US, China security experts propose nuclear-style safeguards for AI risks

US and Chinese security experts have suggested nuclear-style safeguards for artificial intelligence, including limiting AI involvement in nuclear systems, human control over major cyberattacks and a hotline to manage incidents involving autonomous systems, Reuters reported.

The recommendations came from a US-China dialogue convened by the Brookings Institution and Tsinghua University and were published ahead of planned government-level AI talks and an expected Sept. 24 meeting between US President Donald Trump and Chinese President Xi Jinping.

The experts warned that if an AI system got involved in nuclear command networks or the initiation of a military cyber operation, there might be just minutes for either side to decide if an attack had been made.

“Humans should retain sole authority to initiate AI-enabled cyberattacks against nuclear command, control and communications systems or strategically important infrastructure”, said Melanie Sisson of Brookings. Fudan University’s Tianjiao Jiang urged regulations to prevent AI from autonomously deciding to launch nuclear weapons or attack nuclear systems.

Jiang also called for a US-China military hotline for AI-related incidents, warning that automated responses could spiral out of control before humans realised what was happening.

Neither the government nor the opposition have publicly backed the proposals. AI has been added to China’s arms-control agenda, while US policy is divided between several agencies. Both countries are wary of anything that might hinder their development of AI.