Anthropic's Claude watermarks spark user backlash over AI detection fears
Original: Some Claude users are mad that Anthropic’s new watermarks will catch them using it at their jobs, classes
Why This Matters
The backlash highlights growing tension between AI transparency regulations and widespread undisclosed AI use in workplaces and education.
Anthropic introduced invisible watermarks in Claude's text outputs to comply with the EU AI Act's Transparency Code, which mandates machine-readable labels on AI-generated content. The move has drawn anger from some users on Reddit who fear being caught using AI at work or school.
Anthropic has begun embedding invisible watermarks into Claude's text outputs, a policy change driven by the EU AI Act's Transparency Code. The regulation requires tech companies to label AI-generated or AI-edited content in a way that is detectable by computer systems. While the move satisfies European regulatory requirements, it has sparked a backlash among some Claude users who voiced frustration on Reddit. One user, with an account only three weeks old, described the watermarking system as a conspiracy targeting ordinary users, arguing that while sophisticated users could scrub AI traces by paraphrasing through other tools, average users would be exposed. Other Reddit users largely dismissed the complaints. One commenter replied simply, 'Get a load of this guy,' while another told the original poster to take 'a deep breath.' A separate user called the watermarks 'unethical' and 'disgusting,' arguing that because they provided instructions and refinements, Claude was merely a tool and should not be marking its outputs. Critics responded that the watermarks are not about claiming credit but about identifying AI-generated content given the risks it can pose in professional and academic contexts. The article notes that use cases cited by complainers — such as submitting verbatim AI-generated text in essays or articles — are themselves considered unethical practices.