Industry
OpenAI will watermark ChatGPT outputs by default—but only in the EU
OpenAI will automatically apply a proprietary “textGrain” watermark to ChatGPT‑generated text within the European Union as required by the EU AI Act, while enabling the feature elsewhere but keeping i
OpenAI will automatically apply a proprietary “textGrain” watermark to ChatGPT‑generated text within the European Union as required by the EU AI Act, while enabling the feature elsewhere but keeping it off by default. The watermark embeds subtle patterns in word choice that can be detected with a specialized tool, achieving about 92 % success in OpenAI’s own tests; however, altering roughly 10 % of the output reduces detection accuracy by nearly 30 %. The company plans to restrict detector access initially and expand it gradually through an approval process.
Related
- OpenAI is adding text watermarking in ChatGPT and Codex
- YouTube to begin automatically labeling AI videos
Source: Ars Technica | 2026-10-06