On 5 September 2024, during the informal conference of Council of Europe Ministers of Justice in Vilnius (Lithuania), the European Commission signed the Council of Europe Framework Convention on Artificial Intelligence (AI) on behalf of the European Union. This Convention marks a historic milestone as the first legally binding international agreement on AI, aimed at ensuring that AI systems align with human rights, democracy, and the rule of law.
The European Commission Vice-President for Values and Transparency, Věra Jourová, represented the EU at the signing ceremony alongside other key officials from the Council of Europe and international partners. The Convention is fully in line with the EU's pioneering AI Act, the world’s first comprehensive regulation on AI. In her remarks, Jourová highlighted the significance of this agreement, emphasising that it sets the global standard for AI governance while safeguarding fundamental rights.
Andorra, Georgia, Iceland, Norway, the Republic of Moldova, San Marino, the United Kingdom as well as Israel and the United States of America also signed the Framework Convention.
Ministers of Justice from Council of Europe Member States attended the ceremony, along with representatives from the Holy See, the USA, Canada, Japan, and other non-European nations such as Mexico and Uruguay. The presence of these diverse stakeholders underscores the international commitment to shaping AI’s role in society in a way that respects democratic values.
Highlighting the Milestone in AI Governance
The Council of Europe’s Framework Convention on AI is the result of nearly two years of intensive negotiations led by the Committee on AI (CAI), in which the EU actively participated and which brought together experts from civil society, academia, industry, and governments. This agreement stands as a crucial framework that aligns with the principles of the EU AI Act, reinforcing a human-centric approach to AI development. The Convention will be implemented in the EU through the AI Act, which lays down harmonised rules for the placing on the market, deployment, and use of AI systems, as well as other relevant EU legislations where needed.
The Council of Europe’s Framework Convention on AI addresses key areas, including transparency in AI-generated content, rigorous risk management, documentation obligations for high-risk AI systems, and the protection of human rights. It introduces innovative regulatory mechanisms, such as the concept of “regulatory sandboxes” to promote safe AI experimentation, while setting global standards for accountability, safety, and fairness in AI.
Crucially, the Convention also recognises the need for a risk-based approach, ensuring that AI systems considered harmful to fundamental rights are subject to strict oversight. This creates a balanced approach that encourages innovation but places safeguards to prevent misuse and harmful consequences of AI technologies.
Setting the Stage for Global AI Governance
The Convention is a significant step forward in the international AI landscape, offering a common and global approach to govern AI systems and mitigate the risks they pose to human rights, democracy, and the rule of law. By focusing on principles of transparency, accountability, and innovation, it ensures AI systems are designed to support ethical governance while fostering public trust.
Additionally, the Convention provides room for flexibility. While public authorities are required to comply fully, private actors can adopt alternative measures to meet the same objectives, allowing for responsible innovation. Certain exemptions apply, notably for research, development, and national security purposes, ensuring that AI advances continue without compromising core democratic values.
The Next Steps
With the Convention now signed, the European Commission will submit a proposal for a Council decision to formalise the EU's participation in the Convention. The European Parliament will also be called upon to provide its consent, marking the next crucial phase in implementing this ground-breaking agreement.
As the world looks to the future of AI, the Council of Europe’s Framework Convention stands as a testament to the commitment of international actors to secure a fair and ethical AI landscape for all. The Convention, however, has sparked debate, with some arguing that it could stifle innovation while others see it as a crucial step towards responsible AI development.
Navigating the AI Landscape: Striking a Balance
Switzerland, a nation at the forefront of AI development, has a unique perspective on this global debate. While Switzerland endorses the Convention, it has not yet signed it, emphasizing the need for a careful and deliberate approach to regulating AI.
The Swiss Federal Office of Communications is currently drafting recommendations for how Switzerland could regulate AI in the future, with a focus on finding a balance between promoting innovation and ensuring AI’s responsible development. This process involves consulting with stakeholders from across the country, including representatives from the economy, civil society, the cantons, and other interested parties.
The Convention itself has also faced criticism, with some NGOs expressing concerns that it lacks legal teeth and focuses too heavily on state activities, potentially leaving private companies with too much freedom to operate. There are concerns that the Convention could be too vague to be effective, and that it may be difficult to enforce. However, the Convention is a significant step forward in the international AI landscape. It sets an important precedent for the global community, and it is likely to be a crucial factor in shaping the future of AI development.
The treaty also highlights the need for a global approach to AI regulation, as the technology has the potential to impact people around the world. The Convention is a positive step in this direction, and it is likely to spur further dialogue and collaboration between nations as they work to ensure that AI is developed and used in a way that benefits all of humanity. Only time will tell whether the Convention will be a success, but it is certainly a significant step in the right direction.
