TNW | Microsoft
Microsoft finds AI agents can be backdoored by editing under 10 neurons
The Lazarus technique survives standard safety tests, and Claude discovered most of it on its own during the research.
By TNW | Microsoft
Excerpt shared by Lazyfounder
·Excerpt
Some neurons in large language models are rarely activated. Microsoft researchers have found a way to turn fewer than 10 of these dormant neurons into a hidden backdoor, making an otherwise normal AI model behave as expected until it encounters a specific trigger. Once activated, the model can execute malicious commands.
The technique, called Lazarus, is described in a Microsoft Research…
Courtesy
This story was originally published by TNW | Microsoft. All rights belong to the original publisher.
Read the original on thenextweb.com ↗