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

Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

daringfireball.net

Post reply on HN
Print view View original

661–670 of 776 posts

Re: Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

#661

Post by cindyllm »

inigyou wrote: post #536
summarybot wrote: post #508

Earlier quoted context omitted.

Copyright law was updated in a very helpful way in the last twenty years sometime so that as soon as you post something to the internet you have copyright. If you need a citation don't hesitate to ask someone else.

So you can just say some text is copyright to you and it is? Here is some text that is copyright to me. As you infringed my copyright, please pay my $5000 license fee for every user who has read it: > Copyright law was updated in a very helpful way in the last twenty years sometime so that as soon as you post something to the internet you have copyright. If you need a citation don't hesitate to ask someone else.

[dead]

Re: Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

#662

Post by micromacrofoot »

ghrl wrote: post #246

My biggest concern is that checking any text for watermarks requires sending the entire text to Anthropic. And even that is not sufficient, as the text might have been generated with ChatGPT, Gemini, Grok, Mistral, ... So every check requires sending the text to as many AI providers as offer a watermarking detection API, almost all of which have a very dubious track history with obtaining training data through illici…

You've discovered the perpetual mutually assured destruction money generator — guess what the best defense against LLM spam also is?

LLMs cause a wide number of problems, the good news for our investors is that they're all solvable with LLMs.

Re: Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

#665

Post by mholm »

npilk wrote: post #86
mholm wrote: post #39

Earlier quoted context omitted.

This was true in the ChatGPT era. Now we're in a world with reasoning tokens, where a model can thoroughly plan out the response it wants to make. If anything, it makes the style worse.

Yes, models can reason and plan, which helps them write more coherently. But when they write the final output, it’s still a single generation. It would be like letting a human make notes and write an outline, but not let them use the backspace once they start typing their response. Presumably you could use the same reasoning trace, run multiple generations, and get different outputs (if the temperature is >0). But no…

There's been a lot of effort into the writing space, and the models genuinely prefer this style. You can let them iterate on the same idea 100 times, rewrite sentences, determine what works best — and they'll still verb the noun, do rule of 3, and keep the same monotonous structure.

Re: Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

#667

Post by mossTechnician »

summarybot wrote: post #505
nemo wrote: post #376

Earlier quoted context omitted.

>A lot of people use Claude as a friend/therapist/romantic partner People developing a para-social (pseudo-social?) relationship with a corporate robot have far bigger problems than the word-chooser in their robot "friend".

Very intelligent people need intelligent-others to bounce ideas off of, and the LLM can be that.

Even intelligent people have demonstrated they are not immune to the damaging effects of AI sycophancy. The most recent example that comes to mind is Hank Green.

Re: Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

#668

Post by cadamsdotcom »

inigyou wrote: post #540
cadamsdotcom wrote: post #418

Earlier quoted context omitted.

That could be how it works, but in practice it takes into account all previous tokens when producing the next-token distribution to sample from. So a switch back is more likely than your explanation supposes.

No, if you switched to Japanese the LLM wouldn't ignore it, it would "assume" there's a reason for that. The same if the previous iteration of the LLM switched to Japanese. Else you're expecting an LLM to ignore its own previous outputs and restart "thinking" from scratch with every token?

It's situational and I suspect there are situations where it would and others where it wouldn't. Would depend on the almost infinite variables of how training was done. You'd be right that it'd be likely to switch but while it's possible it's due to temperature, there are just so many things going on. But it would be one sensible explanation among many.

Re: Anthropic's ‘watermark’ text adulteration in Claude is a perversion of writing

#669

Post by ComputerGuru »

kergonath wrote: post #496
ghrl wrote: post #246

My biggest concern is that checking any text for watermarks requires sending the entire text to Anthropic. And even that is not sufficient, as the text might have been generated with ChatGPT, Gemini, Grok, Mistral, ... So every check requires sending the text to as many AI providers as offer a watermarking detection API, almost all of which have a very dubious track history with obtaining training data through illici…

> And even that is not sufficient, as the text might have been generated with ChatGPT, Gemini, Grok, MistraL I would think that these things would eventually converge and we’d get one watermarking algorithm as an industry standard. That way, all major provider would follow it and we’d get independent software for checking. This would partly limit the efficacy of the watermarks, but on the other hand if it’s done corr…

You can’t check it with only the algorithm, you need the secret seed key. Which will be different for each provider (and they’ll probably have and use multiple). And you need the llm itself, to generate the potential tokens at each step.
Post reply on HN