DEVOURED
OpenAI is adding invisible watermarks to ChatGPT and Codex for users in the EU
OpenAI is rolling out invisible textGrain watermarks for EU users to comply with the EU AI Act’s transparency requirements.
What: OpenAI will apply statistical patterns, labeled 'textGrain,' to text output generated by ChatGPT and Codex in the EU. API customers globally can opt-in to the feature. The company warns that the signal can be degraded by post-generation editing or short-form responses.
Why it matters: This is a direct response to the EU AI Act, forcing AI labs to implement technical solutions for content provenance that may become the global standard for enterprise compliance.
Takeaway: If you are integrating OpenAI models into an application for the EU market, review the API documentation for how to toggle these watermarking features.
Decoder
- EU AI Act: A landmark European Union regulation that mandates transparency and risk management for artificial intelligence providers.
Original article
OpenAI will begin adding invisible, machine-readable watermarks called textGrain to eligible ChatGPT and Codex output in the EU to comply with transparency requirements in the EU AI Act. The technology embeds a statistical pattern into generated language rather than adding hidden characters. API customers worldwide can optionally enable it for supported models. OpenAI acknowledges that watermarking remains imperfect because editing and short outputs can weaken the detectable signal.
Explore all 55 articles in today's digest
Fresh dev news, AI-digested daily. Never miss a trend at devoured.fyi