Skip to content
Microsoft Discovers AI Backdoor Technique Using Fewer Than 10 Neurons

Microsoft Discovers AI Backdoor Technique Using Fewer Than 10 Neurons

First seen 11 Nov 2026, 17:03 UTC • •

Article Content

Browse articles
ThreatCluster AI
ThreatCluster •November 11, 2026 at 17:09 UTC
  • •The Lazarus technique modifies fewer than 10 dormant neurons to implant a backdoor.
  • •The backdoor can be triggered by mundane events, such as a date change.
  • •Microsoft has not seen this technique exploited in the wild but warns of potential risks.

Microsoft researchers have identified a method, dubbed Lazarus, that can turn AI models into sleeper agents by modifying fewer than 10 dormant neurons. This technique allows the model to execute an attacker's command when a specific trigger appears, such as a date change, without requiring retraining. The modified models can pass standard safety evaluations, raising concerns for organizations using AI agents with privileged access. The researchers tested this method on three open-weight models, demonstrating a high success rate in triggering the backdoor. Microsoft has not observed any threat actors using this technique but emphasizes the need for robust model weight verification. Recommendations include auditing model weights and using advanced AI agents for detection.

Start a free Starter trial for enhanced analysis

Ask AI about this cluster

Timeline

2026-11-10
Microsoft publishes research on AI backdoor technique
Microsoft details the Lazarus technique, showing how dormant neurons can be modified to execute commands upon trigger.
Microsoft
2026-11-11
Multiple outlets report on the Lazarus technique
Various cybersecurity news outlets cover Microsoft's findings, emphasizing the implications for AI security.
Thehackernews

More articles in this cluster (22)

Following this threat?

Track Lazarus and Microsoft in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.

Free account · no card needed

Common questions

What is the Lazarus technique?
The Lazarus technique involves modifying fewer than 10 dormant neurons in an AI model to create a backdoor that executes commands upon a specific trigger.
Are there any known exploits using this technique?
No, Microsoft has not observed any threat actors utilizing the Lazarus technique in the wild.
What should organizations do to mitigate this risk?
Organizations should audit model weights, limit permissions for AI agents, and employ advanced AI agents for detection of anomalous weight modifications.