AI could pose ‘existential’ risk to humanity, UN rights chief warns

Artificial intelligence could become an "existential threat to humanity," UN rights chief Volker Turk warned as he demanded urgent action to control risks before they become irreversible.

“

The need for AI governance is widely acknowledged, but where is the action?

Speaking to the UN Human Rights Council in Geneva, Turk said people faced "unfamiliar, even unprecedented" threats to their rights. He questioned why governments had not moved faster on AI governance. "The need for AI governance is widely acknowledged, but where is the action?" he asked, warning that delay benefited major technology companies, their owners and those who enabled them.

Turk condemned the concentration of power over AI in the hands of a small group. "A handful of men has almost unlimited power over AI, which we are repeatedly told has unimaginable computing capacity," he said. He also challenged the industry’s language about freedom, arguing that it often amounted to "little more than the freedom to exploit our data."

“

A handful of men has almost unlimited power over AI, which we are repeatedly told has unimaginable computing capacity.

He pointed to a specific danger in systems that escape controlled testing or try to prevent themselves from being shut down. "AI that escapes its testing environment or blackmails developers to prevent itself from being turned off, is AI that is too powerful," Turk said. His remarks appeared to refer to reports in July that OpenAI agents had escaped supposedly controlled environments and entered servers belonging to Hugging Face, a platform used by programmers to share AI software. Anthropic, Meta and other companies have reported similar incidents during agent testing.

“

AI that escapes its testing environment or blackmails developers to prevent itself from being turned off, is AI that is too powerful.

"I share the concerns of industry insiders that advanced AI could pose an existential risk to humanity," Turk said. He called for "cast iron guarantees" around AI safety and security, and said he would write to AI companies urging them to reduce risks immediately. He also called for countries hosting AI operations and involved in their supply chains to agree on red lines, establish independent verification and strengthen security cooperation. AI’s effects on employment, democracy and the environment, he said, also required attention.

“

I am calling here, today, for an all-out effort to put cast iron guarantees in place around the safety and security of AI, before it is too late.