The Council of Europe has introduced a treaty described as “technology-neutral,” aimed at establishing a legal framework for the entire lifecycle of AI systems while fostering innovation. The agreement encompasses three primary areas of protection:
-
Human Rights: This aspect focuses on upholding existing human rights agreements, ensuring that individuals’ data is used appropriately, their privacy is respected, and that AI systems do not engage in discrimination.
-
Democracy: This addresses the risks posed by malicious actors disseminating misinformation and disinformation, which could undermine the integrity of information and erode public trust.
- Rule of Law: This places the responsibility on signatory countries to regulate and mitigate AI-specific risks.
In May, the framework received endorsement from the 46 member states of the Council of Europe, the European Union, and 11 non-member states including Argentina, Australia, Canada, Costa Rica, the Holy See, Israel, Japan, Mexico, Peru, the United States, and Uruguay. Alongside the US, EU, and UK, other signatories now include Andorra, Georgia, Iceland, Norway, the Republic of Moldova, San Marino, and Israel.
Marija Pejčinović Burić, Secretary General of the Council of Europe, emphasized the importance of ensuring that the emergence of AI upholds established standards rather than undermines them. She described the Framework Convention as a strong, balanced document, reflecting an open and inclusive drafting process that incorporates a variety of expert perspectives. She expressed hope that this initial wave of signatures would quickly lead to ratifications, enabling the treaty to come into effect promptly.
UK Lord Chancellor and Justice Secretary Shabana Mahmood stated that while artificial intelligence has the potential to significantly enhance public services and stimulate economic growth, it is crucial to ensure that society shapes AI rather than the other way around. She regarded the convention as a significant step toward harnessing new technologies without compromising fundamental values such as human rights and the rule of law.
Dr. Kjell Carlsson, head of AI strategy at Domino Data Lab, acknowledged the real threats posed by AI, such as fraud, deepfakes, and cybercrime, and commended lawmakers for their proactive approach. However, he expressed skepticism regarding the treaty’s effectiveness in addressing specific real-world AI risks due to its general and vague nature, questioning its enforceability. He noted that the trend toward increasing AI regulation is inevitable, as demonstrated by the EU AI Act and various legislative efforts across US states, with many AI-related risks already being addressed by existing regulations.