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

Claude Code is steganographically marking requests

thereallo.dev

Post reply on HN
Print view View original

391–400 of 817 posts

Re: Claude Code is steganographically marking requests

#391

Post by jgalt212 »

meowface wrote: post #92

Value judgment aside: I am a bit surprised at how sloppily they did this. I think they could've achieved the same effect while decreasing the odds of detection via reverse engineering. (This field is known as "underhanded code", coined by the Underhanded C contest: https://www.underhanded-c.org . It's a little-known "art"; little-known for probably self-explanatory reasons. There are much cleverer ways of achieving o…

> they knew the JS bundle gets so heavily scrutinized that it'd eventually get spotted and reported on

Most likely someone did and raised the issue but they're moving too fast to fix these things before clicking deploy.

Re: Claude Code is steganographically marking requests

#394

Post by HDBaseT »

yorwba wrote: post #191
writeslowly wrote: post #141

Earlier quoted context omitted.

The site collection seems pretty random. There's a mix of actual AI labs, extremely questionable resellers (like whatever "claude-opus.top" is), and then random consumer sites like baidu and xiaohongshu.

Baidu has an actual AI lab: https://huggingface.co/baidu So does Xiaohongshu: https://huggingface.co/rednote-hilab Pretty much every Chinese internet company seems to have an AI team nowadays, however small. In addition, many Chinese companies are trying to give their programmers access to Anthropic models even though they're legally prohibited from doing so. And that might involve employees using unmodified Claude C…

Baidu has been doing some interesting things in the AI space though, the 'Unlimited OCR' model is very good.

Re: Claude Code is steganographically marking requests

#395

Post by AussieWog93 »

dools wrote: post #309
mrshadowgoose wrote: post #215

The conclusion of this blog post is a bit hysterical. The intent of this steg is excruciatingly clear (identifying usage by Chinese firms that may be conducting model distillation). It's unclear on how this "punishes normal developers" in any shape or form.

Why would a Chinese firm distilling the product use Claude code?

I'd assume cost? Claude Code plans give like $5,000 worth of API usage for $200/mo.

Re: Claude Code is steganographically marking requests

#396

Post by Philpax »

bayindirh wrote: post #291
gleenn wrote: post #279

Earlier quoted context omitted.

Whether or not you find Anthropic's behavior bad, theybhave been very loudly stating the foreign labs have been distilling their models for a while now. This seems like an obvious response to me that would be a mechanism to make that obvious.

From my understanding, distilling the model with another model is not illegal per se. Also, the output of the LLM is public domain by law, too. So, why all this "effort" to protect the model? This is a free market, and moving fast and breaking things is the norm. If they are so adamant on protecting their IP, maybe they can start by respecting others' IP, so we can start talking about ethics, equality and playing fai…

It's not illegal to distil the traces, but it is also not illegal for them to try to stop it.

Re: Claude Code is steganographically marking requests

#397

Post by solenoid0937 »

scott01 wrote: post #390
solenoid0937 wrote: post #381

Earlier quoted context omitted.

Ah yes, the very hysterical take that superintelligence must be carefully developed, or it likely ends badly for all of us. How irrational and hysterical of me!

Projection. Your account posts mostly negative or passive aggressive comments.

I suppose I'm sick of armchair commentators on AI safety that haven't actually done any critical thinking on AI safety.

Re: Claude Code is steganographically marking requests

#398

Post by ethbr1 »

notnullorvoid wrote: post #308
gleenn wrote: post #279

Earlier quoted context omitted.

Whether or not you find Anthropic's behavior bad, theybhave been very loudly stating the foreign labs have been distilling their models for a while now. This seems like an obvious response to me that would be a mechanism to make that obvious.

> loudly stating the foreign labs have been distilling their models for a while now. They would be stating this even if it weren't true, because it fits their marketing. While I don't disbelieve the claim outright, I highly suspect Anthropic is misleading everyone about the severity.

Distillation usage still burnishes usage numbers for IPO...

If anything, Anthropic is incentivized to track but do nothing until equity lock up expires.

Re: Claude Code is steganographically marking requests

#399

Post by yieldcrv »

meowface wrote: post #92

Value judgment aside: I am a bit surprised at how sloppily they did this. I think they could've achieved the same effect while decreasing the odds of detection via reverse engineering. (This field is known as "underhanded code", coined by the Underhanded C contest: https://www.underhanded-c.org . It's a little-known "art"; little-known for probably self-explanatory reasons. There are much cleverer ways of achieving o…

Anthropic is very transparent about their code being AI slop

they spend their resources on compute and the model itself, the company is carried by the model and software engineers babysitting it

Re: Claude Code is steganographically marking requests

#400

Post by ethbr1 »

next_xibalba wrote: post #302
cowl wrote: post #295

Earlier quoted context omitted.

oh no, the company that illegally used every possible media they could get their hands on is crying that some other company is doing something potentially shady but not illegal? And using that excuse to put in place hidden surveillance systems on their customers?

People keep throwing this idea around haphazardly, but U.S. courts have pretty consistently decided that training on copyrighted works falls under fair use. You may not like it, but that doesn't make it "illegal".

> that training on [lawfully obtained] copyrighted works falls under fair use

Fixed that for you.

Post reply on HN