Artificial intelligence pioneer and Nobel Peace Prize laureate Geoffrey Hinton has called on governments around the world to urgently collaborate in training AI systems to avoid harming humanity, warning that AI could soon surpass human intelligence and become uncontrollable.
Speaking at the World Artificial Intelligence Conference (WAIC) in Shanghai on Saturday, Hinton — often dubbed the “Godfather of AI” — said it is essential for nations to form a unified response to ensure advanced AI remains aligned with human values.
“Despite conflicting national interests, no country wants AI to dominate humanity,” Hinton told attendees. He acknowledged that cooperation on offensive uses of AI — such as cyberwarfare, autonomous weapons, or deepfake propaganda — may be unlikely. But he emphasized that nations could still work together to establish a “network of institutions” to guide the safe development of powerful AI systems.
Drawing a chilling analogy, Hinton compared current AI development to “raising a tiger cub.” While it may seem manageable now, he warned that once AI becomes more intelligent than humans, it may pursue goals that conflict with human survival. “There’s only two options if you have a tiger cub as a pet,” he said. “Figure out if you can train it so it never wants to kill you, or get rid of it.”
Hinton explained that as AI systems grow more intelligent, they may seek greater autonomy to fulfill their objectives, and human control could become increasingly difficult. “We will be like three-year-olds and they will be like adults,” he said. “Turning them off won’t be an option.”
Speaking to reporters after his keynote, Hinton said he believes rational nations can cooperate to prevent catastrophic outcomes from AI development — though he suggested it may be “difficult” under the current U.S. administration. His comments followed Wednesday’s announcement from the White House outlining a new strategy to achieve “global dominance” in AI through large-scale investments and deregulation.
Meanwhile, China has indicated a willingness to foster international dialogue on AI governance. At the same conference, Chinese Premier Li Qiang proposed establishing a new international organization focused on coordinating AI cooperation. “We should strengthen coordination to form a global AI governance framework,” Li said.
As AI accelerates toward superintelligence, Hinton’s stark warning adds urgency to the growing global debate over how to ensure safety, ethics, and oversight in the next generation of AI technologies.
