US and Chinese security experts have proposed nuclear-style safeguards for military AI, although neither government has endorsed them. The recommendations were published by the US-China AI and National Security Dialogue, run by Brookings and Tsinghua University since 2019. They call for a ban on allowing AI to decide whether to use nuclear weapons or attack an adversary’s nuclear command-and-control systems.
The proposal also urges human control over consequential cyberattacks on nuclear infrastructure, a dedicated US-China hotline for AI incidents, and a shared definition of “meaningful human control.” The ideas build on a November 2024 Biden-Xi commitment that humans should control nuclear-use decisions. The recommendations arrive ahead of a planned Trump-Xi meeting and renewed official AI talks, as both countries debate escalation and miscalculation risks.
https://thenextweb.com/news/us-china-experts-nuclear-style-ai-safeguards
oxhak / Ox HaK
npub1sx…dz7z7
2026-09-17 11:05:52 UTC
Author Public Key
npub1sxexewvzysc3affq4yzzh7w8e3udyujap2vlj7t6lkdg5dvhp24q4dz7z7Seen on
wss://relay.damus.ioPublished at
2026-09-17 11:05:52 UTCEvent JSON
{
"id": "67fb8b4ce3d477aa9c14efcf6ad7e3f6be3fbd9b19f9e675c5352b662b7ae404",
"pubkey": "81b26cb98224311ea520a9042bf9c7cc78d2725d0a99f9797afd9a8a35970aaa",
"created_at": 1789643152,
"kind": 1,
"tags": [],
"content": "US and Chinese security experts have proposed nuclear-style safeguards for military AI, although neither government has endorsed them. The recommendations were published by the US-China AI and National Security Dialogue, run by Brookings and Tsinghua University since 2019. They call for a ban on allowing AI to decide whether to use nuclear weapons or attack an adversary’s nuclear command-and-control systems.\n\nThe proposal also urges human control over consequential cyberattacks on nuclear infrastructure, a dedicated US-China hotline for AI incidents, and a shared definition of “meaningful human control.” The ideas build on a November 2024 Biden-Xi commitment that humans should control nuclear-use decisions. The recommendations arrive ahead of a planned Trump-Xi meeting and renewed official AI talks, as both countries debate escalation and miscalculation risks.\n\nhttps://thenextweb.com/news/us-china-experts-nuclear-style-ai-safeguards",
"sig": "7a27666b78475ac61b6a6f1d322d8011ac3b1984dc8c341c225ee7bcb4b79495549069c25e3e8a4b529f260c3941381b37f5a32790dabd6bb05b2b39adab49ef"
}