EUTA Letter to Support and Enable Data-driven Innovation in Europe

CONFIDENTIAL Minister Ivan Bartos Deputy Prime Minister for Digitisation Chair of the EU Telecommunications Council Minister Kristjan Järvan Minister of Entrepreneurship and Information Technology Brussels, 16 November 2022 Dear Minister Bartoš, Dear Minister Järvan, Ahead of the Telecommunications Council’s upcoming vote on the Artificial Intelligence Act (COM/2021/206), the European Tech Alliance (EUTA) invites you to support and enable data-driven innovation in Europe which is built on, and operates according to, European values. EUTA gathers major European digital champions and scaleups from many different sectors, business models and sizes. Our 36 members were created in 16 European countries and have grown in the EU by leveraging the power of artificial intelligence (AI) to innovate, improve their products and services, optimise their daily operations and better serve their customers. The AI Act will shape how European companies are able to compete globally. Ensuring a vibrant, digital and innovative ecosystem has never been more important, as Europe grapples with an increasingly unstable global environment and unprecedented economic challenges. We need a regulatory framework setting clear rules for the development, testing and use of AI in order to guarantee the safety and respect of fundamental rights while enabling innovation. We would particularly like to draw your attention to the use cases currently considered as "high risk" by Annex III of the AI Act. “High risk AI” use cases should be selected with care to ensure that only situations that convey high risks fall into scope. The EU rules need to avoid inadvertently capturing AI systems that pose no significant risk to health, safety, or fundamental rights. Otherwise, the ability of European companies to innovate will unnecessarily be put at risk. For instance, the inclusion of the AI use case to evaluate creditworthiness for low-value credits for the purchase of consumer products would be disproportionate (Annex III point 5(b)). A low-value credit has a low risk associated with it. The real risk levels vary for different values and purposes of the credits. The Council’s broad understanding may be misleading and could harm several growing sectors in Europe such as e-commerce and fintechs. The AI Act should only consider “high risk AI” as systems related to creditworthiness assessment, that is above a certain credit threshold, and not the ones with low-value credit. Another example would be the inclusion of placing targeted job advertisements as “high risk AI” (Annex III point 4 (a)). European companies of all sizes, rely on AI to propose goods or services tailored to and desired by their customers and users. Such a use enables customers and users to see relevant content or ads without facing a significant risk to their health, safety or fundamental rights. If advertising is to be generally European Tech Alliance | eutechalliance.eu | [email protected] 1  CONFIDENTIAL considered as a “high risk AI” use case, this would completely dilute the notion of high-risk AI and render the exception in Article 6.3. useless.1 It is important that the AI Act keeps a high threshold limited to use cases where AI can have severe consequences on individuals. Such an approach has been taken by the Council where it has distinguished potential high-risk task allocation systems as systems based on individual behaviour or personal traits or characteristics (Annex III point 4 (b)).This is a positive development, and we would like to emphasise the importance of similar and nuanced clarifications in Annex III. As the European Commission indicates in its proposal, the identification of high-risk AI should be based not only on the severity of the possible harm and its probability of occurrence, but also on whether the AI system “has already caused harm to health and safety, an adverse impact on fundamental rights, given rise to significant concerns in relation to the materialisation of such harm or adverse impact, as demonstrated by reports or documented allegations”. A broader understanding might alienate the concept of “high risk AI” and limit European companies’ ability to compete internationally. The European Tech Alliance urges you to support pro-innovation rules that adopt a risk-based and proportionate approach. We hope you can take our concerns into account and that we can work together to ensure that the AI Act supports entrepreneurship and technological advancements in Europe while ensuring EU values are fully respected. We look forward to exchanging views with you. Please let us know if you need any additional information from us. Sincerely, Victoria de Posson Secretary General, European Tech Alliance 1 Article 6.3.: “AI systems referred to in Annex III shall be considered high-risk unless the output of the system is purely accessory in respect of the relevant action or decision to be taken and is not therefore likely to lead to a significant risk to the health, safety or fundamental rights.” European Tech Alliance | eutechalliance.eu | [email protected] 2  CONFIDENTIAL Our Members European Tech Alliance | eutechalliance.eu | [email protected] 3 European Tech Alliance About Us NAME European Tech Alliance MEMBERS 36 Companies INDUSTRY Technology and Digital Companies PRESENCE 16 European Countries FOUNDING YEAR 2015 VISION MISSION Develop smart policies promoting European tech Create a better future for Europe through innovation, investments and competitiveness. It is technology and based on shared EU values. important to create the right regulatory conditions We aim to contribute to our local economies which both enable European tech champions to and build a sustainable, greener, innovative grow and empower consumers. and inclusive Europe for future generations. LEADERSHIP Kristin Skogen Lund Aurélie Caulier Victoria de Posson EUTA President EUTA Chair EUTA Secretary General CEO of Schibsted Head of Public Affairs EU, Zalando EUTA INTERESTS OUR MEMBERS Online Platforms (e.g., Digital Services Act) Competition (e.g., Digital Markets Act, Antitrust) Artificial Intelligence (e.g., AI Act, AI liability) Privacy and Data Protection (e.g., ePrivacy, Data Act, GDPR) Sustainability (e.g., Ecodesign, Empowering consumers for the green transition) CONTACT Secretary General: [email protected] | EUTA Secretariat: [email protected] Saatja: Victoria de Posson <[email protected]> Saadetud: 16.11.2022 16:19 Adressaat: MKM info <[email protected]> Koopia: <[email protected]>; <[email protected]> Teema: AI Act | EUTA Letter to Support and Enable Data-driven Innovation in Europe Dear Minister Järvan, Ahead of the Telecommunications Councils upcoming vote on the Artificial Intelligence Act (), the (EUTA) invites you to support and enable data- driven innovation in Europe which is built on, and operates according to, European values. EUTA gathers major European digital champions and scaleups from many different sectors, business models and sizes. Our were created in 16 European countries and have grown in the EU by leveraging the power of artificial intelligence (AI) to innovate, improve their products and services, optimise their daily operations and better serve their customers. Please find our letter attached. We look forward to exchanging views with you. Please let us know if you need any additional information from us. Sincerely, Victoria Victoria de PossonSecretary General&#43;32 476 25 08 16 | | EU transparency register: 189607519323-76