OpenAI ships ChatGPT text watermarking to comply with EU AI Act
A finished 2023 detector with 99.9% accuracy sat shelved for three years after a survey found ~30% of users would use ChatGPT less.
Archived Updated
Everything it touches
15 names · 5 voices · 6 source posts
The story’s neighborhood
NAMESVOICESPOSTS
Explore the graph, or choose a connection below.
Follow the evidence
The source posts
This brief comes from Tech Twitter’s edited headlines. The posts below are the sources grouped with the story; statements in them belong to their authors.
We're expanding our approach to content provenance to include text in response to EU regulatory requirements, while recognizing the significant limitations of current text watermarking technology. Our tools already help verify whether an image or audio file was created with our models. This work builds on those efforts to help people better understand when content may have been generated or edited with an OpenAI model. In the EU, we’ll start watermarking eligible text from ChatGPT and Codex over the coming weeks to comply with the EU AI Act. Customers using our API can turn on text watermarking for select models worldwide today.
Text watermarking in ChatGPT and Codex is rolling out. The new development is that, unlike Anthropic and Google, OpenAI (for now) is launching their watermarking - textGrain - 𝘰𝘯𝘭𝘺 in the EU, not globally.
"Having to set up and fiddle with your loops… I don't think this is the way that it's going to work." @thsottiaux (Head of ChatGPT & Codex) says we are heading towards Dots being the primary way you talk to AI. Full conversation:
Day 1/ We have optimized the default speed to be ~50% faster across GPT-6 Astra and GPT-6.1 Sol through the subscription across all our products and partners using Sign in With ChatGPT (including OpenCode, Pi, Amp, Devin, ...). No changes needed on your end and this should be felt within the next two hours.
OpenAI built a 99.9% accurate ChatGPT text detector back in 2023. Then a customer survey found almost 30% of users would use ChatGPT less if it ever shipped. So the finished tool sat in a drawer for three years. Today it shipped, and the only thing that changed is the law. The EU AI Act now requires AI-generated text to be machine-readable as AI, with fines up to 3% of global revenue for providers who skip it. Look at how it shipped, because every detail follows from that 30% number. The watermark turns on by default only in the EU, the one region where leaving it off is illegal. In the API it exists worldwide but sits off by default. And the detector, the part that would let a teacher check an essay or an editor check a submission, is locked to approved researchers who apply for access…
Day 2.1/ We have made Auto-review free for all users signed in through a ChatGPT account. You can enable it in settings > permissions > auto-review. Auto-review improves upon the default sandbox setting that requires you to approve everything, which is prone to decision fatigue unless you spend a lot of time configuring specific rules. It allows you to run long tasks while having a second agent review all actions taken by the primary agent. Its only goal is to prevent high-risk actions from being taken and to protect against unwanted actions that are not aligned with the original user intent. This Auto-review feature is now free and does not draw usage from your plan.