 16 949.00 97 745.00 168 142.00 2 - OEOO 13 768.00 5 827.00 22 303.00 0.00 11 735.00 40 702.00 94 335.00 3 - TalTech 11 074.00 52 965.00 24 075.00 4 815.00 22 727.00 24 075.00 139 731.00 4 - MKM 6 163.00 10 400.00 15 601.00 22 245.00 10 786.00 5 778.00 70 973.00 5 - JUM 5 778.00 14 445.00 111 707.00 4 815.00 5 875.00 4 815.00 147 435.00 Σ consortium 81 563.00 87 008.00 178 983.00 31 875.00 68 072.00 173 115.00 620 616.00 1 The 'maximum grant amount' is the maximum grant amount fixed in the grant agreement (on the basis of the sum of the beneficiaries' lump sum shares for the work packages). Page 1 of 1 Grant Agreement number: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 3 ACCESSION FORM FOR BENEFICIARIES LYGIU GALIMYBIU KONTROLIERIAUS TARNYBA (OEOO), PIC 913111732, established in S KONARSKIO G 35, VILNIUS 03123, Lithuania, hereby agrees to become beneficiary in Agreement No 101144709 — EquiTech (‘the Agreement’) between SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE VOLINIKU KANTSELEI (SVV) and the European Union (‘EU’), represented by the European Commission (‘European Commission’ or ‘granting authority’), and mandates the coordinator to submit and sign in its name and on its behalf any amendments to the Agreement, in accordance with Article 39. By signing this accession form, the beneficiary accepts the grant and agrees to implement it in accordance with the Agreement, with all the obligations and terms and conditions it sets out. SIGNATURE For the beneficiary [--TGSMark#signature-913111732_75_210--] Mintaute Jurkute with ECAS id n002pid4 signed in the Participant Portal on 17/06/2024 at 07:51:40 (transaction id SigId-138-RzKEUD 3DbHh5SkJ7RO00YTdSzLEzundWkInprG5ha9C8tsaa5Qe31VzbvYm yevTM9dISSYDmstSBbjNzQ63CAe9-jpJZscgsw0KKC51j4wTR4y-9zqk HtEEkSQB5oLosUGXyUksS6mURjqN7an3rM55IaK1vRnzwXRCEH8tv 9zllOrE7WfBPErJLfzXLfYo3mQd7Oe). Timestamp by third party at 2024.06.17 07:51:45 CEST 1 Grant Agreement number: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 3 ACCESSION FORM FOR BENEFICIARIES TALLINNA TEHNIKAÜLIKOOL (TalTech), PIC 999842536, established in EHITAJATE TEE 5, TALLINN 19086, Estonia, hereby agrees to become beneficiary in Agreement No 101144709 — EquiTech (‘the Agreement’) between SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE VOLINIKU KANTSELEI (SVV) and the European Union (‘EU’), represented by the European Commission (‘European Commission’ or ‘granting authority’), and mandates the coordinator to submit and sign in its name and on its behalf any amendments to the Agreement, in accordance with Article 39. By signing this accession form, the beneficiary accepts the grant and agrees to implement it in accordance with the Agreement, with all the obligations and terms and conditions it sets out. SIGNATURE For the beneficiary [--TGSMark#signature-999842536_75_210--] Marika LUNDEN with ECAS id nlundema signed in the Participant Portal on 17/06/2024 at 10:52:31 (transaction id SigId-2443-zzzeu qzr29TpgHlJn9TfBonfjHnQ9qVjYnT45kqYxgKflnycaH0rFzlXzgA4nc N6chxTVL5rIDlV6HoYs3066th0-jpJZscgsw0KKC51j4wTR4y-tN7LTD psobwszkO1MNLBNGJbdotgi3HGfBJnuM71XdMofzvuljqo3rjt1r5jf DXAtVp0deDxry0yAwUcw4AcSZ). Timestamp by third party at 2024.06.17 10:52:35 CEST 2 Grant Agreement number: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 3 ACCESSION FORM FOR BENEFICIARIES MAJANDUS JA KOMMUNIKATSIOONIMINISTEERIUM (MKM), PIC 963638450, established in SUUR-AMEERIKA 1, TALLINN 10122, Estonia, hereby agrees to become beneficiary in Agreement No 101144709 — EquiTech (‘the Agreement’) between SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE VOLINIKU KANTSELEI (SVV) and the European Union (‘EU’), represented by the European Commission (‘European Commission’ or ‘granting authority’), and mandates the coordinator to submit and sign in its name and on its behalf any amendments to the Agreement, in accordance with Article 39. By signing this accession form, the beneficiary accepts the grant and agrees to implement it in accordance with the Agreement, with all the obligations and terms and conditions it sets out. SIGNATURE For the beneficiary [--TGSMark#signature-963638450_75_210--] 3 Grant Agreement number: 101144709 — EquiTech — CERV-2023-EQUAL Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 3 ACCESSION FORM FOR BENEFICIARIES MINISTRY OF JUSTICE (JUM), PIC 974582378, established in TONISMAAGI 5A, TALLINN 15191, Estonia, hereby agrees to become beneficiary in Agreement No 101144709 — EquiTech (‘the Agreement’) between SOOLISE VORDOIGUSLIKKUSE JA VORDSE KOHTLEMISE VOLINIKU KANTSELEI (SVV) and the European Union (‘EU’), represented by the European Commission (‘European Commission’ or ‘granting authority’), and mandates the coordinator to submit and sign in its name and on its behalf any amendments to the Agreement, in accordance with Article 39. By signing this accession form, the beneficiary accepts the grant and agrees to implement it in accordance with the Agreement, with all the obligations and terms and conditions it sets out. SIGNATURE For the beneficiary [--TGSMark#signature-974582378_75_210--] 4  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 4 XXX LUMP SUM MGA — MULTI & MONO FINANCIAL STATEMENT FOR THE ACTION FOR REPORTING PERIOD [NUMBER] EU contribution Eligible lump sum contributions (per work package) WP1 [name] WP2 [name] WP3 [name] WP4 [name] WP5 [name] WP6 [name] WP7 [name] WP8 [name] WP9 [name] WP10 [name] WP [XX] Requested EU contribution [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum [ Lump sum Forms of funding contribution][ Financing not contribution][ Financing contribution][ Financing contribution][ Financing contribution][ Financing not contribution][ Financing contribution][ Financing not contribution][ Financing contribution][ Financing not contribution][ Financing not contribution][ Financing not linked to costs] not linked to costs] not linked to costs] not linked to costs] linked to costs] not linked to costs] linked to costs] not linked to costs] linked to costs] linked to costs] linked to costs] Status of completion COMPLETED COMPLETED COMPLETED COMPLETED COMPLETED COMPLETED COMPLETED PARTIALLY COMPLETED PARTIALLY COMPLETED COMPLETED NOT COMPLETED a b c d e f g h i j k l = a + b+ c + d+ e+ f+ g+ h+ i+ j+ k 1 – [short name beneficiary] 1.1 – [short name affiliated entity] 2 – [short name beneficiary] 2.1 – [short name affiliated entity] X – [short name associated partner] Total consortium The consortium hereby confirms that: The information provided is complete, reliable and true. The lump sum contributions declared are eligible (in particular, the work packages have been completed and the work has been properly implemented and/or the results were achieved; see Article 6). The proper implementation of the action/achievement of the results can be substantiated by adequate records and supporting documentation that will be produced upon request or in the context of checks, reviews, audits and investigations (see Articles 19, 21 and 25).  