Council of Europe Framework Convention on Artificial Intelligence, Human Rights, Democracy and the Rule of Law

The Council of Europe Framework Convention on Artificial Intelligence, Human Rights, Democracy and the Rule of Law was opened for signature on 5 September 2024.

Often overshadowed by Regulation (EU) 2024/1689 of 13 June 2024 on artificial intelligence in the European Union (AI Act), it is helping to strengthen Europe's position in this area.

The Framework Convention lays the foundations for the ethical development of artificial intelligence, respectful of human rights, in a multilateral perspective, beyond Europe.

* * *

The Framework Convention on AI in brief

On 17 May 2024, the foreign ministers of the 46 member states of the Council of Europe adopted the Framework Convention on Artificial Intelligence, Human Rights, Democracy and the Rule of Law, the first legally binding international instrument in this field. It was opened for signature on 5 September 2024.

The Framework Convention aims to guarantee the protection of fundamental rights at every stage in the life cycle of artificial intelligence systems, and to respond to the risks of discrimination, threats to democratic processes, breaches of privacy and repressive use of AI by certain governments.

Its principles and obligations relating to the use of AI by public authorities and private players are thus focused on the protection of human rights, support for democracy and respect for the rule of law, and responsible technological innovation:

• transparency requirements (with exceptions in the interests of public order, national security or in the context of research and development activities) and control;
• decision-making processes and the operation of AI models must be understandable (better understanding should help determine responsibility in the event of negative impacts on human rights, democracy or the rule of law) and accessible to the stakeholders concerned;
• informing users when they interact with AI;
• enabling the persons concerned to challenge the decisions taken by an AI.

The Framework Convention requires States to:

• take measures to ensure that those responsible for AI systems anticipate and mitigate risks;
• assess the need for moratoria or bans if the use of AI systems threatens fundamental rights;
• provide remedies for victims of human rights violations;
• set up an independent body to monitor the application of the rules established by the Framework Convention.

The Framework Convention is intended to be technologically neutral in order to remain relevant in a context of rapid technological advances.

Complementarity between the Framework Convention on AI and the European Union's AI Act

These two bodies of law share the same objective of regulating artificial intelligence, while differing in their scope, nature and field of application.

Designed to be compatible with the AI Act, the Framework Convention provides an ethical basis and principles applicable at global level, while the AI Act acts as a technical instrument for the EU focused on its internal market.

The Framework Convention was drawn up by an intergovernmental body bringing together the 46 member states of the Council of Europe, the European Union and 11 non-member states (United States of America, Canada, Japan, Mexico, Holy See, Australia, Argentina, Costa Rica, Israel, Peru and Uruguay).

To date, the Framework Convention has been signed by the European Union (on behalf of its 27 Member States), Andorra, Georgia, Iceland, Montenegro, Norway, the Republic of Moldova, the United Kingdom, San Marino, Israel and the United States of America.

* * *