On 11 May 2023, the European Parliament announced in a press release that the Internal Market Committee and the Civil Liberties Committee have adopted a draft negotiating mandate on the proposal for a regulation on harmonised rules on artificial intelligence (the AI Act). If approved, the AI Act would be the first regulation of artificial intelligence (AI) introduced in the world.

In their amendments to the Commission’s original proposal, MEPs are aiming to ensure that AI systems are overseen by people, are safe, transparent, traceable, non-discriminatory, and environmentally friendly.

The proposed AI Act follows a risk-based approach, prohibiting AI practices that pose unacceptable risks to safety, including manipulative techniques, social scoring, and discriminatory uses. MEPs expanded the list of prohibited practices to include real-time and post remote biometric identification, predictive policing systems, emotion recognition systems, and indiscriminate scraping of biometric data. High-risk areas have been broadened to include health, fundamental rights, environment, political campaigns, and recommender systems used by social media platforms.

Transparency measures have been introduced for providers of foundation models, such as GPT, including disclosure, prevention of illegal content generation, and publishing summaries of copyrighted data used for training. To support innovation, research activities and AI components provided under open-source licenses are exempt from the AI Act.

The proposed law promotes regulatory sandboxes for testing AI before deployment. MEPs have also emphasised citizens' rights, enabling them to file complaints about AI systems and receive explanations regarding decisions made by high-risk AI systems affecting their rights. The newly-formed EU AI Office would be responsible for monitoring the implementation of the AI rulebook.

Commentators, including the co-rapporteur of the Committee, Dragoş Tudorache, commented that “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 draft mandate requires endorsement from the whole Parliament during its 12-15 June session, before negotiations with the Council on the final wording of the law can commence.

The official press release can be found here.