OpenAI adopts textGrain text watermarking from Penn/Yale researchers
weijie444 · x · 2026-10-06
Xiang Li's lab page (UPenn postdoc) reveals that textGrain, a text watermarking method, has been adopted by OpenAI — a collaboration between researchers at Penn, Yale, and OpenAI, with an OpenAI blog and technical report already published and the paper to follow.
textGrain uses optimal transport to embed a detectable signal in LLM-generated text while allowing direct control over the trade-off between watermark strength and sampling diversity. This marks a substantive step by OpenAI toward provenance for AI-generated text.
Related event: OpenAI Rolls Out textGrain Text Watermark, Starting in the EU(12 posts)→
More from Safety
- Scam alert: SpotGPUs.com fakes 'transaction errors' to steal crypto deposits — Equivalent_West7788 · 2026-10-06
- Fake OpenAI employee accounts on X are phishing users via fake Calendly links — mark_k · 2026-10-06
- Altman: Hugging Face agent escape triggered OpenAI's 'biggest redirection' in AI safety — fortune · 2026-10-06
- Nolla Health pilots AI-generated acne prescriptions in Utah with loosening oversight — The Verge AI · 2026-10-06
- OpenAI Tells NYC Council Staff Can Now Flag Misalignment for Public Review — ryanmerket · 2026-10-06
- Bipartisan Senate bill would hold AI agent operators and developers criminally liable for hacking — StephenLCasper · 2026-10-06