Anthropic has quietly rolled out a new policy that embeds invisible watermarks into Claude’s text outputs, marking them as machine-generated. This move is designed to comply with the EU AI Act’s Transparency Code, which now mandates that tech firms label AI-generated or AI-edited content in ways that computer systems can recognize. While regulators in Europe might be pleased, a number of AI users are far from thrilled—and Reddit has become the go-to spot for airing those grievances, though opinions there are sharply divided.

One particularly dramatic post came from a user named visionode, whose account is notably just three weeks old. Visionode framed the watermarking system as a draconian conspiracy intended to punish innocent chatbot users across the globe. Their core argument seems to be that while experienced Claude users might dodge detection by paraphrasing or running their outputs through other AI tools, the average person will end up flagged. “Who will get caught. You. The student who used Claude to reorganize a paragraph. The journalist who asked the AI to summarize a two-hundred-page transcript. The writer who had creative block and asked for synonyms. Those guys come out of the process with a digital tattoo on their forehead.”

I’m not here to dismiss visionode’s frustration, but those examples don’t hold up well under scrutiny. A journalist who uses AI to distill a lengthy transcript isn’t going to mind a watermark on that summary—unless they’re planning to lift it verbatim into their piece, which is clearly unethical and shouldn’t happen in the first place. The same goes for a student who pastes Claude’s output straight into an essay after asking for a “reorganized paragraph.”

Other Redditors weren’t exactly rallying behind the poster’s outrage. “Get a load of this guy,” one commenter wrote. Another advised the original poster to take “a deep breath.”

Visionode wasn’t alone in their complaints, though. A different critic called the watermarks “unethical” and “disgusting,” insisting that by using Claude, they had done the bulk of the work themselves. In their view, the chatbot was just a “tool” that helped carry out their labor. “I gave the instructions, context, decisions, and countless refinements, claude was the tool. If Claude starts watermarking the code or anything else it generates, what exactly is it claiming credit for.” the poster asked. Once again, other users piled on. “It’s not claiming credit though,” someone shot back. “It’s about being able to detect AI generated outputs because of the risks AI generated outputs can cause in various situations.”

“Bro couldn’t even complain about Claude without using Claude to write it,” another user joked. Some critics, however, avoided the victimhood angle and raised more measured objections. One poster pointed out a perceived hypocrisy in watermarking an editorial product that was itself built by scraping other people’s work. “I think it’s a very sinister direction to take,” they said. “I don’t use Claude to write anything but having an AI that watermarks your work is terrifyingly ironic given how many of the frontier models came by their training data.”

Overall, though, the broader sentiment on Reddit seems to lean in favor of the watermarking approach, with many seeing it as a practical way to track algorithmically generated material. “There is literally no good argument for why this isn’t a good idea,” a user on another thread remarked. “The only reason you wouldn’t want this is to lie to people.”