SecurityNov 11, 2026

Detecting sleeper agents hiding in AI model neurons

Microsoft Research has released new threat research detailing how a capable agent model could be turned into a sleeper agent by editing fewer than ten neurons to inject a short payload; Microsoft Research has also found that this threat is detectable by frontier agents, which can identify anomalous weight edits. The p…

Published by Microsoft Security BlogOpen original story

Magazine Core indexes and summarizes the source. The complete article remains with its original publisher.

Keep reading

More from Security