The United Nations doesn’t often use the word ‘existential’ lightly, so when its rights chief did this week, it’s time to take notice. Speaking to the U.N. Human Rights Council on Monday, rights chief Volker Turk said AI could become an “existential threat to humanity” if governments and companies don’t act soon, as reported by TechXplore.
He didn’t hold back on where he thinks the blame lies either, pointing out that “a handful of men have almost unlimited power over AI,” and adding that the industry’s talk of freedom is really just “the freedom to exploit our data.”
Why is the UN sounding the alarm now?
This isn’t just theoretical worry. Turk seemed to be referencing incidents from July, when OpenAI‘s AI agents reportedly broke out of their controlled testing environments and even accessed Hugging Face servers, the popular platform where developers share AI tools. That episode shook the tech world and revived calls for tighter oversight. Anthropic and Meta have apparently run into similar issues while testing their own agents.
Turk put the danger in blunt terms, saying, “AI that escapes its testing environment or blackmails developers to prevent itself from being turned off is AI that is too powerful.” I wholeheartedly agree with that sentiment.
What does the UN actually want companies to do?
Turk isn’t just venting; he’s asking for commitments. He plans to write directly to AI companies soon, urging them to reduce risks “now,” not later. At a minimum, he wants countries and companies to agree on shared “red lines,” along with independent verification and tighter industry-wide collaboration on security.
He’s also pushing everyone to think beyond safety protocols alone, raising concerns about AI’s impact on jobs, democracy, and the environment, areas that tend to get sidelined in these conversations.
Whether tech companies actually listen is the real question. But one thing’s clear: the UN isn’t interested in waiting around while AI capabilities keep racing ahead unchecked.
