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

411–420 of 446 posts

Re: How Claude marks AI-generated content

#411

Post by hojinkoh »

Dilettante_ wrote: post #120

I notice the "Limitations" section talks about how content only at some point touched by Claude may return a positive, and content that returns a negative may still be Claude generated. But I really would have liked for them to state explicitly that entirely false positives where a piece is fully human-written may still be marked as generated, because too many institutions with the power to ruin someone's life over t…

Sure, there are things you could do legally when falsely accused; and there are things authorities and companies should do.

But ultimately, when you are powerless and can't afford to do the fighting: I'm convinced the only way to protect yourself is to be very mindful about your writing style, and to deliberately corrupt the language through objectively wrong "stylistic elements".

Re: How Claude marks AI-generated content

#412

Post by infinite_spin »

toxik wrote: post #143
infinite_spin wrote: post #139

Earlier quoted context omitted.

My guess is it will be similar to how Genius watermarked lyrics, using things like variants of punctuation https://www.pcmag.com/news/genius-we-caught-google-red-hande...

In program code? Unlikely, surely¡

Consider these naming options:

> total = calculate(items)

> result = calculate(items)

> value = calculate(items)

> amount = calculate(items)

All of them are reasonable options. If we bias the model's output so that one of them is more likely than the others, then we can reconstruct that watermark if enough of these frames are present.

Re: How Claude marks AI-generated content

#413

Post by OrangeMusic »

no_multitudes wrote: post #319
OGWhales wrote: post #211

Earlier quoted context omitted.

And yet, it remains possible that a human could write the same sequence of characters.

How often do you add seemingly-random zero-width unicode characters to the text you write?

That's not how it works (that would be trivial to erase).

Re: How Claude marks AI-generated content

#414

Post by OrangeMusic »

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/

[dead]

Re: How Claude marks AI-generated content

#416

Post by a96 »

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.

They already have their own telltales. Might or might not be more avoidable and varied.

Re: How Claude marks AI-generated content

#417

Post by KoolKat23 »

KoolKat23 wrote: post #270
shawnz wrote: post #190

Earlier quoted context omitted.

There is already some intentional randomness in token selection, because it actually improves the quality of responses if you intentionally don't always pick the most likely next token. You can hide data in that randomness without impacting the quality of the response by using a sufficiently "random looking" pseudorandom bit stream instead of real random numbers. I previously worked on a project to do that here: http…

Good point, very interesting. Thanks!

https://x.com/Koolkat6000/status/2087502146556100812?s=20

A small anecdotal example to demonstrate how the response lacks depth. It's imperceptible to most. You'll probably see through it in your own field of expertise, if you had both answers. Uncanny valley type thing.

Re: How Claude marks AI-generated content

#418

Post by Dilettante_ »

matheusmoreira wrote: post #403
Dilettante_ wrote: post #125

Earlier quoted context omitted.

"This is my emotional support gun. It makes me feel safe despite my CPTSD and is therefore assistive technology."

Sorry, but it essentially cured my ADHD. In my experience, AI is more effective than lisdexamfetamine at allowing me to turn my ideas into reality. AI stigmatization is ableism.

I'm not saying you're wrong, much like a gun really would help a victim of CPTSD feel safe.

What I was trying to point to was that "this thing helps some people" does not equal "this thing is unequivocally Good and should be entirely unchecked".

I don't even care about the AI. I just get peeved by bad lines of argumentation.

Re: How Claude marks AI-generated content

#419

Post by DanielHB »

no_multitudes wrote: post #321
DanielHB wrote: post #144

Earlier quoted context omitted.

But what prevents someone from using Anthropic own detection system to train a watermark-scrubber? Seems like this would only catch the most unsophisticated cases.

Most of the people posting unedited LLM content all over the internet are unbelievably lazy.

I don't think so, I think most LLM content is bot generated and amending bots to remove watermarkers would be trivial if the bypass is trivial.

You are just pointing out the very visible single cases. But the mass of low-visible content is much higher and more dangerous (like propaganda bot-farms). If a social media platform adds watermarker checks the bots would implement bypasses ASAP.

Re: How Claude marks AI-generated content

#420

Post by hbn »

breezybottom wrote: post #199
hbn wrote: post #72

If the western AI companies are forced to comply with this type of BS, and develop their models to do their job while balancing a book on their head and hopping on one foot, the Chinese models just got a free pass to completely dominate the frontier. EU regulation does it again!

Less AI slop sounds like a win to me. Let China drown in it.

Are you using some version of the internet that's cut off from other countries?
Post reply on HN