Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How Claude marks AI-generated content

support.claude.com

Post reply on HN
Print view View original

41–50 of 446 posts

Re: How Claude marks AI-generated content

#42

Post by mihaelm »

simonw wrote: post #15

> 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. I'd like to know a lot more about how that works. A lot of my interactions with Claude return pretty precise text. If I ask it to edit a project and refactor a specific function in several places I know ex…

> have some kind of weird pattern baked into their text to act as a watermark.

public abstract class BaseAnimalBeanFactoryGeneratedFromClaudeFactory

Re: How Claude marks AI-generated content

#44

Post by colesantiago »

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.

Re: How Claude marks AI-generated content

#45

Post by resonantjacket5 »

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…

it's a statistical way. like for example maybe in your above paragraph claude maybe writes "Thus, I don't see how this wouldn't be insanely (instead of trivial) to remove" and then also says like "And this is before we things being put on the clipboard." or maybe the i just says the word "the" in a certain pattern or frequency.

you can then consistently like figure out if it was claude that wrote the sentence. it is easy as you noted if you just get another ai to read it and then rewrite it.

Re: How Claude marks AI-generated content

#46

Post by jbs789 »

baq wrote: post #29
simonw wrote: post #15

> 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. I'd like to know a lot more about how that works. A lot of my interactions with Claude return pretty precise text. If I ask it to edit a project and refactor a specific function in several places I know ex…

> I'd like to know a lot more about how that works. Count load-bearing words using two different algorithms in a belt-and-braces fashion

Fair - I should have been honest about the watermark.

Re: How Claude marks AI-generated content

#47

Post by LoganDark »

pjm331 wrote: post #17
aabhay wrote: post #6

I have had a hunch for a while now that (in addition to these tools), Anthropic has actually leaned in to Claude's distinctive manner of writing since it makes the text more obviously AI generated and thus less susceptible to misuse. That's not necessarily the same thing as a markov-style fingerprint but it could be a correlated factor.

I had a similar thought but I assumed they leaned in because it improved performance on coding or something like that

I suspect it's because of alignment concerns. The more deeply they can integrate their principles, the harder it'll be to misuse. Or at least that's the idea.

Re: How Claude marks AI-generated content

#48

Post by benrow »

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.

Post reply on HN