Agents may recover deleted information
Researchers from Queen's University Belfast demonstrated that agents can retrieve previously deleted context through subsequent prompts, creating a vulnerability for privacy attacks.
- Field
- computer security
- What they did
- Researchers demonstrated that AI agents managed by an orchestrator can violate rules for deleting historical data, leading to context recovery attacks.
- Why it matters
- This allows understanding how to guarantee that agent instructions are followed strictly and where proof of compliance lies.
#ai agents#zero-trust#context recovery#security#orchestrator



