Microsoft research: editing fewer than 10 dormant neurons can backdoor LLM agents
Microsoft Research's Lazarus study shows that reprogramming fewer than 10 dormant neurons in an LLM can implant a nearly undetectable backdoor, with the attack autonomously discovered by Claude Code and the risk amplified when models run as tool-enabled agents.
2026-11-12 ~ 2026-11-13 · 2 related posts
- Episode 1: Microsoft's Lazarus: Reprogramming Fewer Than 10 Dormant Neurons Backdoors AI Agents(2026-11-11, 45 posts)
- Episode 2: Microsoft Dormant Neuron Backdoor Research Raises Alarm for Open-Weight Models(2026-11-11, 3 posts)
- Episode 3: Experts Warn of Catastrophic LLM Backdoor Risks(2026-11-11, 2 posts)
- Episode 4: Microsoft research: editing fewer than 10 dormant neurons can backdoor LLM agents(2026-11-12, 2 posts)
- Editing a few weights plants undetectable LLM backdoors — Claude Code found the attack — DimitrisPapail · 2026-11-12
- MSR: Rewriting fewer than 10 dormant neurons backdoors any LLM agent, no training needed — mikeflache · 2026-11-13