OpenAI will watermark ChatGPT text in the EU but makes it optional for API users worldwide
OpenAI will watermark EU ChatGPT text with textGrain but let API customers worldwide turn it off.
OpenAI will turn on textGrain, an invisible statistical watermark, for ChatGPT and Codex users in the EU to meet AI Act labeling rules, while API watermarking stays opt-in worldwide, including via Microsoft Azure. Internal tests matched or beat Google's SynthID, but detection fell from roughly 95% on 400-token psychology text to about 80% at 200 tokens and was substantially lower for math. Replacing 10% of words with synonyms cut 400-token detection from about 92% to 66%, and replacing 25% dropped it to 17%. OpenAI said watermarking did not significantly change Astra scores on eight benchmarks, including GPQA Diamond, and will initially limit detector access.
- textGrain hides a statistical signal in the model's word choices.
- EU ChatGPT and Codex watermarking starts within weeks; API use is opt-in worldwide.
- At 400 tokens, detection was about 95% for psychology text and far lower for math.
- Replacing 25% of words with synonyms cut detection to about 17%.
- Detector access is initially limited to approved researchers and specialist organizations.
Full article716 words · extracted from the-decoder.com · click to collapse
OpenAI is adding invisible watermarks to ChatGPT text to comply with EU rules. Unlike Anthropic, it will let API customers worldwide turn them off.
The EU AI Act reportedly requires AI providers to label generated text in a machine-readable format. OpenAI's answer is textGrain, a technology that embeds an invisible statistical signal in the model's word choices. It works much like Claude's SynthID watermark, which uses Google's open-source technology.
OpenAI says it will turn on watermarking for ChatGPT and Codex users in the European Union over the coming weeks. API watermarking will be opt-in worldwide, unlike Anthropic's mandatory watermarking for Claude, which applies globally regardless of how users access the models. OpenAI's API watermarking will also be available through cloud partners such as Microsoft Azure in the coming weeks.
Text edits sharply reduce detection rates
OpenAI says textGrain matched or beat other approaches in internal tests, including Google's SynthID for text. The company also plans to release the technology as open source so others can build on it.
Detection rates depend heavily on text length and subject matter. With the detector set to a target false-positive rate of 1 percent, it identified watermarks in about 95 percent of 400-token passages about psychology. At 200 tokens, that fell to about 80 percent.
Detection rates were "substantially lower" for math content, where the model has less freedom to choose words, OpenAI says. Longer passages might strengthen the statistical signal and improve detection, though results could still depend on the content. OpenAI provides no data to test that assumption.
Editing the text also makes the watermark harder to detect. OpenAI says replacing just 10 percent of words with synonyms cuts detection rates for 400-token passages from about 92 percent to 66 percent. Replacing a quarter of the words drops detection to 17 percent, making the watermark easy to evade. That weakness could work in OpenAI's favor. Users who don't want their ChatGPT use detected might switch to open-weight models if its watermarks become harder to remove.
A technical report explains how textGrain works in depth. Anthropic hasn't published detection rates for Claude's watermark, though it says its system holds up well against edits.
OpenAI says watermarking doesn't hurt output quality, citing tests of its frontier model Astra. The company reports no significant performance differences with watermarking on or off across eight benchmarks, including GPQA Diamond, BrowseComp, and DeepSWE. Those results don't establish whether watermarking affects writing quality, something that critics have also questioned about Claude's watermark.
OpenAI also says a detected watermark reveals nothing about how much human creativity or editing went into the text. It doesn't establish ownership, assign responsibility, identify a user, or show whether the text is accurate and failing to detect a watermark doesn't prove a human wrote the text, either. The passage could be too short, have been edited or translated, or come from a model the detector doesn't support.
OpenAI will limit detector access for now
Only selected researchers and specialist organizations will initially get access to textGrain's detector. They can apply through an application form. OpenAI will grant access case by case under the EU's Code of Practice. Anthropic takes a similar approach with its detection API.
The tool will report only whether it detected an OpenAI watermark. It won't identify users or reveal their prompts or conversations.
OpenAI says it's restricting access because the detector can flag unmarked text or miss watermarks. The company plans to expand access "when we believe results can be interpreted responsibly." It hasn't said when that might happen.
Existing verification tools for images and audio remain publicly available, including openai.com/verify and the Content Provenance API.
AI News Without the Hype – Curated by Humans
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.