UN AI governance bodies face pressure to accelerate action

The United Nations is expanding its role in artificial intelligence governance at a moment when the technology's own architects are calling for restraint — but the institution's deliberate pace may not match the urgency of the moment.

World leaders gathering in New York this week will hear OpenAI CEO Sam Altman brief the Security Council on AI's implications for international security. The timing is notable: Anthropic CEO Dario Amodei recently called for AI developers to "pace the frontier," slowing development until safety work and independent oversight can catch up. Altman, Elon Musk, and Google DeepMind's Demis Hassabis have endorsed the concept without committing to specifics.

Why it matters

The gap between AI companies' public calls for government coordination and their actual engagement with UN processes reveals a fundamental tension in global AI governance. While frontier labs urge international action, they've largely avoided meaningful participation in the multilateral forums designed for exactly that purpose — leaving governments to regulate technologies they struggle to understand.

The UN's AI architecture takes shape

The UN has built substantial infrastructure for AI governance over the past three years. Secretary-General António Guterres convened a High-level Advisory Body on AI in 2023. All 193 member states adopted the Global Digital Compact in 2024, committing to greater international cooperation. In August 2025, the General Assembly formally established two bodies: the Independent International Scientific Panel on AI and the Global Dialogue on AI Governance.

The first Global Dialogue convened in Geneva in July, drawing more than 3,000 participants from 163 countries and over 1,000 stakeholder organizations. The scientific panel presented its first report, intended to give governments a common evidence base.

Industry has some representation through the ITU's AI for Good Global Commission, launched during the Geneva meetings, which includes members from Anthropic, Amazon, Cohere, Google, Microsoft, and Nvidia.

Speed versus deliberation

"The brakes sit with a handful of AI companies and the few governments that host them. What the UN can do is make sure the rest of the world is not simply the track that the AI race is run on," said Nighat Dad, a member of the UN's Scientific Panel, according to Tech Policy Press.

Tony Oweke, a Council on Foreign Relations research fellow and former digital adviser to Kenya's UN mission, told Tech Policy Press that recent developments "create more political impetus towards action." But he cautioned that "as far as the United Nations is concerned, it isn't likely to change anything materially, at least in the short run."

Both the panel and dialogue can only be reviewed — potentially gaining "more teeth to start working on norms and principle-building" — in 2027 or 2028, Oweke noted.

Where international coordination could work

Friederike Schüür, a senior policy researcher at the Ada Lovelace Institute who attended the Geneva dialogue, argues the UN should focus on problems that cross national and regional boundaries. She suggests AI incident monitoring as one area where international coordination could prove achievable, rather than requiring every country to build separate surveillance systems.

Doreen Bogdan-Martin, secretary-general of the UN's International Telecommunication Union, argued last week that standards can translate broad AI safety commitments into practical requirements.

Yet a fundamental challenge remains: governments pursue different goals, and some countries benefit more from rapid AI development than others. Within the United States, skepticism about the UN's role persists. "Any role for the UN beyond convening blunt discussions will entirely depend on shared political will," Oweke explained, calling it "an extremely uphill task."

The details in this article were first reported by Chris Stokel-Walker for Tech Policy Press.