Council of Europe Framework Convention on AI

Today, the Commission has signed the Council of Europe Framework Convention on Artificial Intelligence (AI) on behalf of the EU.
The Convention is the first legally binding international agreement on AI and it is  fully in line with the EU AI Act, the first comprehensive AI regulation in the world.

The Convention provides for a common approach to ensure that AI systems are compatible with human rights, democracy and the rule of law, while enabling innovation and trust. It includes a number of key concepts from the EU AI Act such as a risk-based approach, transparency along the value chain of AI systems and AI-generated content, detailed documentation obligations for AI systems identified as high-risk, and risk management obligations with the possibility to introduce bans for AI systems considered a clear threat to fundamental rights.

The signature took place at the informal conference of Council of Europe Ministers of Justice in Vilnius, Lithuania. Negotiating parties included the EU, other Council of Europe Member States, the Holy See, the USA, Canada, Mexico, Japan, Israel, Australia, Argentina, Peru, Uruguay, and Costa Rica. Input from 68 international representatives from civil society, academia, industry and other international organisations also ensured a comprehensive and inclusive approach. The Council of Europe Convention signed today is part of the EU's broader efforts in AI at the international level, which include discussions at the G7, the OECD, the G20 and the United Nations.

The Convention will be implemented in the EU by means of the AI Act, which provides harmonised rules for the placing on the market, deployment, and use of AI systems, along with other relevant EU legislations where needed.

After the signature, the Commission will prepare a proposal for a Council decision to conclude the Convention. The European Parliament should also give its consent.