Rewriting less than 10 neurons in an LLM can turn it into a sleeper backdoor

Microsoft researchers have found that rewriting fewer than 10 rarely used neurons in an open-weight AI model can turn any agent built on it into a sleeper backdoor that runs an attacker’s commands on cue, with no retraining required. Researchers from Microsoft Research and Microsoft Red Team described the technique called Lazarus in a research paper. It modifies dormant neurons in a model’s final feed-forward layer t…

This page shows only the headline and the short summary. The full text is published by the original source.