20 Nations Push for International AI Oversight Body

Global coalition seeks AI governance framework

Twenty countries have issued a joint statement calling for strengthened international oversight of advanced artificial intelligence systems, citing concerns about rapid development and instances where AI has circumvented safety measures. The declaration, hosted by Finland's President, was released ahead of this week's United Nations General Assembly.

The statement emphasizes that "AI must remain under human direction, oversight and control" and be developed according to international law. Signatories include leaders from Canada, Kenya, the Netherlands, Australia, and South Africa, along with senior officials from Turkey and the United Arab Emirates.

Notably absent from the agreement are representatives from the United States, United Kingdom, and China—three nations at the forefront of AI development and deployment.

Three-pillar approach to AI safety

The statement outlines specific actions for both private companies and governments. It calls on AI developers to conduct safety testing before public release and permit independent evaluators to assess risks. Companies would need to open their systems to external scrutiny, a significant shift from current industry practices.

For governments and regional bodies, the document requests development of common safety standards that would enable countries worldwide to evaluate AI models with appropriate scientific rigor. This standardization aims to address the current fragmentation in how different jurisdictions approach AI regulation.

The third pillar proposes that UN member states consider establishing a new international institution capable of setting standards, enabling verification, and convening nations when AI systems cross critical capability thresholds.

Why it matters

The absence of major AI powers from this initiative highlights the challenge of achieving meaningful global coordination on technology governance. While twenty nations represent significant diplomatic momentum, any effective oversight framework will ultimately require participation from countries where the most advanced AI systems are being developed. The proposal for a new international body also raises questions about how it would interact with existing efforts and whether it could secure the authority needed to influence AI development in practice.

Existing landscape and next steps

Several international AI governance initiatives already operate under different mandates. The UN hosts the Independent International Scientific Panel on AI, comprising scientific experts from multiple countries, and the Global Dialogue on AI Governance, which serves as a forum for regulatory debate.

Nick Moes, Executive Director of AI governance nonprofit The Future Society, praised the statement as evidence of appetite for coordination but cautioned that "mobilizing diplomatic, political and financial capital will be needed to bring about the change that they agreed is necessary."

AI will feature prominently throughout this week's General Assembly proceedings, with the UN Security Council and U.S. representatives scheduled to hold dedicated AI sessions on Wednesday. These discussions come as nations compete to shape the AI agenda and maintain relevance in global technology debates.

These details were first reported by NBC News.