U.S., China security experts propose nuclear-style safeguards for AI risks

U.S. and Chinese security experts participating in a U.S.-China dialogue are urging Washington and Beijing to adopt nuclear-style safeguards for artificial intelligence, warning that an AI system interfering with a nuclear command network or launching a military cyber operation could leave either government with only minutes to decide whether it had been attacked. The experts' proposals include red lines around nuclear systems, human control over consequential cyberattacks, and a hotline for incidents involving autonomous AI. The dialogue's work helps inform official discussions between the two governments.