How OpenAI's text watermark works: invisible statistical signals, no performance hit
OpenAI · x · 2026-10-06
OpenAI details how its text watermarking works:
- Answers one question: "Was this likely generated by an OpenAI model?"
- Embeds an invisible statistical signal during generation; a detector can look for it, with no visible marks or special characters
- Does not identify any person, organization, account, conversation, or prompt, nor who wrote or owns the text
- Testing reportedly shows no effect on model capability, speed, or how responses read
Related event: OpenAI launches textGrain text watermarking, starting in the EU(25 posts)→
More from Models
- Google silently disables logprobs on Gemini Hydrogen, Neon, Argon and Carbon models — dejanseo · 2026-11-20
- Bug Hunt chart: score vs subscription cost, Gemini 4 Neon xhigh tops at 63 for $1.80 — PawelHuryn · 2026-11-20
- Measured: Claude Max delivers 46.4x API value on Opus 5.5, beating ChatGPT Pro's 10.25x — PawelHuryn · 2026-11-20
- Dev complains GPT keeps skipping steps and ignoring instructions — lxfater · 2026-11-20
- GPT-6.2 Astra and Sol show OpenAI optimizes performance per dollar, not raw capability: analysis — VraserX · 2026-11-20
- Mamba+MoE hybrid reasoning model Olala-v1.0-mini-7A1B trends on Hugging Face — OVHaiLLM · 2026-11-20