Steganeur – Hide secret messages in LLM-generated text (Rust)

Model output can carry a payload that reads as ordinary prose, which is worth understanding if your product publishes anything a model generated.

• LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
• token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.

• postThe encrypted reasoning blobs your secret scanner can't readAlphaSignal
• articleControllable Neural Text GenerationLilian Weng
• articleStealing Reasoning Traces from Proprietary LLM APIssimonwillison.net

Checking sign-in…

Loading comments…