UN Security Council AI Oversight

The CEOs of OpenAI and Anthropic asked the UN Security Council for global AI oversight this week. The US government's answer came from the same room.

Dario Amodei told the Council that if managed poorly, AI "could be a risk to humanity as a whole." Sam Altman said humanity could "lose control of the future of AI" and that the most important decisions cannot be made by labs in San Francisco alone.

Michael Kratsios, speaking for the Trump administration, said the US "totally rejects" efforts by international bodies to assert centralized control and global governance of AI.

So the two most influential labs are asking for rules, and their home government is refusing the venue. That leaves enterprise leaders in an unusual position. The regulatory framework you will operate under is being decided in real time, and no single jurisdiction is writing it.

The most practical detail came from Hugging Face CEO Clement Delangue. His company was attacked by AI agents, and it defended itself with a Chinese model because that model had fewer restrictions than comparable US tools. Whatever gets regulated, the tools will not sit still.

If you buy, build or deploy AI, plan for a patchwork: EU rules, US state rules, and voluntary lab commitments that may or may not hold. Governance built around one regulator's definition of "compliant" will age badly.

Which is more likely to shape your AI governance over the next two years: government regulation, or the standards the labs set for themselves?