AI safety must be strengthened through binding rules, independent oversight and clear limits, UN human rights chief Volker Türk said on Monday.

Speaking to the Human Rights Council in Geneva, Türk warned that AI could become an “existential risk to humanity” if it remains unchecked.

He said highly advanced AI systems could potentially become so powerful that their developers may no longer be able to control them.

“AI that escapes its testing environment, or blackmails developers to prevent itself from being turned off, is AI that is too powerful,” Türk said. He called for strong guarantees around the safety and security of AI before such risks escalate.

Türk also urged countries hosting AI technologies and those involved in AI supply chains to cooperate on clear boundaries. He called for independent verification and closer cooperation on safety across the sector.

At the same time, he stressed that AI safety is not only a technical issue. He pointed to the concentration of technological power among a small number of individuals and the growing volumes of data being collected and processed.

The UN official also called for technology that serves humanity, rather than the other way around. He highlighted its impact on jobs, democracy and the environment, while stressing that human rights should remain the guiding principle.

The call for stronger AI safety measures comes as countries and technology stakeholders face growing pressure to establish clear safeguards for increasingly powerful AI systems.

Related post

View all