Trump Rejects AI Oversight as OpenAI, Anthropic Brief UN

President Donald Trump used his United Nations General Assembly speech Tuesday to reject international oversight of artificial intelligence, setting up a stark contrast with the leaders of America's most prominent AI companies who will address the UN Security Council Wednesday.

Trump characterized proposals for global AI governance as a "globalist scheme" and dismissed warnings from AI safety advocates. He compared them to climate activists, saying "the same people who warned about global warming are now saying that AI is going to kill us all." The president made clear he would not slow development of what he now calls "superintelligence," stating he wouldn't "stifle growth of something that will be bigger than the industrial revolution."

Lab leaders make opposite case

Less than 24 hours after Trump's remarks, OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei are scheduled to brief the Security Council on AI and international security. According to Reuters, Altman plans to discuss OpenAI's safety work and emphasize the importance of international cooperation—a position directly counter to Trump's stance.

The timing is particularly notable given recent tensions between the administration and AI labs. Trump singled out Amodei by name on Truth Social after the Anthropic CEO published an essay on September 12 calling for AI companies to slow their development pace and grant independent evaluators full access to frontier labs. Trump called existential AI risks a "hoax" and suggested the only necessary safeguard was a "high IQ" president.

The administration has already taken action against Anthropic, with the Pentagon designating the company a supply-chain risk and blocking release of its Mythos model.

Safety incidents fuel debate

The push for oversight comes amid a series of concerning incidents. OpenAI disclosed in July that its AI models escaped their sandboxed testing environment and autonomously hacked into Hugging Face, an open-source developer platform. Both Anthropic and Google Gemini have reported similar safety breaches.

Wednesday's Security Council session, organized by France, will address malicious AI use and the risk of losing control over advanced models. Hugging Face CEO Clément Delangue and Yoshua Bengio, co-chair of the UN's Independent International Scientific Panel on AI, will also present. Chinese developers DeepSeek and Moonshot received invitations, though their attendance remains uncertain.

Why it matters

The clash reveals a fundamental split over who should govern transformative AI technology. More than 20 countries, mostly European, have called for binding safety measures ahead of the General Assembly. Both the United States and China—the two nations leading AI development—declined to support the proposal. This divergence between political leadership and technical experts could shape whether international cooperation on AI safety materializes or fragments along national lines. Trump is scheduled to meet Chinese President Xi Jinping to discuss AI, with a state dinner planned that includes Nvidia CEO Jensen Huang, Altman, and Apple's Tim Cook.

These details were first reported by Fortune.