UN Rights Chief Calls for Mandatory AI Safeguards, Not Reliance on Goodwill

AI companies’ voluntary safety commitments are not enough, UN High Commissioner for Human Rights Volker Türk warned on October 5, 2026. Speaking at the AI and Human Rights Summit in Oxford, he called for mandatory safeguards as competition accelerates. His appeal follows a White House accord that carries no penalties for companies that ignore it.

Türk acknowledged AI’s potential to advance science, health, education and human-rights work. But he warned that control of the technology rests with a small group, without public oversight, while its creators cannot fully predict or manage its behavior. He described competition among companies and countries as a “ruthless race.”

His warning extended beyond a possible future loss of control. He said AI could erode equality, freedom, privacy and public debate in less visible ways. He also warned that entire populations could lose access to communications, water, heating and financial services, and urged a prohibition on machines deciding who lives or dies.

Governments would carry the accountability duty

Türk’s proposed response centers on human-rights due diligence: assessing and addressing risks to people’s rights as AI is developed and used. He called for those safeguards to be compulsory and adjusted as the technology evolves. These were policy recommendations in his summit address, not an announcement that new requirements had taken effect.

Countries hosting AI companies, or able to influence their supply chains, should require responsible conduct and hold companies accountable when they fall short, he said. That assigns a role to governments with leverage over the industry, rather than leaving responsibility solely with developers.

He did not argue that existing law could answer every problem. Some areas would need new legal approaches, while updates to liability, cybersecurity and data-protection rules could provide protection in the meantime. He also sought stronger security cooperation, independent industry monitoring and more input from civil society and people whose rights are affected.

We can no longer afford to wait or rely on goodwill

Volker Türk, UN High Commissioner for Human Rights

Momentum, but no penalty for ignoring the accord

Türk welcomed the UN’s Global Dialogue on AI Governance and Independent International Scientific Panel on AI. He also endorsed a call for tighter controls on frontier models—the most advanced AI systems—backed by 29 states and the European Union.

He acknowledged recent industry steps on safeguards, self-regulation and risk assessment. UN News interpreted that as an apparent reference to the September 29 White House accord signed by President Donald Trump and six AI company heads. Trump described the voluntary agreement as “morally binding”; it imposes no penalties for noncompliance.