> Microsoft Researchers Reveal How Tiny Weight Edits Can Hijack AI Models

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.