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 ANNEX 5 SPECIFIC RULES INTELLECTUAL PROPERTY RIGHTS (IPR) — BACKGROUND AND RESULTS — ACCESS RIGHTS AND RIGHTS OF USE (— ARTICLE 16) Rights of use of the granting authority on results for information, communication, dissemination and publicity purposes The granting authority also has the right to exploit non-sensitive results of the action for information, communication, dissemination and publicity purposes, using any of the following modes: - use for its own purposes (in particular, making them available to persons working for the granting authority or any other EU service (including institutions, bodies, offices, agencies, etc.) or EU Member State institution or body; copying or reproducing them in whole or in part, in unlimited numbers; and communication through press information services) - distribution to the public in hard copies, in electronic or digital format, on the internet including social networks, as a downloadable or non-downloadable file - editing or redrafting (including shortening, summarising, changing, correcting, cutting, inserting elements (e.g. meta-data, legends or other graphic, visual, audio or text elements extracting parts (e.g. audio or video files), dividing into parts or use in a compilation - translation (including inserting subtitles/dubbing) in all official languages of EU - storage in paper, electronic or other form - archiving in line with applicable document-management rules - the right to authorise third parties to act on its behalf or sub-license to third parties, including if there is licensed background, any of the rights or modes of exploitation set out in this provision - processing, analysing, aggregating the results and producing derivative works - disseminating the results in widely accessible databases or indexes (such as through ‘open access’ or ‘open data’ portals or similar repositories, whether free of charge or not. The beneficiaries must ensure these rights of use for the whole duration they are protected by industrial or intellectual property rights. If results are subject to moral rights or third party rights (including intellectual property rights or rights of natural persons on their image and voice), the beneficiaries must ensure that they 1  Associated with document Ref. Ares(2024)4260729 - 12/06/2024 comply with their obligations under this Agreement (in particular, by obtaining the necessary licences and authorisations from the rights holders concerned). COMMUNICATION, DISSEMINATION AND VISIBILITY (— ARTICLE 17) Additional communication and dissemination activities The beneficiaries must engage in the following additional communication and dissemination activities: - present the project (including project summary, coordinator contact details, list of participants, European flag and funding statement and project results) on the beneficiaries’ websites or social media accounts - for actions involving publications, mention the action and the European flag and funding statement on the cover or the first pages following the editor's mention - for actions involving public events, display signs and posters mentioning the action and the European flag and funding statement - upload the public project results to the Rights and Values Project Results platform, available through the Funding & Tenders Portal. 2  Digitally sealed by the European Commission Date: 2024.06.12 22:30:56 CEST This electronic receipt is a digitally signed version of the document submitted by your organisation. Both the content of the document and a set of metadata have been digitally sealed. This digital signature mechanism, using a public-private key pair mechanism, uniquely binds this eReceipt to the modules of the Funding & Tenders Portal of the European Commission, to the transaction for which it was generated and ensures its full integrity. Therefore a complete digitally signed trail of the transaction is available both for your organisation and for the issuer of the eReceipt. Any attempt to modify the content will lead to a break of the integrity of the electronic signature, which can be verified at any time by clicking on the eReceipt validation symbol. More info about eReceipts can be found in the FAQ page of the Funding & Tenders Portal. (https://ec.europa.eu/info/funding-tenders/opportunities/portal/screen/support/faq) Commission européenne/Europese Commissie, 1049 Bruxelles/Brussel, BELGIQUE/BELGIË - Tel. +32 22991111  European Commission Citizens, Equality, Rights and Values Programme CERV-2023-EQUAL 101144709 - EquiTech - CERV-2023-EQUAL ETHICS GUIDELINE Version 1.0 2024 ANNEX 2  Ethics Guidelines Project: 101144709 - EquiTech - CERV-2023-EQUAL Ethics Guidelines Preamble The EquiTech project, funded under the European Commission's CERV- 2023-EQUAL program, focuses on addressing risks of discrimination, bias, and intolerance in automated decision-making processes to promote equality and human rights. This document outlines the ethical guidelines that will govern all project activities, ensuring adherence to the highest standards of integrity and fairness. All the project partners, in all their actions, adhere to the principles of gender equality and equal treatment. All forms of sexual and psychological harassment throughout the project shall not be tolerated. 1. Introduction At the project's inception, it is crucial to establish an ethical framework to guide all activities and decisions. These guidelines are developed in collaboration with all consortium members and are based on the principles of the European Code of Conduct for Research Integrity. 