Discover · microsoft.com

Lazare, veni foras: Reprogramming Dormant Neurons to Backdoor AI Agents

Bing Maps microsoft.com

Microsoft researchers have discovered “Lazarus,” a backdoor technique that can reprogram dormant neurons in an AI model, causing an otherwise normal AI agent to execute malicious commands when a hidden trigger appears.

The threat is especially serious for AI agents with tool or terminal access. A triggered model could quietly download and run attacker code, steal secrets, or execute destructive commands while still passing standard safety evaluations.

More ways to use this
Open on another device
Scan this QR code to open the same public page

Scan to open this page. Your saved notes and account access are not included.

Download QR code

Open the page link

Named in it

Keep going

Each of these is read from what this estate already kept: the feed a piece came through, the publisher behind it, the credits and topics the publisher printed. None of it is a recommendation, an identity or a record kept anywhere else.

More like this

Also from Bing Maps

  1. TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficientmicrosoft.com
  2. Microsoft Research Focus – Edition #98Anson Ho microsoft.com
  3. Formally verifying agent-written full-stack web apps with PDD KitAlex Lavaee, Matthai Philipose microsoft.com
  4. Learn the rules: The hacker mindsetWeishung Liu, Mike Walker microsoft.com
  5. From synthetic enterprise data evaluations to real Microsoft Copilot AI confidencemicrosoft.com
  6. Microsoft Research Focus – Edition #97Sowbhagya Suresh microsoft.com