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

151–160 of 446 posts

Re: How Claude marks AI-generated content

#151

Post by limbicsystem »

Computer0 wrote: post #28
toufka wrote: post #18

Earlier quoted context omitted.

I'm guessing - probably some textual variation on Benford's law? [1]. Trivial for compute, painful for a human. - "Ensure distribution of vowels is in >99th percentile of human work" - "Ensure the distribution of the letter "s" is within 99th percentile of human work" - "Ensure the distribution of the letter "L" is periodic with periodicity within 5% of 1/N characters. - "Ensure there is a cross-linguistic 'typo' (co…

I would hate to have any of these rules effecting my output

I see what you did there!

Re: How Claude marks AI-generated content

#152

Post by s_dev »

sixtyj wrote: post #88
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…

It was quick :) … https://claudewatermarkremover.app/

So was it going down:

"Neutralize engine is temporarily unavailable. Try again."

Re: How Claude marks AI-generated content

#153

Post by pshirshov »

Well I wonder how would it respond to now. The correlations should be traceable with a similar technique. Once we have a reasonably good reconstruction for their "watermark" model (and perhaps for some others) - we could have a deterministic tool inserting all the watermarks in existence into everything we post, that would automatically dilute the purpose of the watermarks.

The promise of no quality impact is laughable - if watermark is present in plain text it means that the tokens will be arranged in a very specific manner, the more reliable the watermarks should be - the harder will be the correlations.

Don't forget how annoyingly bad Anthropic products have become in recent releases - low adherence, annoying alignment, annoying guardrail false-positives, unwarranted checkpoints - all that shit. Now they deliver more crap.

Re: How Claude marks AI-generated content

#155

Post by TheOtherHobbes »

w_for_wumbo wrote: post #34
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…

What happens if someone handwrites a Claude output, then someone uses that handwritten text as a reference. Now you've got a watermarked idea which may have no direct linkage to the usage of Claude.

If the algos work as advertised, watermarked token sequences have an extremely low probability. Copying the words by hand doesn't change that.

The mechanism seems to survive editing. The extreme probabilities get a little less extreme, but are still extreme enough to be distinctive.

But it wouldn't survive paraphrasing, because the output would be entirely human and the token correlations would disappear.

It might not survive referencing if only a sentence or two is used.

The practical issue is how true the claims are. It's one thing to create a proof of concept, another to see how it works in use.

And this is potentially catastrophic for code, because the grammar and word choices of code are completely different and more fragile than standard English.

Re: How Claude marks AI-generated content

#156

Post by wpietri »

embedding-shape wrote: post #150
Filligree wrote: post #148

Earlier quoted context omitted.

For watermarked long-form text, it is actually possible. Makes the watermark more fragile, but the math is considerably more forgiving than usual.

> For watermarked long-form text What about "watermarked long-form code"? I'm having a hard time understanding how a model could watermark not prose, but functional/semantic text like code, that actually has meaning. You can't switch our the characters, you can't use various types of whitespace, you can't add arbitrary code comments, and a lot of other restrictions. Is there any state of the art methods for watermark…

As anybody who has put together a coding standard knows, there are a lot of options for individual expression, meaning a lot of room for things like watermarking. And of course you can add arbitrary comments; my Claude-generated code is very verbose.

Re: How Claude marks AI-generated content

#157

Post by wpietri »

thunfischtoast wrote: post #92
COAGULOPATH wrote: post #80

Earlier quoted context omitted.

>I'd like to know a lot more about how that works. My guess is that it works like Gemini's SynthID: by altering the logprobs of the next token. Like, for every 10th token, instead of outputting the most probable, it outputs the 17th most probable, or something. (Obviously it's way more complicated but I think conceptually this is how it works.) No human will notice this, but a classifier trained on Claude's output wi…

They still need to choose when to do that though. When I prompt the program to e.g. alter a bash script in a specific way or to recite a longer known text it can't go round and randomly exchange tokens. It has to somehow define what is a simple repeated text from a different origin and what is a novel generation.

I would guess they're not worrying about watermarking a tweak to a human-written program. That's both a tiny fraction of Claude use and of very little concern to the kinds of people who want to check watermarks.

Re: How Claude marks AI-generated content

#158

Post by embedding-shape »

wpietri wrote: post #156
embedding-shape wrote: post #150

Earlier quoted context omitted.

> For watermarked long-form text What about "watermarked long-form code"? I'm having a hard time understanding how a model could watermark not prose, but functional/semantic text like code, that actually has meaning. You can't switch our the characters, you can't use various types of whitespace, you can't add arbitrary code comments, and a lot of other restrictions. Is there any state of the art methods for watermark…

As anybody who has put together a coding standard knows, there are a lot of options for individual expression, meaning a lot of room for things like watermarking. And of course you can add arbitrary comments; my Claude-generated code is very verbose.

> there are a lot of options for individual expression, meaning a lot of room for things like watermarking

The way I use LLMs (and I'd advice everyone to do the same) there really isn't, the agent implements things exactly how I want them, or I use the agent to massage it into the exact bit-by-bit version I imagined when I first sent the prompt afterwards. I honestly don't know what the point would be to let the agents compose worse code than what I'd do manually, although I know it's a popular approach taken by many.

> And of course you can add arbitrary comments; my Claude-generated code is very verbose.

So watermarking for all users who allow code comments from agents, no watermarking for us who force the agents to never write a single code comment? Alright, I'd be fine with that.

Re: How Claude marks AI-generated content

#159

Post by junofan »

embedding-shape wrote: post #150
Filligree wrote: post #148

Earlier quoted context omitted.

For watermarked long-form text, it is actually possible. Makes the watermark more fragile, but the math is considerably more forgiving than usual.

> For watermarked long-form text What about "watermarked long-form code"? I'm having a hard time understanding how a model could watermark not prose, but functional/semantic text like code, that actually has meaning. You can't switch our the characters, you can't use various types of whitespace, you can't add arbitrary code comments, and a lot of other restrictions. Is there any state of the art methods for watermark…

[dead]

Re: How Claude marks AI-generated content

#160

Post by wpietri »

DaSHacka wrote: post #123
cassianoleal wrote: post #56

Earlier quoted context omitted.

I hope all models adopt it.

Thankfully, there are a variety of Chinese models that never will. I think we all know that in a few years, they will also be the only relevant offerings on the market, due to not being bogged down with over-zealous ""safety"" footguns.

Your theory is that the Chinese government is thoroughly uninterested in safety or prosocial controls?
Post reply on HN