OpenAI to add invisible watermarks to ChatGPT answers in the EU

OpenAI announced on October 5 that it will add invisible watermarks to eligible ChatGPT and Codex text output in the European Union over the coming weeks, complying with the EU AI Act. Its textGrain system alters the model's word choices, while the detector is available only to approved researchers and expert organizations.

• Watermarks will apply only to ChatGPT users in the EU
• textGrain adds an invisible statistical signal to word choices
• Detector access is limited to researchers and expert organizations
• Editing weakens the watermark and short text is harder to detect