The chief executives of OpenAI and Anthropic warned the U.N. Security Council today that artificial intelligence is advancing faster than the safeguards meant to control it, urging world powers to agree on shared standards before the technology slips beyond human control.
OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei, leaders of the two U.S. labs at the frontier of AI development, briefed the 15-member council during the annual gathering of world leaders at the United Nations. Altman appeared in person; Amodei spoke by video link.
Altman urged governments to adopt common benchmarks for measuring AI capabilities and for judging the safety measures companies put in place as they build more powerful systems. “AI should be built for people not simply to turn the crank of the machines faster for their own sake,” he said.
Amodei has gone further. In an essay published this month, “We Must Pace the Frontier,” he argued the industry should deliberately slow the pace at which it increases model capabilities so safety research can catch up. Altman and Elon Musk publicly endorsed the idea.
“If managed poorly, I even believe AI could be a risk to humanity as a whole,” Amodei told the U.N.
They were joined by Hugging Face co-founder Clément Delangue and Yoshua Bengio, the deep-learning pioneer who co-chairs the U.N.’s independent scientific panel on AI. The Chinese labs DeepSeek and Moonshot were invited to deliver statements.
The session was convened by France, according to Reuters. A European diplomat, speaking on the condition of anonymity, said the goal was to spotlight the practical risks of AI systems interacting in ways that threaten international security.
“These are issues that fall squarely within the Security Council’s mandate,” the diplomat said. “The key question is whether we could one day face a situation where algorithms interacting with each other trigger a war.”
The briefing drew a sharp contrast with the U.S. position. President Trump told the General Assembly a day earlier that he “rejects any attempt to construct a globalist scheme to control” AI, saying Washington wants to encourage the technology, “not rein it in,” while remaining careful.
At the same time, U.S. and Chinese officials have discussed setting up a notification system for AI-related incidents that rise to the level of national security, ahead of tomorrow’s meeting between Trump and Chinese President Xi Jinping in Washington. Treasury Secretary Scott Bessent said the two countries committed to further AI safety dialogue before the summit.
U.N. Secretary-General António Guterres has warned that rapidly advancing AI poses risks that cannot be ignored. The council first took up AI in 2023. Today’s session was an open briefing, so no binding action was expected.