OpenAI will start watermarking ChatGPT text in the EU

OpenAI will begin adding an invisible watermark to text produced by ChatGPT and Codex inside the European Union, the company said in a blog post on Monday, to comply with the bloc's AI Act transparency rules. TechCrunch reported the move the same day.

The watermark will roll out over the coming weeks to eligible ChatGPT and Codex users on all plans, but only in the EU. Developers using OpenAI's API anywhere in the world can switch it on for select models starting today; it is off by default, and OpenAI said it is not making text watermarking a global default at launch.

The mark is not a symbol. It works by subtly shaping the model's word choices, leaving a pattern that readers cannot see but a detector can pick up. Because it lives in the words themselves, it travels with the text when it is copied and pasted. OpenAI said the watermark does not identify the user and that it saw no meaningful change in model performance with the feature enabled.

Alongside the announcement, OpenAI published a technical report for its method, called textGrain, co-written with researchers from the University of Pennsylvania and Yale. It walks through using a secret key to sort next-word predictions, so that after hundreds of these nudges a detector can flag AI-generated content from the text and the key alone.

OpenAI's own tests suggest the watermark can be removed by editing. In one test, replacing 10 percent of words with synonyms dropped detection from about 92 percent to 66 percent, and short passages, math answers and translated text proved harder to detect. Those limits are why initial detector access is restricted to approved researchers and expert organizations.

The company also cautioned that a missing watermark does not prove human authorship. The text could be too short, too heavily edited, or generated by another company's AI. Watermarks, it said, can indicate that an OpenAI system generated or processed part of a passage, but not how much human judgment, editing or creativity went into it.

The move lands about two months after Anthropic said it would watermark text generated by Claude, and follows the EU AI Act's transparency obligations taking effect on August 2, which require AI companies to mark AI-generated content in a way other systems can identify. Rolling it out first in the EU, rather than globally by default, lets OpenAI answer the regulatory requirement while keeping room to adjust.

The questions to watch are how reliable detection proves in real-world use, whether developers choose to enable the API option, and how EU regulators judge whether the marking works in practice.