Bill Gates says AI has already crossed the danger thresholds he spent years tracking, and he is stunned by how little society seems to care. In a new interview with MIT Technology Review, the Microsoft co-founder said the field has passed the bio-capability, cyber-capability, psychosocial, and job-market thresholds, with hints that even the control threshold may be in sight.
Gates said he is in a state of shock that the field has crossed these thresholds, and voiced surprise that so little concern has surfaced outside the industry. He argued that past technology revolutions are misleading guides, since AI that can replace cognition for a large share of jobs at modest cost marks a different kind of shift.
Gates said data center protests miss the point, since stopping US builds would not change the underlying issues. He called for monitoring any model capable of making novel molecules and backed a US-China agreement on the same rules. In his telling, a deliberate biological attack is about 50 times likelier than a naturally occurring pandemic.
He also pressed for more AI expertise inside government and warned against the issue becoming partisan. The Gates Foundation continues to fund AI for health, including Biomni at Stanford, while a spun-out group called NextLadder targets government-services navigation for low-income families.
Asked whether he hopes readers feel some alarm, Gates said he had not expected to be among the loudest voices warning that wider society is not paying attention.