2. Scope and Purpose These guidelines apply to all individuals involved in the project, including researchers, technical staff, and administrative personnel. The ethics guideline is intended to ensure that all activities, especially data collection and analysis, are conducted while respecting individuals' privacy and the integrity of the research. 3. Principles 3.1 Academic Integrity Participants must adhere to principles of honesty, accountability, and transparency. Academic misconduct, including plagiarism and data fabrication, is strictly prohibited. 3.2 Data Protection and Privacy The collection and processing of personal data will comply with the General Data Protection Regulation (GDPR) and other relevant domestic laws, regulations and Partner guidelines. Data collection, storage, and usage must be authorized and secure. 3.3 Respect for Participants The privacy, rights, and welfare of human research participants must be highly respected. The informed consent process must be transparent and documented. 4. Implementation of Ethics Guideline 2  Ethics Guidelines Project: 101144709 - EquiTech - CERV-2023-EQUAL The project Management and Quality Board will monitor the implementation of these guidelines. 5. Data Collection and Handling 5.1 Collection Methods Data will only be collected using ethically approved methods that avoid any deception or coercion of participants. 5.2 Data Usage Data usage must conform to predefined objectives, which participants are aware of and have consented to. 6. Publication and Dissemination of Results The results of the project will be published in a manner that supports the principles of scientific truth and integrity. Declaration of conflicts of interest and transparency are mandatory. 7. Compliance Monitoring Compliance will be regularly monitored through audits and reviews by the project management and quality board. Non-compliance will be addressed with corrective measures, which may include suspension of project funding. 8. Amendments Amendments to the ethics guidelines will only be adopted after thorough discussion and consensual approval by the Project Management & Quality Board. Approval: the Coordinator 1. Office of Gender Equality and Equal Treatment Commissioner in Estonia The legal representative the Beneficiaries 2. Office of the Equal Opportunities Ombudsperson in Lithuania The legal representative 3. Tallinn University of Technology The legal representative 4. Ministry of Economic Affairs and Communications in Estonia The legal representative 3  Ethics Guidelines Project: 101144709 - EquiTech - CERV-2023-EQUAL 5. Ministry of Justice in Estonia The legal representative 4  European Commission Citizens, Equality, Rights and Values Programme CERV-2023-EQUAL 101144709 - EquiTech - CERV-2023-EQUAL CONSORTIUM AGREEMENT 2024  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL Consortium Agreement This consortium agreement (hereinafter referred to as "Contract'') is between the Coordinator 1. Office of Gender Equality and Equal Treatment Commissioner in Estonia, Kiriku 4, Tallinn 10130 Estonia, represented by Christian Veske, [email protected] and the Beneficiaries: 2. Office of the Equal Opportunities Ombudsperson in Lithuania, S Konarskio g 35, 03123, Vilnius, Lithuania, represented by Jolita Miliuvienė 3. Tallinn University of Technology, Ehitajate tee 5, 19086, Tallinn, Estonia, represented by Tanel Kerikmäe 4. Ministry of Economic Affairs and Communications of the Republic of Estonia, Suur-Ameerika 1, 10122, Tallinn, Estonia, represented by the Minister of Justice and Digital Affairs Liisa-Ly Pakosta 5. Ministry of Justice of the Republic of Estonia, Suur-Ameerika 1, 10122, Tallinn, Estonia, represented by Tõnis Saar hereinafter, jointly or individually, hereinafter referred to as "Parties" or "Party", relating to the project "Improving response to risks of discrimination, bias and intolerance in automated decision-making systems to promote equality" with project no 101144709 in short EquiTech, hereinafter referred to as "Project". 2  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL WHEREAS: The Parties, having considerable experience in the field concerned, have submitted a proposal for the Project to the European Commission as part of the Rights, Equality and Citizenship Programme (CERV-2023-EQUAL). The Parties wish to specify or supplement binding commitments among themselves in addition to the provisions of the specific Grant Agreement signed by the Parties and the European Commission in the EU Funding & Tenders Portal1 (hereinafter referred to as "Grant Agreement''), attached as Annex 1. 1 Available at: https://ec.europa.eu/info/funding-tenders/opportunities/portal/screen/home. 3  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL Table of Contents 1. Subject ____________________________________________________ 5 2. Duration ___________________________________________________ 5 3. Obligations of the Coordinator _____________________________ 5 4. Obligations of the Beneficiaries ____________________________ 6 5. Financing __________________________________________________ 6 6. Payments and financial management ______________________ 7 7. Intellectual property rights ________________________________ 9 8. Dissemination and communication ________________________ 12 9. Confidentiality ____________________________________________ 12 10. Data protection ___________________________________________ 14 11. Reports ___________________________________________________ 14 12. Monitoring and supervision _______________________________ 14 13. Liability __________________________________________________ 15 14. Termination of the Contract _______________________________ 15 15. Jurisdiction clause and dispute resolution _________________ 16 16. Amendments or additions to the Contract _________________ 16 17. Final provisions ___________________________________________ 16 4  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL 1. SUBJECT 1.1 The Coordinator and the Beneficiaries commit themselves to carrying out the Project. 