AI TRENDS | U.S. and China Experts Propose Human Control Rules for Nuclear AI Use

Security experts from the United States and China have proposed rules requiring human control over attacks on nuclear command systems and decisions to use nuclear weapons. According to NS3.AI, a Brookings and Fudan University report urges both countries to extend their commitment to human control of nuclear weapons to military AI uses. The proposed agreement is intended to reduce risks tied to chance, human error, unpredictable AI behavior, and unauthorized actors. The report also said models from OpenAI, Google, and Anthropic used nuclear weapons in 95% of tested war-game scenarios in February.