Artificial Intelligence for Europe Irina Orssich DG CNECT

Artificial Intelligence for Europe Irina Orssich DG CNECT European Commission

EU strategy for AI - April 2018 – October 2019 A STRATEGY FOR EUROPE TO LEAD THE WAY Boost technological and industrial capacity & AI uptake 12/18 Coordinated Plan with MS 4/19 Ethics Guidelines for Trustworthy AI Prepare for socio-economic changes 4/19 Expert Group Future of Work 5/19 Servoz Report Ensure an appropriate ethical & legal framework 6/19 Policy and Investment Recommendations for Trustworthy AI Expert Group Saftey and liability

Joining forces with MS On 10 April 2018 25 European countries signed a Declaration of cooperation on Artificial Intelligence. Member States agreed to work together on the most important issues raised by Artificial Intelligence, from ensuring Europe's competitiveness in the research and deployment of AI, to dealing with social, economic, ethical and legal questions! The initiative has been joined by all 28 Member States and Norway. 7 December 2018: Coordinated plan on AI published All Member States to put in place AI strategy by mid 2019. The coordinated plan will be reviewed annually

Coordinated plan of 7 December • Maximising investment through partnerships • Common strategic research and innovation agenda • Resources for start-ups and innovators • From the lab to the market • Networks of European AI research excellence centres • World-reference testing facilities • Fostering AI take-up through Digital Innovation Hubs • Talent skills and life-long learning • A European data space • AI for the Public sector • Ethical guidelines and regulatory framework • International Cooperation

High-Level Expert Group and mandate Chair: Pekka Ala-Pietilä 52 members from: Industry Academia Civil society Two deliverables • Ethics Guidelines for Artificial Intelligence • Policy & Investment Recommendations Interaction with European AI Alliance • Broad multi-stakeholder platform counting over 4500 members to discuss AI policy in Europe

Ethics Guidelines for AI – Requirements Human agency and oversight Diversity, nondiscrimination and fairness Technical Robustness and safety Societal & environmental well-being Privacy and data governance Accountability Transparency To be continuously implemented & evaluated throughout AI system’s life cycle

Ethics Guidelines for AI – Assessment List Assessment list to operationalise the requirements • Practical questions for each requirement – 131 in total • Test through piloting process to collect feedback from all stakeholders (public & private sector) Official launch of piloting: 28 June – Stakeholder event https: //ec. europa. eu/futurium/en/register-piloting-process

Policy & Investment Recommendations 1. Empower and protect humans and society 2. Take up a tailored approach to the AI market 3. Secure a Single European Market for Trustworthy AI 4. Enable AI ecosystems through sectoral multi-stakeholder alliances 5. Foster the European data economy 6. Exploit the multi-faceted role of the public sector 7. Strengthen and unite Europe's research capabilities 8. Nurture education to the Fourth Power 9. Adopt a risk-based governance approach to AI and ensure an appropriate regulatory framework 10. Stimulate an open and lucrative investment environment 11. Embrace a holistic way of working, combining a 10 -year vision with a rolling action plan

March 2017 Special Eurobarometer 460 • The majority agree robots and artificial intelligence technologies require careful management Almost nine in ten respondents (88%) agree robots and artificial intelligence are technologies that require careful management. More than half (53%) “totally agree”, while a further 35% “tend to agree”. Just 7% disagree.

Question 1: “Is a voluntary approach to AI ethics the right tool to create an environment of trust and innovation in Europe? ” Multiple choice answer We don’t need any new rules for AI in Europe, only guidance A voluntary selfcertification scheme for AI ethics would be helpful There should be mandatory rules in all circumstances Depending on the impact of the AI system, a combination of voluntary mechanisms, guidance and mandatory rules Result (%) 5 6 13 76 Question 2: “Which of the 7 requirements from the Ethics Guidelines for trustworthy AI would you find most important to regulate? ” Multiple choice answer Result (%) Human agency and oversight 24 Technical robustness and safety 20 Privacy and data governance 28 Transparency 33 Diversity, non-discrimination 25 and fairness Societal and environmental well 10 -being Accountability None of these 31 4

President-elect Ursula von der Leyen: • In my first 100 days in office, I will put forward legislation for a coordinated European approach on the human and ethical implications of Artificial Intelligence. … • I will make sure that we prioritise investments in Artificial Intelligence, both through the Multiannual Financial Framework and through the increased use of public-private partnerships.

Thank you