The head of the United Nations human rights office said on Monday that the world is running out of time to set rules for artificial intelligence. Volker Türk warned that a "ruthless race" between companies and countries is raising concerns about human safety and even survival.
Türk spoke at an AI and Human Rights Summit in Oxford, in the United Kingdom. He said AI could bring major advances in science, health, education and even human rights work.
However, he said the tools are being built and controlled by a small group without public oversight. He added that AI systems are already doing things their creators cannot fully predict or manage.
Warnings About Rights and Warfare
Türk said AI could weaken equality, freedom, privacy and public debate in quiet ways. He described this as "letting dystopia creep in unnoticed."
He warned that whole populations could be cut off from communication, clean water, heating and financial services "at the click of a button."
He also called for urgent action to stop machines from deciding who lives and who dies. He noted that AI is already changing how wars are fought.
Türk pointed to studies showing that students who rely on AI remember less and think less on their own. He asked whether the world is heading toward a future where no one can understand, question or guide the technology.
Calls for Mandatory Safeguards
Türk called for mandatory human rights safeguards and checks on how AI is built and used. He said these rules should change as the technology changes.
He said countries that host AI companies, or have influence over their supply chains, must require them to act responsibly. Those countries should also hold companies to account when they do not, he added.
Türk said existing law does not have every answer, and some areas will need new legal approaches. In the meantime, updated rules on liability, cybersecurity and data protection could offer protection.
He urged closer teamwork on security and independent monitoring within the industry, "not in words, but in practice." He also asked for a bigger voice for civil society and people whose rights are affected.
"We can no longer afford to wait or rely on goodwill," he said.
Türk welcomed two UN efforts, the Global Dialogue on AI Governance and the Independent International Scientific Panel on AI. He also welcomed a joint call for tighter control of frontier AI models, backed by 29 countries and the European Union.
He said AI industry leaders took new steps last week on safeguards, self-regulation and risk assessment. This appeared to refer to a voluntary accord signed at the White House on September 29 by President Donald Trump and the heads of six AI companies.
The accord carries no penalties for companies that ignore it. President Trump has described it as "morally binding."
"Too few are walking the talk," Türk said. Three weeks ago, he wrote to countries and frontier AI companies urging them to act now to reduce risks.
As Türk spoke in Oxford, the New York City Council held a rare Committee of the Whole to examine AI risks to New Yorkers. The council is considering 10 bills to regulate AI in the nation's largest city, with testimony from Anthropic, OpenAI, Google and Meta.
