TechCrunch· ReportingProduct launch

Some Claude users are mad that Anthropic’s new watermarks will catch them using it at their jobs, classes

Anthropic is watermarking Claude's outputs to comply with the EU AI Act, inserting invisible code that marks text as AI-generated, which has sparked user backlash over potential detection of AI usage.

From the source

Anthropic recently made the decision to watermark Claude’s outputs — inserting invisible code into the chatbot’s editorial text that marks it as AI-generated.

techcrunch.com

Announced

12 Aug 2026 22:26 UTC

Detected

14 Aug 2026 13:14 UTC (+38h after)