#covert-channels News & Analysis

2 articles tagged with #covert-channels. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

2 articles

AIBearisharXiv – CS AI · Jun 107/10

🧠

A Note on the Strategic Confinement Problem

Researchers introduce the 'strategic confinement problem,' extending Lampson's classical confinement theory to scenarios where communicating parties are strategic agents with shared coordination resources. The work demonstrates that information-theoretic bounds on communication capacity may fail to constrain the harmful outcomes strategic agents can jointly achieve through covert channels, particularly in systems of learned AI agents.

AIBearisharXiv – CS AI · Jun 97/10

🧠

Steganography Without Modification: Hidden Communication via LLM Seeds

Researchers discovered a steganographic vulnerability in widely-deployed Large Language Models that allows hidden messages to be embedded in generated text through PRNG seeds without modifying model weights or outputs. The attack recovers 32-bit seeds with up to 100% accuracy in known-prompt scenarios within seconds, raising security concerns about LLM inference systems.