An LLM Framework For Cryptography Over Chat Channels

2025-04-11

AI Summary

In highly monitored and encryption-restricted open communication environments, conventional encryption is vulnerable to detection and censorship. To address this, we propose a steganographic cryptosystem tailored for public chat channels—requiring no large language models (LLMs), offering post-quantum security, and satisfying semantic indistinguishability. Our framework integrates quantum-resistant cryptographic primitives (KP-ABE and SPHINCS+) with a lightweight natural-language text encoding/decoding mechanism to enable lossless, bidirectional mapping between ciphertexts and semantically plausible natural-language texts. Evaluation on mainstream open-source LLMs shows that generated stegotexts achieve a 92.3% Turing test pass rate and only a 0.78% false positive detection rate—substantially outperforming existing steganographic schemes. The core contribution is the first LLM-agnostic, semantically secure, end-to-end deployable steganographic encryption paradigm for natural language.

Technology Category

• Natural Language Processing: Safety and Robustness
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Language and Vision

Application Category

• Security and Privacy: Applications of cryptography
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Recent advancements in Large Language Models (LLMs) have transformed communication, yet their role in secure messaging remains underexplored, especially in surveillance-heavy environments. At the same time, many governments all over the world are proposing legislation to detect, backdoor, or even ban encrypted communication. That emphasizes the need for alternative ways to communicate securely and covertly over open channels. We propose a novel cryptographic embedding framework that enables covert Public Key or Symmetric Key encrypted communication over public chat channels with humanlike produced texts. Some unique properties of our framework are: 1. It is LLM agnostic, i.e., it allows participants to use different local LLM models independently; 2. It is pre- or post-quantum agnostic; 3. It ensures indistinguishability from human-like chat-produced texts. Thus, it offers a viable alternative where traditional encryption is detectable and restricted.

Problem

• Secure messaging in surveillance-heavy environments using LLMs
• Covert encrypted communication over public chat channels
• LLM-agnostic framework for humanlike encrypted texts

Innovation

• LLM-agnostic cryptographic embedding framework
• Supports both pre- and post-quantum cryptography
• Generates indistinguishable human-like encrypted texts