OpenAI will add invisible, machine-readable watermarks to text generated by ChatGPT and Codex for users in the European Union to meet the EU AI Act’s transparency requirements. The rollout will reach all EU users, across all plans, over the coming weeks.
Starting October 5, 2026, OpenAI will offer text watermarking as an opt-in feature to global API customers. It will be off by default.
OpenAI described text watermarking as an early technology with limitations. The phased rollout will allow the company to gather real-world feedback. Access to a detection tool will initially be limited to approved researchers and expert organizations, who will help evaluate and improve the technology.