The Artificial Intelligence Act Draft has been adopted by the European Parliament’s Internal Market Committee and Civil Liberties Committees with 84 votes in favour, 7 against and 12 abstentions. The AI Act, launched by the Commission on April 2021, aims at strengthening rules around data quality, transparency, human oversight and accountability of AI, as well as at addressing ethical questions and implementation challenges in various sectors. 

The draft adopted differs in some points substantially from the original Commission proposal: 

• The Act now defines AI as a ‘machine-based system designed to operate with varying levels of autonomy and that can, for explicit or implicit objectives, generate outputs such as predictions, recommendations, or decisions, that influence physical or virtual environments’. 
• The list of prohibited practices has been substantially extended to include biometric categorisation, predictive policing, and scrapping facial images for building databases. 
• Providers of foundation models will have to comply with new obligations to guarantee robust protection of fundamental rights, health and safety and the environment, democracy and rule of law. Generative models will have to fulfill additional transparency requirement. 
• AI Applications need to pose a significant risk to harm people’s health, safety, or fundamental rights in order to be classified as high risk. 
• High risk AI applications providers now face more prescriptive obligations, especially regarding risk management, data governance, technical documentation and record keeping. 
• The mandate of the AI Office, the main enforcement body, have been limited to a supportive role, such as providing guidance and coordinating joint investigations. The Commission is now in charge of settling disputes among national authorities on dangerous AI systems. 

The Act is a pioneering initiative to regulate AI that could establish itself as a global standard and put the EU at the forefront of regulation in this area. In the words of Co-rapporteur Dragos Tudorache (Renew, Romania): “Given the profound transformative impact AI will have on our societies and economies, the AI Act is very likely the most important piece of legislation in this mandate. It’s the first piece of legislation of this kind worldwide, which means that the EU can lead the way in making AI human-centric, trustworthy and safe’’. 

The final adoption of the act will be debated in the plenary of the Parliament in mid-June. After that, following the ordinary legislative process, negotiations between Parliament, the Commission and the Council of the EU will begin.