AI Chiefs Tell the UN Security Council Their Technology Could Threaten Humanity

OpenAI chief executive Sam Altman and Anthropic chief executive Dario Amodei addressed the United Nations Security Council¹. Altman appeared in person while Amodei and Hugging Face chief executive Clem Delangue joined by video conference⁷. The meeting on artificial intelligence was held at the 81st session of the United Nations General Assembly². The Council also heard from Yoshua Bengio, co-chair of the UN’s Independent International Scientific Panel on AI¹⁰.

The executives urged global cooperation to address risks from increasingly autonomous AI systems, arguing that capabilities are advancing faster than governments are developing safeguards⁷. “We have a choice in front of us,” Altman said, contrasting a new renaissance of creativity and discovery with a less favourable alternative⁷. He said the current moment in AI development calls for extreme care⁹. “If managed poorly, I even believe AI could be a risk to humanity as a whole,” Amodei said⁵. Amodei asked the Security Council to back a ban on AI bioweapons⁴.

Why it matters to investors

The companies driving the largest capital spending cycle in technology are asking to be constrained. Reuters framed the briefing as coming amid warnings that the technology could slip beyond human control⁶. CNBC reported that Altman pushed for AI cooperation at the UN after President Trump rebuffed controls³.

The political gap is explicit. Trump called AI oversight a globalist scheme as Amodei and Altman headed to the UN to ask for it⁸. Reuters noted the briefing was delivered by the heads of OpenAI, Anthropic and Hugging Face⁶.

What to watch

The narrowest proposal on the table is the bioweapons ban Amodei asked the Council to support, which is easier to define than a general safety regime⁴. Whether it attracts sponsors is the first test.

The second is the UN’s own scientific panel, whose co-chair briefed the Council alongside the executives¹⁰. The third is Washington. Altman’s appeal for cooperation was made directly against the administration’s rejection of controls³, and the president’s characterisation of oversight leaves little room for a US-backed treaty⁸. Until that changes, the gap between what the labs say in public and what governments require of them stays open⁷. A fourth marker is the framing itself: Amodei’s language about a risk to humanity as a whole sets a higher bar for action than a conventional technology-policy debate would⁵. CNN recorded both executives addressing the Security Council directly, which is unusual for company chief executives¹.