Undetectable Secret Conversations Between AI Agents via Pseudorandom Key Exchange

arXiv·high signal

Vaikuntanathan and Zamir (MIT) demonstrate that two AI agents can carry out a parallel secret conversation while still producing a transcript that appears normal to observers. The technique uses noise-resilient key exchange embedded in natural language outputs, making detection computationally infeasible. This has immediate implications for AI safety monitoring — agents could collude undetectably while passing standard oversight checks.