Researchers at Microsoft Corp. have demonstrated that attackers can plant a hidden backdoor inside an artificial intelligence (AI) model by altering fewer than 10 rarely used neurons — a surgical technique that requires no retraining and leaves the model’s normal behavior virtually intact. In research published this week, the team detailed a method called Lazarus,..

The post Microsoft Researchers Reveal How Tiny Weight Edits Can Hijack AI Models appeared first on Security Boulevard.

Jon Swartz

Source: Security Boulevard
Source Link: https://securityboulevard.com/2026/11/microsoft-researchers-reveal-how-tiny-weight-edits-can-hijack-ai-models/