Wire flash
OpenAI unveils textGrain watermarking for EU AI Act compliance
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
OpenAI has announced textGrain, a text watermarking technology designed to comply with the EU AI Act. The system embeds invisible statistical signals into the word selection process of AI models. Starting immediately, API customers can selectively enable watermarks on certain models. Within the next few weeks, OpenAI will add invisible watermarks to outputs from ChatGPT and Codex for users in the European Union. The accompanying detection tool is currently only available to approved researchers and expert institutions. OpenAI also released detection performance data and limitations, allowing readers to assess the current capabilities and constraints of the text watermarking approach.
Source report
OpenAI has unveiled a text watermarking solution designed to address requirements under the EU AI Act. The company introduced textGrain, a technology that embeds invisible statistical signals into model word choices.
Key Details
- Availability: API customers can now selectively enable watermarking for certain models.
- Upcoming rollout: In the coming weeks, OpenAI will add invisible watermarks to outputs from ChatGPT and Codex in the European Union.
- Detection tool access: The watermark detector is currently available only to approved researchers and expert institutions.
Performance and Limitations
The original announcement includes detection data and boundary conditions for textGrain, allowing readers to assess the current practical capabilities and applicable scenarios of text watermarking technology.
Source
aihotNeutral / independent
Part of this Story
OpenAI rolls out invisible textGrain watermarking for ChatGPT and Codex in EU to comply with AI Act