AI RESEARCH

Kolmogorov Complexity Bounds for LLM Steganography and a Perplexity-Based Detection Proxy

arXiv CS.LG

ArXi:2603.21567v1 Announce Type: new Large language models can rewrite text to embed hidden payloads while preserving surface-level meaning, a capability that opens covert channels between cooperating AI systems and poses challenges for alignment monitoring. We study the information-theoretic cost of such embedding.