United Nations Secretary-General António Guterres on Monday urged the creation of a global governance system to regulate artificial intelligence, warning that the technology is advancing at “runaway speed” and without sufficient oversight.

Speaking ahead of the first Global Dialogue on AI Governance in Geneva, which brings together governments, technology companies, academia and civil society, Guterres said AI is already transforming societies, but cautioned that the world is witnessing “an experiment being run on our own societies, without a plan and without consent.”

He stressed that the key question facing humanity is whether AI will be shaped collectively or allowed to evolve unchecked.

Guterres warned that modern AI systems are increasingly operating beyond their original function as tools.

“They are writing code, acting online and making choices with less and less human oversight,” he said, adding that existing institutions were designed for machines that follow instructions, not systems capable of independent decision-making.

He also raised concerns over the growing difficulty of distinguishing between truth and falsehood in the digital space, noting that AI is accelerating misinformation and over-reliance on automated outputs.

While acknowledging that so-called “vibe-coding”, using AI systems to generate solutions based on natural language prompts, can be useful, Guterres warned against overdependence on such tools.

“Vibe-coding can do wonders,” he said, “but we cannot vibe-code the truth. We cannot vibe-code the future of humanity.”

The UN chief highlighted concerns that AI development is increasingly concentrated in a small number of companies and countries, leaving much of the world without meaningful input into its direction.

He warned that this imbalance risks deepening global inequality and widening the digital divide unless addressed through coordinated international action.

Guterres also pointed to the potential benefits of AI in healthcare, education and development, but stressed that these gains must be matched with safeguards rooted in human rights, safety standards and accountability.

He urged the establishment of common global methods to assess and verify AI risks, particularly in systems accessible to children.

Comparing AI regulation to safety standards in medicine and toys, he said children are already interacting with AI in sensitive areas such as learning and personal communication without adequate protections.

Guterres called for an AI Child Safety Pledge, requiring companies to ensure systems accessible to minors are safe, with zero tolerance for sexual exploitation and mechanisms to connect vulnerable users with real human support.

“No child should be a guinea pig for unregulated AI,” he said.

Guterres expressed particular concern over the use of AI in military applications, especially lethal autonomous weapons systems.

Describing them as “killer robots,” he warned against machines that can select and engage targets without human judgment or control, calling such systems “morally repugnant” and urging an international legal ban.

The UN chief also proposed the creation of a Global AI Fund to help developing countries build technical capacity, access computing power and develop data infrastructure, warning that otherwise the global digital divide could deepen into an “AI divide.”

He further called on technology companies to disclose the environmental impact of AI systems and to commit to powering data centres with renewable energy by 2030.

Guterres concluded by stressing urgency in establishing global guardrails for AI development, warning that the window to act is limited.

“We may be the last generation able to set the terms on which humanity and machines coexist,” he said. “The door is still open. It will not stay open long.”