Business Insider reports Claude users canceling over the watermark
5 Aug 15 8:46 AM · 63d ago · 26 articles · 38 posts · 5 comments · 7 sources · development 5 of 8
Business Insider interviewed four users who said they canceled Claude Max subscriptions specifically citing the watermark, worried it could misattribute authorship on lightly-edited or human-written work in professional and academic settings.
“I do use it for everything, mostly coding apps and tools”
Arturo VillarroyaAnthropic AI lab, maker of ClaudeJohn Gruber Daring Fireball writerArturo Villarroya Claude Max subscriber, public policy workerVladislav Rajtmajer Freelance developer, Czech RepublicJames Padolsey Author of interactive watermarking explainer
The whole story articlespostscomments the bright band is this development · numbered dots are the others · click one to jump
What was reported 2 claims about this development
-
first by Forbes, 67d ago
-
first by City A.M., 67d ago
What people said 24 voices · best of 28 · verbatim
-
wonder why! https://www. businessinsider.com/claude-use rs-cancel-subscriptions-citing-anthropic-new-ai-watermark-2026-8
-
E
Also, Anthropic's just the first one out of the gate. The other platforms will be doing this, too, in short order, to comply with the EU AI Act Transparency Code (ref:
-
Anthropic has explained its text watermarking approach.
-
I may be wrong but in my opinion this is not technically possible. They can say a text is not AI written with more confidence, but it should be absolutely impossible to say a text is 100% AI written. There is nothing that stops a human to write something AI also wrote. It could be less probable, it could be not common, but there is no way to be…
-
T
Finally a good explanation of the text watermarking Anthropic will apply to Claude-generated content: https://www. anthropic.com/news/claude-text -watermark . They even link to the SynthID paper their approach is based upon, and have a real-world example so even lay persons should be able to understand the technique. I don't blame them for using…
-
John Gruber dislikes Anthropic's watermarking because Claude may choose words that make AI text easier to detect instead of simply writing the best answer. — I personally don't care. Claude's writing was already too clunky to send without editing. If editing also defeats the watermark, bonus.
-
🔴 So Anthropic just wrote a statement just to confirm all concerns about watermarking. - It's permanent and not removable. - It will be applied to all new models from now on, and existing after December. - They will apply it also to translations. - They will provide a binary AI
-
> Can a watermark be traced back to me or my organization?> No. The watermarking applies to Claude and its outputs. It doesn’t identify anything to do with individual usersBut there's nothing stopping them from doing this in the future. In fact, couldn't they append the user's ID to the key in some manner?
-
H
Claude will soon be adding an invisible watermark to all the content it generates, and OpenAi and the other major models are soon to follow thanks to new laws from places that have FUNCTIONING governments https:// tech.yahoo.com/ai/claude/artic…
-
S
Anthropic is rolling out new watermarks for Claude, a move that could make AI generated content easier to trace and verify as synthetic media keeps spreading. - techcrunch.com/2026/08/15/a... #AI #Anthropic #Claude #TechCrunch
-
sounds like Anthropic's watermarking is MUCH more interesting than hiding invisible Unicode, as most people assumed. the watermark is baked into the token selection itself! at a high level: normal Claude: context → probability cloud + random sampling → token watermarked
-
2 years ago i created nonfungibledocs.com based on the best of multi variated testing, statistics and nlp. The broad idea was that even a simple page can subtly be written in N ways for N recipients, such that a leak can be traced back to the version.
-
S
Engadget (@engadget) Anthropic이 생성 텍스트의 출처를 식별하기 위한 텍스트 워터마킹 접근법을 설명했습니다. AI 생성물의 추적성·신뢰성 확보와 콘텐츠 출처 검증에 관련된 기술로, LLM 배포 및 안전성 정책을 설계하는 개발자에게 참고할 만합니다. https:// x.com/engadget/status/20886906 70181450140 # anthropic # watermarking # llm # aisafety # provenance
-
M
This one is getting attention: Anthropic shares more details about how Claude’s new watermarks will work
-
Do i understand correctly - to check watermark you need full model weights, of all org models. So running the check is basically the same as running every model once? That's really expensive
-
S
Anthropic plans to watermark AI-generated text from Claude Source: BleepingComputer https://www. bleepingcomputer.com/news/arti ficial-intelligence/how-anthropic-plans-to-watermark-claudes-ai-generated-text/ # AI # Anthropic # Claude
-
If it’s an EU law can you atleast only enable it for EU customers? I don’t even serve eu anymore because of gdpr
-
D
This is a cool explainer, and it’s interactive, too! # ai # watermarking
-
A
Anthropic just started watermarking every single word that Claude writes, invisible, embedded in the text itself, and it rides along when you paste it onto your website. So, if you've been publishing AI content, there's now a permanent mark on it saying where it came from. It's on everything, every single model launched since August 2nd across the…
-
N
Turns out the easiest way to remove # Claude ’s watermark is to change enough words so that you basically write the text yourself “Nothing is added to the text, and there are no hidden characters or invisible Unicode. Because the pattern sits in the word choices themselves, it travels with copied and pasted text in a way attached metadata cannot”…
-
A
From yesterday. "Some users, who shared confirmation of their cancellations with us, said they disliked the prospect of a persistent marker in their work and were worried about their AI use showing up in client work, such as coding." "Even if you edit your own work that you wrote, they add the watermark anyway." Business Insider: Claude users are…
-
D
These new word watermarks Claude is implementing seem like a better way to discern AI-written text vs what we have now, which is relying on AI to find similar patterns based on what it’s programmed to find instead of a set rule of order. Seems all AI companies will do this. https:// techcrunch.com/2026/08/15/anth…
-
S
Anthropic plans to watermark Claude text by subtly biasing word choices, not hidden markers. Google's test on 20M Gemini responses showed no quality loss. But detection has real limits: short text or editing can weaken the signal. What the watermark proves depends on how much Claude actually writes. https://www. implicator.ai/claude-watermark…
-
K
Meanwhile, anthropic. The people canceling are a problem anyway. https://www. businessinsider.com/claude-use rs-cancel-subscriptions-citing-anthropic-new-ai-watermark-2026-8 # ai # llm # anthropic # aislop
All 8 developments of Claude's AI text watermark sparks backlash, cancellations… →
Hacker NewsNewswiresMastodonRedditXBlueskyGoogle NewsLobsters