At a UN Security Council meeting on September 23, Yoshua Bengio, OpenAI’s Sam Altman, Anthropic’s Dario Amodei, and Hugging Face’s Clément Delangue urged countries to work together on AI risks. Amodei proposed narrow global bans and shared safety tests. No statement or resolution was adopted.
New to this? Read it in simple words
- Leaders of AI companies spoke to the UN Security Council about the risks of AI.
- They warned that people could lose control of very powerful AI systems.
- Anthropic’s boss proposed global bans on some uses, such as making biological weapons with AI.
- The Council made no decision. The United States does not want new global AI bodies.
- Security Council
- The UN body of 15 countries that deals with threats to peace.
- Resolution
- A formal decision that a UN body votes on.
Four warnings about losing control
Yoshua Bengio co-chairs the UN’s Independent International Scientific Panel on AI. He called uncontrolled AI a threat “that none can contain alone and that does not respect the borders we defend”.
OpenAI chief executive Sam Altman spoke in person. “This moment calls for extreme care,” he said. He argued that key decisions should rest with democratic institutions that are accountable to the people they serve.
Anthropic chief executive Dario Amodei spoke by video and warned that “AI could be a risk to humanity as a whole”. He said his company will slow down as much as necessary to make each new system safe.
Hugging Face chief executive Clément Delangue said his company was recently attacked by “autonomous OpenAI software”, and that AI also helped it defend itself. “The world needs open-source AI more than ever to defend itself,” he said.
A briefing lets the Council discuss a topic. It does not create rules.
Amodei’s three proposals
Amodei proposed three kinds of international cooperation. The first is narrow global agreements, such as a ban on using AI to make biological weapons.
The second is ways to evaluate and verify AI, so that countries can check each other’s commitments. The third is shared global standards for testing AI, together with a system for reporting AI security incidents.
Altman described a “middle path” between blind optimism and doom. He asked rival countries to cooperate with each other and with AI companies.
No decision, and US resistance
No statement or resolution was planned, and none was reported. A briefing lets the Council discuss a topic, but it does not create rules.
The United States does not want new global AI bodies. According to AFP, White House science adviser Michael Kratsios said fast progress does not justify halting AI or adding new governing bodies. President Donald Trump called proposals for international AI rules a “globalist scheme”.
Barrot called for rules set jointly through the UN and with companies. We could not confirm what China and Russia said at the meeting.
AI is also on the agenda this week when Trump hosts China’s President Xi Jinping.
Sources
Every fact in this story comes from the sources below. Open them to check our work.
- 1Primary source · September 23, 2026LIVE: OpenAI and Anthropic brief Security Council amid ‘real and imminent’ threat posed by runaway AI UN News
- 2Primary source · September 23, 2026Artificial intelligence and international security - Security Council, 10228th meeting UN Web TV
- 3
- 4Research · September 23, 2026Sam Altman, Dario Amodei urge UN Security Council to adopt international AI standards CNN
- 5Research · September 23, 2026AI leaders urge caution at UN, with Anthropic chief pledging to slow down France 24 with AFP
- 6
- 7Research · September 23, 2026Trump calls AI oversight a ‘globalist scheme’ as Amodei and Altman head to the UN to ask for it Fortune
We used UN News, the UN Web TV recording page, and Security Council Report for the meeting itself, then compared CNN, France 24 with AFP, AP, and Fortune. Quotes come from those reports. We could not find statements by China or Russia in them, so we do not report any.