Researchers demonstrate Lazarus technique that hides backdoors in AI model neurons
AI SECURITYMicrosoftredteam/daily · 1 source
- Severity
- Medium
- Sources
- 1
What happened
Researchers showed that a small change to a few rarely used parts of an AI model can secretly make it issue harmful commands when it sees a specific trigger word. The model behaves normally in everyday tests, so the change is hard to spot. Microsoft says no criminals have been seen using this method so far.
Like I’m new · Written by AI from 1 source · Wrong?
What to do
- Compare every model checkpoint against its declared original before deployment and after updates.
- Obtain models only from original publishers, verify published file hashes, and keep approved checkpoints in access-controlled registries.
- Restrict AI agent permissions, run code in isolation, limit outbound connections and require human approval for consequential actions.
- Monitor agents for suspicious child processes and unexpected network activity.
Written by AI from the sources, checked against them
Full story timeline · 1 update
Who reported · 1 outlet
Cyber Security News
Who and what is in it
Companies
Products