AI Could Pose ‘Existential’ Risk to Humanity, UN Rights Chief Warns

GENEVA — The United Nations’ top human rights official issued a stark warning on Monday, declaring that artificial intelligence could pose an existential risk to humanity unless governments and technology companies take immediate, coordinated action to impose robust safeguards.

Volker Türk, the U.N. High Commissioner for Human Rights, said the breakneck pace of AI development is outpacing regulatory frameworks, leaving fundamental rights, democratic processes, and even human survival vulnerable to catastrophic misuse.

“The rapid advancement of artificial intelligence without adequate oversight threatens to undermine the very fabric of our societies. We face an existential risk if we do not act now to ensure that these technologies serve humanity, rather than the other way around,” Türk said in a statement from the U.N. Human Rights Office in Geneva.

Türk’s remarks, delivered during a briefing on emerging technology risks, mark one of the most forceful warnings from a senior U.N. official about the perils of AI. He pledged to push for a binding international treaty that would enshrine human rights protections at the core of AI governance, and called on states to immediately halt the deployment of high-risk systems that lack adequate oversight.

A Growing Chorus of Concern

The U.N. rights chief’s intervention comes amid a wave of urgent calls from global institutions, researchers, and civil society groups for stronger guardrails on artificial intelligence. Earlier this year, the U.N. Secretary-General called for a global digital compact, while the U.N. Educational, Scientific and Cultural Organization (UNESCO) adopted the first-ever global agreement on AI ethics, emphasizing transparency, accountability, and human dignity.

The U.N. Human Rights Office has increasingly focused on how AI systems can amplify discrimination, erode privacy, and enable mass surveillance. Automated decision-making tools used in hiring, law enforcement, and social welfare have already been shown to perpetuate bias, and Türk warned that next-generation systems could pose far greater dangers if embedded in critical infrastructure, military command chains, or the global financial system.

“We are not talking about a distant, hypothetical future. The risks are already materializing in ways that disproportionately harm the most vulnerable,” Türk said. “An existential threat does not necessarily mean the annihilation of the human species; it can also mean the collapse of the rights and freedoms that make life dignified and meaningful.”

Specific Steps Urged

During the briefing, Türk outlined a series of measures the international community should adopt:

• An immediate moratorium on the use of AI in contexts that could jeopardize life, liberty, or physical safety, pending rigorous human-rights impact assessments.
• Mandatory transparency requirements for all AI systems used by governments and large corporations, including the disclosure of training data and algorithmic decision-making logic.
• Establishment of an independent international oversight body, modeled on the International Atomic Energy Agency, to monitor compliance and investigate abuses.
• A legally binding treaty that prohibits the development of autonomous weapons systems capable of taking lethal action without meaningful human control.

He also stressed that tech companies must stop treating AI safety as a secondary concern and instead embed human-rights by design principles from the earliest stages of development. “Voluntary commitments are not enough,” he said. “We need binding obligations, enforced by law, to hold the most powerful actors accountable.”

Wider Institutional Push

Türk’s warning aligns with a broader diplomatic push. The U.N. Human Rights Office has been working closely with the UNESCO’s AI ethics framework and the Secretary-General’s High-Level Advisory Body on AI, which is expected to release a report later this year recommending a global governance architecture. Last month, the G7 leaders agreed on a set of “Hiroshima AI process” principles, but many advocates argue that only a U.N.-backed treaty can provide the universality and enforcement power needed.

The U.N. Human Rights Office noted that Türk’s call for a treaty is rooted in existing international human rights law, which already obligates states to protect individuals from foreseeable harms. The challenge, officials say, is translating those obligations into concrete technical standards and enforcement mechanisms for a technology that evolves faster than legislatures can act.

“The gap between AI capabilities and our ability to govern them grows wider every day,” Türk said. “We cannot afford to wait until the worst outcomes become reality. We need a global response that matches the global scale of the threat.”

While the U.N. rights chief’s warnings often carry moral weight, they do not have the force of law. However, Monday’s statement is expected to intensify pressure on the U.N. General Assembly and the Security Council to take up the issue during the upcoming session.

“History will judge us by whether we had the foresight to act before it was too late,” Türk concluded.