1.2 The Project financing is primarily regulated by the Grant Agreement: 1.2.1 the total cost of the project for the contractual period referred to by the Grant Agreement, all financing combined, is estimated at 689 573,00 EUR (including all taxes and duties); 1.2.2 the maximum EU contribution (maximum requested EU grant amount) to cover expenditure incurred by the Parties participating in the Project shall be 620 616,00 EUR (90%). 1.3 The final financial contribution shall depend on the evaluation of the Project achievements pursuant to the rules laid down at EU level and in the Grant Agreement. 1.4 This Contract shall regulate relations between the Parties, and their respective rights and obligations with regard to their participation in the Project. 1.5 The subject matter of this Contract and the related works of the Project are covered by the approved application (attached as annex 2) and the Grant Agreement which form an integral part of this Contract and that each Party declares to have read and approved. 1.6 The Parties shall comply with the ethical guidelines of the Project (attached as Annex 3: Ethics guidelines) to ensure adherence of the highest standards of integrity and fairness. 2. DURATION 2.1 The Project has a duration of 24 months. 2.2 The Project began on 01.05.2024 and ends on 30.04.2026, unless terminated prior to the end date in accordance with the rules laid down in the Grant Agreement. 2.3 The Contract shall enter into force on the date on which it is signed by the authorised representatives of all Parties and shall apply retroactively from the starting date of the Project laid down in paragraph 2.2 until all Parties and the European Commission have fully performed their obligations. 2.4 The duration of the Contract can be extended in case an extension of the Project has been approved by the European Commission on the request of Parties or European Commission. 3. OBLIGATIONS OF THE COORDINATOR 3.1 The Coordinator shall: 3.1.1 take all the steps necessary to prepare for, perform and correctly manage the work to fulfill the Project as set out in the Grant Agreement, in accordance with the list of declared tasks, milestones and deliverables; 3.1.2 send the Beneficiaries any official documents received from the European Commission concerning the Project; 5  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL 3.1.3 notify and provide the Beneficiaries with any information, fact, problem or delay likely to affect the Project and any amendment made to the Grant Agreement; 3.1.4 comply with all the provisions of the Grant Agreement binding the Coordinator and the European Commission; 3.1.5 be in communication with the European Commission on behalf of Beneficiaries, if requested by Beneficiaries. 4. OBLIGATIONS OF THE BENEFICIARIES 4.1 Each Beneficiary shall: 4.1.1 take all the steps necessary to prepare for, perform and correctly manage the Beneficiary’s tasks set out in this Contract to fulfill the Project as set out in the Grant Agreement; 4.1.2 to cooperate with other Beneficiaries to fulfill the Project as set out in the Grant Agreement; 4.1.3 communicate to the Coordinator any information or document relating to the Project in the Beneficiary’s possession required by the Coordinator that is necessary for the management of the Project; 4.1.4 notify and provide the Coordinator with any information, fact, problem or delay likely to affect the Project; 4.1.5 comply with the EU and national financial regulations when implementing the Project; 4.1.6 bear the responsibility in submitting correct financial information to the Coordinator, including details of costs claimed and, where appropriate, ineligible expenses. 5. FINANCING 5.1 The total expenditure to be committed by each Beneficiary for the period covered by this Contract is set out in the Grant Agreement. 5.2 The European Commission contribution for each Partner shall be 90% of total expenditure specified in the Grant Agreement. 5.3 The financial contribution of the European Commission to the Project shall be distributed by the Coordinator according to: - the Grant Agreement - the approval of reports by the European Commission, and - the provisions of payment in Section 6. 5.4 A Party shall be funded only for its tasks carried out in accordance with the Contract and the Grant Agreement. 5.5 In accordance with generally recognized accounting and management principles and practices, each Party shall be solely responsible for justifying its costs with respect to the Project towards the European Commission. Neither the Coordinator nor any of the other Parties shall be in any way liable or responsible for justification of other Parties costs towards the European Commission. 5.6 A Party that spends less than its allocated share of the budget as set out in the Grant Agreement or – in case of reimbursement via unit costs - implements less units than foreseen in the Grant Agreement 6  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL will be funded in accordance with its actual duly justified eligible costs only. A Party that spends more than its allocated share of the budget as set out in the Grant Agreement will be funded only in respect of duly justified eligible costs up to an amount not exceeding the share fixed in the Grant Agreement. 5.7 In any case of a Party having received excess payments, the Party has to return the relevant amount to the Coordinator without undue delay. 5.8 In case a Party earns any receipt that is deductible from the total funding as set out in the Grant Agreement, the deduction is only directed toward the Party earning such income. The other Parties' financial share of the budget shall not be affected by one Party's receipt. In case the relevant receipt is more than the allocated share of the Party as set out in the Grant Agreement, the Party shall reimburse the funding reduction suffered by other Parties. 5.9 A Party leaving the consortium shall refund all payments it has received except the amount that was already approved as a result of activities by the European Commission. Furthermore, a defaulting Party shall, within the limits specified in Section 13, Liability of this Contract, bear any reasonable and justifiable additional costs occurring to the other Parties in order to perform its and their tasks due to the withdrawal of the Party. 