The chief executives of OpenAI and Anthropic briefed the 15-member UN Security Council on Wednesday, at its first high-level session on the safety risks of increasingly capable AI. France, which holds the rotating presidency, convened it. Sam Altman and Dario Amodei were joined by Hugging Face co-founder Clément Delangue and Yoshua Bengio, co-chair of the UN’s Independent International Scientific Panel on AI.
Amodei warned that AI, if managed poorly, could be a risk to humanity as a whole, and set out three areas for cooperation: narrow global agreements such as a ban on using AI to make biological weapons, systems that let countries verify each other’s commitments, and common testing standards backed by a notification system for AI security incidents. Altman said the most important decisions about AI cannot be left to labs in San Francisco and must run through governments accountable to their citizens. Bengio called the dangers real and imminent, and beyond the reach of any single country.
The timing gave the session its edge. Donald Trump used his address to the General Assembly this week to reject what he called globalist schemes to control AI, saying the United States would encourage the technology’s development rather than restrict it. Two leading American AI companies then asked a UN body for international standards.
What the briefing did not produce was anything binding. No resolution or outcome document was reported, and Altman’s emphasis on democratic accountability sits awkwardly with a council on which China holds a veto. A ban on AI-designed bioweapons is the kind of narrow agreement that might pass. Verification, the piece that would make any commitment enforceable, would require rival governments to let each other inspect their labs, and that is a far harder ask.
