The UN's Human Rights Chief Warns AI Could Become an 'Existential' Risk

The quick version

Volker Turk, the United Nations' top human rights official, gave a speech in Geneva warning that advanced AI could grow powerful enough to threaten humanity. He said too few people control it, named four US companies, and announced he plans to contact them directly rather than working only through governments.

What happened

Speaking to the UN Human Rights Council, Turk said: "I share the concerns of industry insiders that advanced AI could pose an existential risk to humanity." He called for "an all-out effort to put cast iron guarantees in place around the safety and security of AI, before it is too late."

His sharper point was about concentration of power. He said just "a handful of men" held "almost unlimited power" over the direction of AI, and singled out Meta, OpenAI, Google, and Anthropic. He said he would reach out to those companies himself in the coming days — an unusual move for a UN human rights office, which normally deals with national governments.

Why it matters

"Existential risk" language has mostly come from AI company leaders and researchers. Hearing it from the UN's human rights machinery signals the debate has moved into mainstream international institutions. The framing also shifts the conversation from "will AI be dangerous" to "who gets to decide" — treating the small number of decision-makers at a few companies as a human rights concern in itself.

What's next

Turk's office says it will approach the companies directly. There is no binding mechanism behind the request, so its impact depends on whether the companies engage and whether member states pick up the theme. It also lands as the EU implements its AI Act and other governments weigh their own rules.