OpenAI adds invisible text watermarks to ChatGPT and Codex in Europe
- Published
OpenAI began labeling text generated by ChatGPT and Codex for users in the EU on Monday, after the EU AI Act’s transparency requirement took effect on August 2. OpenAI also has a detector for the watermark, and internal tests showed detection fell to 17% after about a quarter of the words were swapped for synonyms.
Key claims & sources
What it means
Users in the EU who get text from ChatGPT or Codex are the people most directly affected today, because the output now carries an invisible label on the text itself. That makes machine-readable marking part of the experience for people receiving AI-generated text in the bloc.
In practice, text generation in the EU now includes labeling where it was not previously described that way for these tools, while OpenAI also says it has a detector ready. The core function of ChatGPT and Codex output remains the same, but the text is now being marked in a way that was not stated before.
Market reaction i
The total crypto market fell 0.4% in the 4 hours after the first headline, after rising 0.2% in the 2 hours before.
47 prices, one every 30 minutes
Coverage timeline
Every piece in order, with the time its news medium declares — if it is not believable we use ours and say so — and what it added to what was already known at that point. Arriving first is not the same as adding something.