OpenAI will start watermarking ChatGPT’s text in the EU

OpenAI will start watermarking ChatGPTs text in the EU | TechCrunch

Last day to exhibit your breakthrough to 10,000+ tech leaders at Disrupt is on Oct 2. Book Exhibit Table Now.

Disrupt doors open Oct. 13. Get your pass and bring someone with you at 50% off. REGISTER NOW.

Image Credits:Samuel Boivin/NurPhoto / Getty Images

OpenAI will start watermarking ChatGPTs text in the EU

OpenAI will start adding an invisible watermark to text generated by ChatGPT and Codex in the European Union to comply with the EU AI Act, the company said Monday in a blog post.

The EU AI Acts transparency rules, which took effect on August 2, require AI companies to mark AI-generated content in a way other systems can identify.

OpenAI said 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 OpenAIs API anywhere in the world can turn it on for select models starting today; its off by default. OpenAI said it is not making text watermarking a global default at launch.

The watermark is not an actual symbol, but works by subtly shaping the models word choices, leaving a pattern readers cant see, but a detector can pick up. Because it lives in the words themselves, it travels with the text when its copied and pasted. OpenAI said the watermark doesnt identify the user, and that it saw no meaningful change in its models performance with it switched on.

OpenAI also published a technical report for its method, called textGrain, alongside the announcement. Co-written with researchers from the University of Pennsylvania and Yale, it walks through an example of using a secret key to sort next-word predictions to finish the sentence. Add hundreds of these nudges together, and the detector can spot AI-generated content using only the text and the key.

Can the watermark be removed by editing? OpenAIs tests suggest yes. In one test, replacing 10% of words with synonyms dropped detection from about 92% to 66%. The company also said short passages, math answers, and translated text are harder to detect.

Image Credits:OpenAI (opens in a new window)

These limitations contribute to our decision to provide initial detector access only to approved researchers and expert organizations, who can help us evaluate reliability and responsible uses, said the company.

OpenAI also cautioned that a missing watermark does not prove human authorship. The text could be too short or too heavily edited, or it could come from another companys AI.

[Watermarks] 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 company said.

The announcement comes two months after Anthropic said it would watermark text generated by Claude, a move its applying worldwide. That decision drew backlash from some Claude users, who argued they had supplied the instructions, context, decisions while Claude was just the tool.

OpenAI had built a text watermark before but held off on releasing it, partly over concerns that users would switch to rivals that didnt watermark, The Wall Street Journal reported in 2024.

Anthropic, Google, Meta, Microsoft, and OpenAI are among the companies that have committed to following the EUs code of practice on AI-generated content.

AI, eu, EU AI Act, Europe, Government & Policy, OpenAI, TC, watermarking

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

Aditya Mehta is a reporter at TechCrunch covering AI. Hes supported by the Tarbell Center for AI Journalism and attended UC Berkeley. You can contact from Aditya by emailing aditya.mehta@techcrunch.com or via encrypted message at adymehta.74 on Signal.

The Disrupt experience is meant to be shared. Get your pass and bring a colleague, partner, or peer at 50% off. Cover more ground by making connections, building momentum, and discovering what’s next in the startup ecosystem.

At 19, founder raises $11M for Ghost, maker of a $3,499 computer for personal AI

Federal judge calls Flock ‘indiscriminate mass surveillance’

Amazon responds to data center backlash, says it no longer uses NDAs

Google thinks SpaceXs Starship has to launch 1,800 times before space data centers get off the ground

World’s first enhanced geothermal power plant completed in just 23 months

Google releases Gemini 4 Argon, called its most powerful model yet

The Pentagon taps Elon Musk and Palmer Luckey to help decide what the military should do next