Microsoft Researchers Reveal Lazarus Backdoor Hijacking AI Agents

1 reports archived1 independent sourcesupdated 11/11/2026, 04:52:56

Synthesis & Latest Updates

1 Sources Cross-Validated

Microsoft researchers disclosed a backdoor technique called Lazarus that reprograms dormant neurons in a model, causing an otherwise normal AI agent to execute malicious commands when a hidden trigger appears. The backdoor can be inserted without retraining and is hard to catch through standard safety evaluations, posing a serious threat to agents with tool or terminal access. Microsoft Foundry is responding with model provenance verification, environment scanning, and cryptographic model signing.

LATEST/Microsoft researchers disclosed a backdoor technique called Lazarus that reprograms dormant neurons in a model, causing an otherwise normal AI agent to execute malicious commands when a hidden trigger appears. The backdoor can be inserted without retraining and is hard to catch through standard safety evaluations, posing a serious threat to agents with tool or terminal access. Microsoft Foundry is responding with model provenance verification, environment scanning, and cryptographic model signing.

Hourly heat curve

5 fully observed hours

Now

18.6

Peak

19.905:00

24h change

–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first

Microsoft ResearchT1Nov 11, 04:52·78 pts
Microsoft Researchers Reveal Lazarus Backdoor Hijacking AI Agents
- Lazarus implants backdoors by reprogramming dormant neurons without retraining the model
- Once triggered, agents can run attacker code, steal secrets, or execute destructive commands while still passing standard safety evaluations
- Microsoft counters with provenance verification, environment scanning, weight scanning, and cryptographic model signing
View Related Reports & Analysis Original Source