How Claude marks AI-generated content • Hacker News Board

Re: How Claude marks AI-generated content

#61

Post by pixl97 » Mon, Aug 10, 2026, 11:14 PM UTC

amelius wrote: post #33

A_D_E_P_T wrote: post #13

Earlier quoted context omitted.

If it's that simple and obvious, you'll have 10 "Remove Claude Watermark" web-apps by the end of Day 1. Most of them coded by Claude. Hell, it'll probably happen no matter how sophisticated their watermark is. There's no watermark in text that can't be detected and removed, and no text that can't be converted to generic keyboard ASCII.

You forgot about the cases where (1) people don't care, (2) people want to say "I used an LLM for this". I'm convinced that those cases happen more often than you think. Why not cover them with a simple mechanism? It's also in the interest of AI companies who don't want to train on AI output.

Depends on the pushback in different sets of users. Students for example would clean it up.

Re: How Claude marks AI-generated content

#62

Post by pixl97 » Mon, Aug 10, 2026, 11:17 PM UTC

case540 wrote: post #7

I don’t like the idea of hacking a response to contain a watermark. I also don’t like the idea of false positives detections coming directly from Anthropic. If people read more AI generated content, people will probably start writing more in that style

I have no idea why you were down voted for this. Language is alive and people adopt it from sources they hear a lot.

Re: How Claude marks AI-generated content

#63

Post by IshKebab » Mon, Aug 10, 2026, 11:18 PM UTC

benrow wrote: post #36

I've heard that this kind of watermarking process works by biassing the statistical sampling towards a partition of the set of possible next tokens (red set and green set), at each position. It might only be a slight nudge each time, but over a sequence of tokens, the likelihood of repeating the bias by chance is increasingly improbable. The bias is different for each position and follows a defined RNG, seeded someho…

If it's based on position mod 2, wouldn't inserting or deleting (or splitting/merging) words every now and then trivially defeat it?

If it is based on position mod 2 then wouldn't inserting/deleting (or splitting and merging) words every now and then defeat it?

Re: How Claude marks AI-generated content

#64

Post by pixl97 » Mon, Aug 10, 2026, 11:19 PM UTC

colesantiago wrote: post #44

Good. They should make it easier, to detect slop so we can ignore it quickly. I hope Pangram makes an API or an extension to analyze a page to detect slop on a page and then closes the tab immediately. Nobody should be wasting time on garbage LLM output in code, text, image and videos.

Panagram is a scam.

Re: How Claude marks AI-generated content

#65

Post by colesantiago » Mon, Aug 10, 2026, 11:22 PM UTC

pixl97 wrote: post #64

colesantiago wrote: post #44

Good. They should make it easier, to detect slop so we can ignore it quickly. I hope Pangram makes an API or an extension to analyze a page to detect slop on a page and then closes the tab immediately. Nobody should be wasting time on garbage LLM output in code, text, image and videos.

Panagram is a scam.

(This is the part where you provide extensive extraordinary evidence to your claim)

Re: How Claude marks AI-generated content

#66

Post by ethin » Mon, Aug 10, 2026, 11:28 PM UTC

benrow wrote: post #48

ethin wrote: post #37

Can someone help me understand how exactly this watermarking of text works? Given that text is, well, text, and not some kind of binary format, I don't see how any watermarking can work unless you insert characters which are invalid under Unicode. I further don't really understand how this won't be perceivable by assistive technology (the "watermark" will just appear as either unreadable characters, or if the waterma…

Have a look around for token biasing, or green lists. It's based on a nudge to the choice of the next token (which can always be drawn from a set of possibilities which are all probable enough). At first I thought this approach was just the "LLM flavour" of writing, but it's way more subtle, especially as the bias is applied uniquely for each token position.

Yeah, will do, this sounds interesting since I'm not entirely sure how this would actually be reliable to any degree. Thanks for the help, not sure why I got downvoted since I was genuinely curious.

Re: How Claude marks AI-generated content

#67

Post by akersten » Mon, Aug 10, 2026, 11:35 PM UTC

So my code that Claude makes, which previously was using the best (most probable) tokens for the job, will now be getting worse in random positions, to appease a voluntary EU suggestion. Love that.

Re: How Claude marks AI-generated content

#68

Post by tech234a » Mon, Aug 10, 2026, 11:40 PM UTC

Computer0 wrote: post #2

So this won't be happening in the US, but in the EU: " When a supported Claude model generates text, it weaves an imperceptible watermark directly into the text itself. You won’t see it, and it doesn’t change the meaning, quality, or readability of Claude’s response. Because the watermark is part of the text, it will travel with the text when it’s copied and pasted elsewhere, and may persist through some editing. Wat…

Article specifically says "worldwide"

Re: How Claude marks AI-generated content

#69

Post by pixl97 » Mon, Aug 10, 2026, 11:50 PM UTC

colesantiago wrote: post #65

pixl97 wrote: post #64

Earlier quoted context omitted.

Panagram is a scam.

(This is the part where you provide extensive extraordinary evidence to your claim)

No, they are the ones making claims, especially their CEO saying things like a 1/10000 false positive rate. Their own testing showed a 2% rate, which is insanely high when you talk about the number of papers students turn in. Worse their testing methodology compared it with pre-llm documents and not post llm documents that were human written (much harder and more expensive to verify), by treating language as static.

Re: How Claude marks AI-generated content

#70

Post by slowin » Tue, Aug 11, 2026, 12:00 AM UTC

simonw wrote: post #21

An interesting factor of this is competition. If Claude was the only model family they could ship a change like this and users who want to cheat (or don't like watermarks for other reasons) would just have to put up with it. In a world with many different competing models, the risk of losing customers to other providers over this is much more real. Maybe they've looked at the numbers and the portion of people who cle…

I’m more worried that this will degrade performance. I want the best results from a model, not the results that fit a constraint that’s not defined by me. Any increased cost or latency is also unacceptable.