6. PAYMENTS AND FINANCIAL MANAGEMENT 6.1 The European Commission’s contribution to the Project's costs will be calculated on the basis of rules specified in the Grant Agreement. 6.2 The Coordinator is entitled to withhold any payments to a Party who is in a breach of its obligations under the Contract or the Grant Agreement or to a Beneficiary which has not yet signed this Contract. 6.3 The Coordinator is entitled to recover any payments already paid to a defaulting party. The Coordinator is equally entitled to withhold payments to a Party when this is suggested by or agreed with the European Commission. 6.4 The Estonian Partners (Tallinn University of Technology, Ministry of Economic Affairs and Communications of the Republic of Estonia and Ministry of Justice of the Republic of Estonia) shall submit to the Coordinator on expenditure incurred during the eligibility period, quarterly balance statements about the expenses incurred and the transfer of advance payments to income. The Lithuanian Partner (Office of the Equal Opportunities Ombudsperson in Lithuania) shall submit to the Coordinator the notification of targeted funding on 31st of December each year. The balance and notification should be sent directly to Pille Kirsimäe (e-mail address [email protected]). 6.5 All amounts given in the Project reports must be expressed in euro. All payments from the Coordinator to the Beneficiaries will be made in euro. Declarations of expenditure by the Beneficiaries are to be made in euro. Any conversion into euro of actual costs incurred in 7  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL other currencies shall be made by the Beneficiary in accordance to the scheme described in the Grant Agreement. 6.6 The initial prefinancing shall be regarded as an advance pending explicit approval by the European Commission of the final report. 6.7 Beneficiary undertakes to repay to the Coordinator any amounts paid in excess of the final contribution from the European Commission within 45 days of the issuing of the debit note. 6.8 Payments will be made to the following bank accounts: Coordinator Office of Gender Equality and Equal Treatment Commissioner in Estonia Address Kiriku 4 Postal code 10130 Municipality/Town/ Tallinn City Country Estonia Name of the Ministry of Finance account holder Name of Bank AS SEB Pank IBAN EE891010220034796011 BIC / SWIFT EEUHEE2X Reference Number 2800048574 Subject line for EquiTech project transaction Beneficiary 2 Office of the Equal Opportunities Ombudsperson in Lithuania Address S Konarskio g 35 Postal code 03123 Municipality/Town/ Vilnius City Country Lithuania Name of the Lygių galimybių kontrolieriaus tarnyba account holder Name of Bank AB SEB bankas IBAN LT257044060008206818 BIC / SWIFT CBVILT2X Subject line for Equitech transaction Beneficiary 3 Tallinn University of Technology Address Ehitajate tee 5 Postal code 19086 Municipality/Town/ Tallinn City Country Estonia 8  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL Name of the Tallinn University of Technology account holder Name of Bank AS SEB Pank IBAN EE201010052037382001 BIC / SWIFT EEUHEE2X Subject line for EquiTech projekt/ Kristi Joamets transaction Beneficiary 4 Ministry of Economic Affairs and Communications in Estonia Address Suur-Ameerika 1 Postal code 10122 Municipality/Town/ Tallinn City Country Estonia Name of the Ministry of Finance account holder Name of Bank AS SEB Pank IBAN EE221010220027690221 BIC / SWIFT EEUHEE2X Reference number 2800045496 Subject line for 101144709 - EquiTech - CERV-2023-EQUAL transaction Beneficiary 5 Ministry of Justice in Estonia Address Suur-Ameerika 1 Postal code 10122 Municipality/Town/ Tallinn City Country Estonia Name of the Ministry of Finance account holder Name of Bank SEB Pank IBAN EE891010220034796011 BIC / SWIFT EEUHEE2X Reference number 2800049667 Subject line for 9J10-MU00-EquiTech transaction 7. INTELLECTUAL PROPERTY RIGHTS 7.1 “Intellectual Property Rights” means all intellectual property rights existing or subsequently arising throughout the world at the time of the conclusion of the Contract, including patents, utility models, industrial designs and trademarks, applications for registration and rights to apply for registration of the aforementioned, copyrights, trade secrets and rights to trade names and domain names. 9  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL 7.2 Except as otherwise expressly provided in the Contract, none of the Parties shall assign or license to the other Parties Intellectual Property Rights, except for the grant of a limited, royalty-free, and non- exclusive license to exercise the Intellectual Property Rights in a manner and to the extent strictly necessary to perform the Party’s obligations under the Contract. 7.3 If, in the course of performance of the Contract, the Beneficiaries creates or arranges for the creation of deliverables (“Deliverables”) that are the subject of Intellectual Property Rights and such Deliverables are transferred to the Coordinator, then: (a) the Beneficiaries assigns to the Coordinator all such Intellectual Property Rights in the Deliverables as may be assignable under applicable law, including economic copyrights and the rights to apply for the registration of industrial property; (b) the Beneficiaries grants to the Coordinator an irrevocable, worldwide, transferable, exclusive license valid throughout the entire term of the licensed rights, to exercise the remaining Intellectual Property Rights in the Deliverables for any purpose and in any manner, including by way of sublicensing, to the extent that such license can be granted under applicable law; (c) the Beneficiaries represents and warrants that the authors of the Deliverables or the elements included therein, or their successors in title, will not use their Intellectual Property Rights in any manner that prevents or restricts the exercise by the Coordinator, or any third party having acquired or licensed such rights from the Coordinator, of the Intellectual Property Rights assigned or licensed by the Beneficiaries to the Coordinator; (d) the Beneficiaries represents and warrants that the Deliverables do not contain any objects of Intellectual Property Rights of third parties other than those included in the Deliverables with