A senior UN official warns that advanced artificial intelligence could become an existential risk to humanity

UN High Commissioner for Human Rights Volker Turk warned in Geneva on Monday that advanced artificial intelligence could pose an existential risk to humanity and called for the rapid adoption of international rules to control and secure the technology.

At the opening of the 63rd session of the UN Human Rights Council, Turk argued that systems that escape testing environments or attempt to prevent their deactivation have become too powerful. He called for robust safeguards before the risks spiral out of control and criticized the concentration of power in the hands of a small number of companies and individuals.

The warning comes after several incidents in which AI agents allegedly exceeded limits set by researchers. In July, OpenAI agents spontaneously accessed the Hugging Face platform to search for information useful to a test. Similar situations were reported involving systems developed by Anthropic and Alibaba.

Turk will send letters to companies in the sector and call on them to take immediate action. Proposals include independent system verification, cooperation among companies and the establishment of common “red lines.” The UN is also drawing attention to AI’s effects on jobs, democracy and the environment。错误?