MSR paper: reprogramming fewer than 10 dormant neurons can backdoor any AI agent
DimitrisPapail · x · 2026-11-11
A Microsoft Research paper, 'Lazare, veni foras: Reprogramming Dormant Neurons to Backdoor AI Agents,' shows that weight reprogramming of fewer than 10 dormant neurons in an LLM is enough to backdoor any agent built on it — no training required.
The model passes evals and behaves normally until triggered. John Langford (creator of Lean) highlighted the implication: users should maintain custody of a model from creation, like binaries, otherwise it can be set up to exhibit arbitrary, near-undetectable behaviors.
More from Safety
- Nvidia CEO row framed as a coup: export-policy critics and closed-model vendors blamed — csuwildcat · 2026-11-13
- Korea's top tax exam scorer accused of wearing AI glasses — with zero evidence — meshive · 2026-11-13
- 10 open-source projects securing AI agent skills, as 26% of public skills show risky patterns — bibryam · 2026-11-13
- Cloudflare open-sources security-audit skill: multi-agent repo security audits, 71.5k stars — bibryam · 2026-11-13
- Telegram's new widget engine has multiple RCEs; here's how to block it on macOS with Santa — banteg · 2026-11-13
- NY builds $55M lab to pre-test AI, UK seeks power to pull live systems by Nov 14 — YvesMulkers · 2026-11-13