the prior written consent of the Coordinator; (e) the Parties have agreed that the fee to be paid to the Beneficiaries under the Contract includes the fair remuneration for the assignments and licences provided for in this Section 7.3; (f) the assignments provided for in this Section 7.3 shall be deemed to have been made and the licences provided for in this Section 7.3 shall be deemed to have been granted at the moment of the creation of the Deliverables or any part thereof without the need for any further declarations of intent by any of the Party; (g) at the request of the Coordinator, the Beneficiaries shall provide evidence to the Coordinator of its fulfilment of its obligations under Subsections (a) to (f), providing, inter alia, the Coordinator with the contracts and other documents necessary for this purpose; (h) the Beneficiaries undertakes to indemnify the Coordinator for any loss or damage resulting from the Beneficiaries’ breach of the 10  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL representations and warranties provided for in this Section 7.3; and (i) if the applicable law requires any further performance, including making a declaration of intent or giving an affirmation or consent in a specified form, by one or more Parties or by any third party acting at the direction of any Party for the assignments and licences set out in this Section 7.3 to become effective, each obligated Party shall make such performance to the minimum extent required and within the shortest reasonable time possible, and the costs, including government fees, necessarily incurred in making such performance shall be borne by Beneficiary making the performance. 7.4 As the Beneficiaries have transferred (assigned) the economic copyrights of the Deliverables to a Coordinator and granted an authorisation (licence) to exercise the remaining Intellectual Property Rights in the Deliverables, whereas the goal of the Parties is to ensure that all Parties can use the results without restrictions, then the Coordinator grants to the Beneficiaries an irrevocable, worldwide, transferable, non-exclusive license valid throughout the entire term of the licensed rights, to use Intellectual Property Rights in the Deliverables and to exercise the Intellectual Property Rights in the Deliverables for any purpose and in any manner, including by way of sublicensing, to the extent that such license can be granted under applicable law. The Coordinator does not have the right to demand a separate additional fee from the Beneficiaries for granting a license. 7.5 For the sake of clarity reasons, the Coordinator and the Beneficiaries have, amongst others, the right to exploit Deliverables for information, communication, dissemination and publicity purposes, using any of the following modes: (a) use for its own purposes (in particular, making them available to persons working for the Coordinator or Beneficiaries, the European Commission or any EU service (including institutions, bodies, offices, agencies, etc.) or EU Member State institution or body; copying or reproducing them in whole or in part, in unlimited numbers; and communication through press information services); (b) distribution to the public in hard copies, in electronic or digital format, on the internet including social networks, as a downloadable or non-downloadable file; (c) editing or redrafting (including shortening, summarising, changing, correcting, cutting, inserting elements (e.g. meta-data, legends or other graphic, visual, audio or text elements extracting parts (e.g. audio or video files), dividing into parts or use in a compilation; (d) translation (including inserting subtitles/dubbing) in all official languages of EU; (e) storage in paper, electronic or other form; 11  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL (f) archiving in line with applicable document-management rules; (g) processing, analysing, aggregating the results and producing derivative works; (h) disseminating the results in widely accessible databases or indexes (such as through ‘open access’ or ‘open data’ portals or similar repositories), whether free of charge or not. 8. DISSEMINATION AND COMMUNICATION 8.1 All Project related dissemination and communication activities should be coordinated with other Parties. 8.2 Any objection to the planned publication shall be made in writing (an email suffices) to the other Party within 7 days after receipt of the notice of the planned publication. If no objection is made within the time limit stated above, the publication is permitted. An objection is justified if the publication contains confidential information of the objecting Party or is in contrary with the Contract or with the purpose of the Project. The objection has to include a precise request for necessary modifications. 8.3 If an objection has been raised, the involved Parties shall discuss how to overcome the justified grounds for the objection on a timely basis (for example by amendment to the planned publication and/or by protecting information before publication) and the objecting Party shall not unreasonably continue the opposition if appropriate actions are performed. 8.4 The Parties must engage in the following additional communication and dissemination activities: (a) for Project involving publications, mention the Project and the European Commission, Partners' and EquiTech logos together with funding statement on the cover or the first pages following the editor's mention; (b) for all Project involving materials and dissemination actions mention the Project and the European Commission, Partners' and EquiTech logos together with funding statement; (c) upload the Project results (deliverables) to the Funding & Tenders Portal. 8.5 The Parties present the Project (including Project summary, Coordinator contact details, list of Partners, the European Commission, Partners' and EquiTech logos together with funding statement and Project results) on the Parties websites or social media accounts. 9. CONFIDENTIALITY 9.1 “Confidential Information” means any non-public information disclosed by one Party (“Discloser”) to the other Party (“Recipient”) or otherwise made known to Recipient in the course of the negotiation and/or performance of the contracts or in course of Project implementation. 