Top
Best
New

Posted by mfiguiere 13 hours ago

How Claude marks AI-generated content(support.claude.com)
185 points | 145 commentspage 2
drnick1 12 hours ago|
Seems like an awful idea. I hope that that "watermark" will soon be discovered, reverse-engineered, and that tools to remove it will appear.
cassianoleal 12 hours ago|
I hope all models adopt it.
DaSHacka 2 hours ago|||
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.
wpietri 22 minutes ago||
Your theory is that the Chinese government is thoroughly uninterested in safety or prosocial controls?
hparadiz 12 hours ago|||
Other models will end up diffusing it and making the signal indeterministic and irrelevant.
lorenzohess 13 hours ago||
> Generated text will carry embedded watermarks, and generated files will include digitally signed provenance metadata where supported.

This should make it easier to catch cheaters who use Claude, right? Unless everyone runs their artifacts through some watermark and metadata sanitizer?

uncivilized 13 hours ago||
As long as they’re in the EU.
nonestdeus 12 hours ago||
From the linked article

> Regions. Marking will apply to output from supported models wherever Claude is offered, worldwide.

bossyTeacher 13 hours ago|||
> Unless everyone runs their artifacts through some watermark and metadata sanitizer?

It will happen if Claude tampers the text. Guaranteed.

pixl97 12 hours ago||
Text is too low bandwidth to classify reliably without lots of false positives. Especially as people start talking like LLMs.
reasonableklout 3 hours ago||
The approach Pangram has taken which works pretty well is to simply lower the recall a lot but ensure the precision is very high. Which means potentially high false negative rate but low false positive rate.
dejanseo 2 hours ago||
> "Claude models launched on or after August 2, 2026 support marking at launch."

No Anthropic model has been launched in August.

dalemhurley 12 hours ago||
People with dyslexia and dystrophia, commonly use LLMs to proofread content. Even Anthropic admits this is a limitation.
stranded22 4 hours ago||
Yes, I’m audhd and dyslexic.

I am cancelling my Claude max 5x subscription and moving to ChatGPT pro. I have difficulty enough trying to ensure my meaning comes through correctly, along with everything else; to now have to look out for/analyse watermarks too?

I feel shamed enough by society, thanks Anthropic.

Mashimo 2 hours ago||
Don't you think OpenAI will do this too soon?
bramhaag 2 hours ago|||
How exactly does this impact proofreading? You can manually apply the suggestions (typo here, unnatural sounding sentence there, etc.) the LLM gives you to your own content, and it would stay watermark-free.

Unless with "proofreading" you actually mean having the LLM write your content for you.

matheusmoreira 5 hours ago||
People with executive dysfunction too. LLMs bring execution costs down to near zero and are therefore assistive technology.
Dilettante_ 2 hours ago||
"This is my emotional support gun. It makes me feel safe despite my CPTSD and is therefore assistive technology."
case540 13 hours ago||
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
stranded22 4 hours ago||
The amount of times ‘delve’ appeared in general conversation in the last couple of years shows the influence LLMs have on society.
pixl97 11 hours ago|||
I have no idea why you were down voted for this. Language is alive and people adopt it from sources they hear a lot.
ack_complete 9 hours ago||
Moreover, what if you quote text that happens to have been generated by Claude, does that bump up the AI-ness score of your source file or document?
reasonableklout 3 hours ago||
The flip side of this is that if AI-generated content becomes reliably identifiable and carries a stigma, then people might deliberately change their styles to be more diverse and human.

One example I've seen are junior employees at my company deliberately adopting a lowercase/less punctuation writing style so as to stand apart from AI.

hparadiz 12 hours ago||
I know you're gonna read this so I'll be blunt. This is bad for your brand.
LEDThereBeLight 10 hours ago|
Reactionary emotional advice does no good, it just makes people want to hold their positions more defensively. If you care enough to say something, care enough to say it with reasons that might shift someone’s perspective.
lukewarm707 12 hours ago||
no thanks.
jp0001 12 hours ago||
OpenAI has been watermarking their images with C2PA for some time.
hbn 10 hours ago||
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!

Mashimo 2 hours ago||
If the Chinese want to sell to EU customers, they probably have to do the same.
DimitriBouriez 2 hours ago||
What's the problem, really? Given the direction the U.S. has been heading in recent years, I wonder what really sets it apart from China. Europe needs to maintain an equal distance from both the U.S. and China.
amelius 13 hours ago|
They should just replace the spaces by one of Unicode special space characters.

Can it be circumvented? Of course. Will most people go through the trouble to circumvent it? No.

A_D_E_P_T 13 hours ago||
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.

amelius 12 hours ago|||
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.
pixl97 11 hours ago||
Depends on the pushback in different sets of users. Students for example would clean it up.
amelius 2 hours ago||
Sure, but let's first find out how many % of people are willing to be frank about their AI usage, and/or don't care about it. My guess is it is worthwhile to do this.
selcuka 6 hours ago|||
But the source codes of those web apps will also be watermarked. /s
RataNova 3 hours ago|||
Those invisible spaces get wiped by the first sanitizer in any normal ide. Worse it'll instantly break parsing for configs like yaml where spaces are critical for structure
sixtyj 13 hours ago||
https://unicode-explorer.com/articles/space-characters

U+2800 or U+3164 would be nice.

But as I remove unwanted characters with grep before layout in InDesign, someone will make a skill for removing such space characters.

ack_complete 12 hours ago||
We already have one, our Claude setup already requires output to be 7-bit ASCII clean and scans it for such.
More comments...