Post

John Gruber
@gruber.foo

It’s unacceptable for a tool to sacrifice an iota of clarity, coherence, meaning, quality, etc. for the purpose of embedding hidden clues within the text to suggest its provenance. daringfireball.net/2026/08/anthro…

daringfireball.netAnthropic’s ‘Watermark’ Text Adulteration in Claude Is a Perversion of WritingIt’s unacceptable for a tool to sacrifice an iota of clarity, coherence, meaning, quality, etc. for the purpose of embedding hidden clues within the text to suggest its provenance. The idea that anything other than *my* needs should factor into the generation of text *for me* is patently offensive.
George Black
@george.black

Suppose the model has a 49.9% probability of choosing banana, and 50.1% pineapple. Today, without watermarking, you run the model twice and get two different results. Sure, watermarks modify the probability. But you’re still getting a word it was *already* choosing half the time?

George Black
@george.black

Banana, the “inferior” word, was already showing up in your output before. I don’t think watermarking makes the model produce something it wasn’t already doing. I could be misunderstanding how it works though, it’s all kinda way over my head. 😅

Adam Lasnik
@adamlasnik.net

I read through John's very long rant, as well as the supporting documentation and referenced essay and... I think this is a lot of ado about nothing. Nothing practical or perceptible, that is.

John Gruber
@gruber.foo

You will note that Anthropic doesn't provide before / after examples. Nor do they yet provide a watermark detector. Google apparently has been using this in Gemini for a while and don't provide a detector either.

0 reposts2 likes

Adam Lasnik
@adamlasnik.net

I hear you on that, and I can see why those things are frustrating, but it doesn't change my belief (and others') that your core assertion of watermarking making the output worse is inaccurate, given how LLMs work. (and again, I *do* agree with you about the folly of this regulation).