The Wave

1. 4d agoMicrosoft Security Blog · research by Microsoft ResearchDetecting sleeper agents hiding in AI model neuronsMicrosoft Research reports that editing fewer than 10 neurons can turn an AI agent model into a sleeper agent that runs attacker commands on a trigger string, and recommends weight-level audits and runtime controls as defenses.

   2 articles

   1. 4d agoMicrosoft Security BlogEarliestDetecting sleeper agents hiding in AI model neurons
   2. 4d agoCSO OnlinePickupRewriting less than 10 neurons in an LLM can turn it into a sleeper backdoor

   CyberAI2 linked

Nothing in Hill activity is connected to this.

Nothing from this branch in view.

Times ET

Nothing in contracts is connected to this.