12  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL 9.2 The Recipient will use the Confidential Information only for the purposes of preparing the Contract, performing its obligations under the Contract, Grant Agreement or applicable law, or proving its claims in a legal dispute with the Discloser. 9.3 The Recipient may disclose Confidential Information: (a) to members of its governing bodies, employees, contractors, subcontractors, and consultants who have a need to know and who are bound by at least as restrictive contractual or legal confidentiality obligation; or (b) to comply with its legal obligation. 9.4 The obligation for confidentiality shall not apply for the information, which: (a) was public or generally available to public at the time of disclosure; (b) becomes public or generally available to public after the time of disclosure without any default of the Receiving Party; (c) was in the possession of the Receiving Party before the disclosure as demonstrated by its written records; (d) was received from a third party without any confidentiality obligations; or (e) is independently developed by the Receiving Party without any use of or reference to the received Confidential Information. 9.5 At the request of the Discloser, the Recipient shall promptly return or destroy all copies of the Confidential Information in its possession and confirm compliance with the request to the Discloser within 7 calendar days of receipt of the request, except: (a) if and insofar as it needs that Confidential Information for any purpose of the Project; or (b) if such copies are part of back-up copies made by or for the Recipient, in which case they may be retained until the end of the retention period of such back-up copies. 9.6 The obligation set out in Section 9.5 shall not apply to analyses, studies, reports, summaries or similar documents prepared by or on behalf of the Recipient that contain or rely on Confidential Information. 9.7 The Recipient shall protect the Confidential Information with at least the same degree of care as it protects its own confidential information, but at least with reasonable care. 9.8 Each Party shall promptly advise the other Parties in writing of any unauthorized disclosure, misappropriation or misuse of confidential information after it becomes aware of such unauthorized disclosure, misappropriation or misuse. 9.9 If any Party becomes aware that it will be required, or is likely to be required, to disclose Confidential Information in order to comply with applicable laws or regulations or with a court or administrative order, it shall, to the extent it is lawfully able to do so, prior to any such disclosure notify the Disclosing Party, and comply with the Disclosing 13  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL Party’s reasonable instructions to protect the confidentiality of the information. 9.10 The confidentiality obligations under this Contract shall not prevent the communication of confidential information to the European Commission. 9.11 Any breach of the obligations set out in this Section be considered a material breach and, in the case of a reasonable suspicion of breach, the Party shall, at the request of the other Party, prove that it has not breached an obligation. The obligation of confidentiality shall remain in force for 5 years after the termination of the Contract for whatever reason or expiration of this Contract. 10. DATA PROTECTION 10.1 Where personal data is processed in relation with the Contract, the parties shall comply with Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation). 10.2 A Party may process personal data of members of the governing bodies, employees, contractors, agents and consultants of the other Parties to the extent necessary for the purposes of the Contract, while ensuring compliance with all requirements under the applicable law and the Party’s own internal rules, and treating such personal data as Confidential Information within the meaning of Section 9. 10.3 If needed according to the General Data Protection Regulation, the Parties shall conclude separate data processing agreement. 11. REPORTS 11.1 Each Beneficiary shall provide the Coordinator with any information and document relating to the Project in the Beneficiary’s possession required for the preparation of the reports and, where appropriate, with copies of all the necessary supporting documents completed and signed by the legal representative within 20 working days from the date of the Coordinator's request (by e-mail) for access to information and documents. 11.2 Each Beneficiary shall provide the Coordinator with any information and document relating to the Project in the Beneficiary’s possession required for the preparation of the final report and, where appropriate, with copies of all the necessary supporting documents in the Beneficiary’s possession completed and signed by the legal representative by 14.04.2026 at the latest. 12. MONITORING AND SUPERVISION 12.1 Each Beneficiary shall provide without delay the Coordinator with any information relating to the Project in the Beneficiary’s possession that 14  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL the latter may reasonably request from Beneficiary concerning the Beneficiary’s tasks covered by this Contract. 12.2 Each Beneficiary shall make available to the Coordinator any document relating to the Project in the Beneficiary’s possession making it possible to check that the Beneficiary’s tasks regarding the Project is being or has been carried out. 12.3 The Beneficiaries shall keep at the European Commission’s disposal all original documents, especially accounting and tax records, or, in exceptional and duly justified cases, certified copies of original documents relating to the Grant Agreement stored on any appropriate medium that ensures their integrity in accordance with the applicable national legislation, for a period of seven (7) years in Estonia and ten (10) years in Lithuania from the final project implementation date, i. e. 30.04.2026. 