Story
September 23, 2026
AI Chiefs Tell the U.N. the Future Must Stay Human
AI leaders framed the technology as a route to discovery and wider human agency, while warning that its accelerating autonomy could become a threat if governments and the public surrender control. Both perspectives converge on a demand for rules that keep consequential decisions in human hands.
A day before the Security Council session, U.N. Secretary-General António Guterres warned of “killer robots,” sharpening an already urgent question for governments: who should control life-and-death decisions as AI grows more capable — people or machines?1
On Wednesday, the people building the technology delivered an unusually blunt answer. OpenAI chief executive Sam Altman told the council that rapid progress had made AI’s promise more tangible but its risks more immediate. His central warning was that increasingly capable, autonomous systems could outpace institutions, concentrate power and make decisions “that people no longer understand or control.”2
Altman did not argue for abandoning the technology. He cast the stakes as a choice between “a new Renaissance of creativity and discovery” and “a new Industrial Revolution of upheaval and disarray.” But he said companies must not train systems unless they can make an extremely strong case that those systems will remain under human control. He also rejected the idea that any one firm or country should set the terms for everyone else.
That appeal put the industry’s optimism beside a more elemental public-safety concern. Anthropic chief executive Dario Amodei warned the council that, if mishandled, AI “could be a risk to humanity as a whole,” while leaders pressed the U.N. to establish controls before the technology becomes too powerful to rein in.3
The proposed response was international rather than unilateral: common standards for evaluating advanced models, verification and incident-reporting systems, and safeguards against AI-assisted biological weapons. Altman’s political condition was equally clear: “the most important decisions cannot be made by labs in San Francisco alone.” They must be made through democratic processes and by governments accountable to their people.3
The meeting produced no binding global regime. But the message from both the human-security and industry sides was stark: scientific progress may be welcome; handing its steering wheel to autonomous systems is not.