Low Entropy Makes Code Generation Harder to Watermark Reliably
davidstutz92 · x · 2026-08-11
David Stutz points out a crucial technical caveat regarding AI text watermarking in models like Claude: not all text generation is equally 'watermarkable'.
He notes that code generation is particularly harder to watermark because it has lower entropy to exploit compared to natural language. This lack of randomness means that significantly more tokens are required to reliably detect the watermark in AI-generated code.
More from Safety
- Linux Kernel CVEs Surge From ~500 to 1500+ Per Release, LLMs Blamed for Bulk of the Rise — burny_tech · 2026-10-03
- COLM 2026 Launches DAIH Workshop on Deploying LLMs/VLMs Responsibly in Healthcare — StellaLisy · 2026-10-03
- Trillium Labs wants to do open research on recursive self-improvement and agents — nordicinst · 2026-10-03
- Trillium Labs Wants to Research Self-Improvement and Model Behavior in the Open — Wired AI · 2026-10-03
- Cloudflare Turnstile everywhere: anti-AI scraping walls now hit human users — sethlazar · 2026-10-02
- Filler tokens let frontier models reason invisibly: 13-point gains undetectable by CoT monitoring — PandaAshwinee · 2026-10-02