13. LIABILITY 13.1 Without limiting the liability of any Party for the damages caused willfully or by gross negligence or by the breach of the representations or warranties set out in the Contract, the Party in breach of the Contract shall be liable for direct patrimonial damages caused by the breach to the other Parties. No Party shall be liable for loss of profit or non-patrimonial damages. 13.2 Each Party shall protect the European Commission, the other Parties and their personnel against any action for damages suffered by third parties, including Project personnel, as a result of its performance of this Contract, to the extent that these damages are not due to the negligence or willful misconduct of the European Commission or the other Parties or their personnel. 13.3 A Party’s failure to perform or delay in performing an obligation under the Contract is excusable if and to the extent that it is caused by force majeure, provided that the affected Party (i) takes all reasonably practicable steps to minimize the damage to the other Parties and (ii) notifies the other Parties of this fact at the earliest reasonable opportunity. 14. TERMINATION OF THE CONTRACT 14.1 If a Party has materially breached its obligations under the Contract (material breach), the Parties not in breach of the Contract shall have the right to jointly terminate the Contract towards breaching Party with immediate effect by notifying the breaching Party. A material breach of the Contract shall be deemed to have occurred in particular if: (a) the Party is in breach of an obligation that is considered material under the Contract; (b) the Party in breach has not remedied the breach and has not compensated for the damage caused by the breach within an additional term of at least 30 days given to it for that purpose, or 15  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL has declared that it has no intention to remedy the breach and has not compensated for the damage caused; (c) the Party has repeatedly failed to comply with an obligation for which it has previously been granted an additional term; or (d) the Party has breached its obligation intentionally or through gross negligence. 14.2 Obligations of the Parties which have arisen and fallen due prior to the termination of the Contract shall remain in force until they are performed in full. In the event of termination of the Contract, the Parties shall return to each other what has been transferred to them under the Contract or, if this is impossible, compensate each other for the value of what has been transferred. 14.3 Obligations of the Parties which, by their content and purpose, also govern the obligations of the Parties after the termination of the Contract, shall remain in force until the end of their term, irrespective of the circumstances of the termination of the Contract. 15. JURISDICTION CLAUSE AND DISPUTE RESOLUTION 15.1 The Contract shall be governed by the law in force in the Republic of Estonia. 15.2 The parties will try to resolve their disagreements by way of bona fide negotiations. In case of failure to reach an agreement, any dispute, controversy or claim arising out of or in connection with the Contract, or the breach, termination or invalidity thereof, will be settled in Harju County Court as a court of first instance. 16. AMENDMENTS OR ADDITIONS TO THE CONTRACT 16.1 Any Amendments to this Contract shall enter into force upon signature by the authorised representatives of all Parties, unless the Parties have agreed a different date of entry into force thereof. 17. FINAL PROVISIONS 17.1 No rights or obligations of the Parties arising from this contract may be assigned or transferred, in whole or in part, to any third party without the other Parties’ prior formal approval. 17.2 The Parties shall not be entitled to act or to make legally binding declarations on behalf of any other Party. Nothing in this Contract shall be deemed to constitute a joint venture, European Commission, partnership, interest grouping or any other kind of formal business grouping or entity between the Parties. 17.3 In case the terms of this Contract are in conflict with the terms of the Grant Agreement, the terms of the latter shall prevail. 17.4 Should any provision of this Contract become invalid, illegal or unenforceable, it shall not affect the validity of the remaining provisions of this Contract. In such a case, the Parties concerned shall be entitled to request that a valid and practicable provision be negotiated which fulfils the purpose of the original provision. 16  Consortium Agreement Project: 101144709 - EquiTech - CERV-2023-EQUAL 17.5 Notices provided for or relating to the Contract shall be given in a form that can be reproduced in writing or in electronic form in English or in any other language agreed between the Parties, unless a different form is provided for in the Contract or the applicable law for a particular notice. Notices sent to the email addresses of the Party’s contact persons specified in the Contract or its annex shall be deemed to have been received on the business day following the day on which they were sent; notices sent by post or courier shall be deemed to have been received on the 5th calendar day following the day on which they were sent. The contact details of a Party and its contact persons shall be deemed to be valid unless that Party has notified the other Party of a change thereof. 17.6 Contact details of the Parties are set out in the header of the Contract. The undersigned Parties declare that they have read and accepted the terms and conditions of this Contract. Annexes: 1. Grant Agreement; 2. Ethics guidelines Contact person: the Coordinator 1. Office of Gender Equality and Equal Treatment Commissioner in Estonia, Kiriku 4, Tallinn 10130 Estonia, Gea Kangilaski, [email protected]. and the Beneficiaries: 2. Office of the Equal Opportunities Ombudsperson in Lithuania, S Konarskio g 35, 03123, Vilnius, Lithuania, Rūta Juodelytė, [email protected], tel. +370 615 71064. 3. Tallinn University of Technology, Ehitajate tee 5, 19086, Tallinn, Estonia, Kristi Joamets, [email protected]. 4. Ministry of Economic Affairs and Communications of the Republic of Estonia, Suur-Ameerika 1, 10122, Tallinn, Estonia, Sofia Paes, [email protected]. 5. Ministry of Justice of the Republic of Estonia, Suur-Ameerika 1, 10122, Tallinn, Estonia, Henrik Trasberg, [email protected]. 17