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

81–90 of 817 posts

Re: Claude Code is steganographically marking requests

#81

Post by wolttam »

thih9 wrote: post #70
wolttam wrote: post #36

I used Claude Code for a month because my boss gifted me a sub and wanted me to try it. I used that month to complete a work project and then beef up my personal harness so I'd never have to deal with Anthropic (and these sorts of shenanigans) again.

How do people build something like a personal harness? Are there tools for that or is it done from scratch?

I started mine from scratch in 2023 because I wanted to use LLMs from a terminal and there was nothing else compelling at the time (nowadays there is pi and opencode)

Harnesses are/can be incredibly simple things, not much more than a HTTP client that renders things in a way that suites your taste.

Re: Claude Code is steganographically marking requests

#82

Post by SubiculumCode »

applfanboysbgon wrote: post #54
ahmedehab_01 wrote: post #41

Frankly, I don't see this as the concerning behaviour the article describes. It is fine to try to protect against distillation through a technique like this. This will also allow them to, instead of blocking the distillation agents, respond with a poorer result/model, hindering the progress of distillation, momentarily at least. I would guess that's their first line of defense; they should have more techniques to ide…

> This will also allow them to, instead of blocking the distillation agents, respond with a poorer result/model, i.e. this will allow them to literally commit fraud against paying customers

1st, this technique is not fraud, and fraud is a separate accusation. 2nd, paying customers can legally and legitimately be banned and monitored for breaking terms of service, which probably includes things like using the model against U.S. export restrictions.

Re: Claude Code is steganographically marking requests

#83

Post by hakunin »

thih9 wrote: post #70
wolttam wrote: post #36

I used Claude Code for a month because my boss gifted me a sub and wanted me to try it. I used that month to complete a work project and then beef up my personal harness so I'd never have to deal with Anthropic (and these sorts of shenanigans) again.

How do people build something like a personal harness? Are there tools for that or is it done from scratch?

Not the comment author, but I use pi and customize it with my own extensions. Pi automatically tells models how to customize itself, so it's a pretty easy process.

Re: Claude Code is steganographically marking requests

#84

Post by nowittyusername »

thih9 wrote: post #70
wolttam wrote: post #36

I used Claude Code for a month because my boss gifted me a sub and wanted me to try it. I used that month to complete a work project and then beef up my personal harness so I'd never have to deal with Anthropic (and these sorts of shenanigans) again.

How do people build something like a personal harness? Are there tools for that or is it done from scratch?

Build it from scratch. Understanding fundamentals of how agentic coding harnesses is a must though if you gonna go that route. I think everyone should take time and learn these things, maybe reverse engineer Codex Cli or something like that as a starter. That info is very valuable in this day and age.

Re: Claude Code is steganographically marking requests

#85

Post by echelon »

thih9 wrote: post #70
wolttam wrote: post #36

I used Claude Code for a month because my boss gifted me a sub and wanted me to try it. I used that month to complete a work project and then beef up my personal harness so I'd never have to deal with Anthropic (and these sorts of shenanigans) again.

How do people build something like a personal harness? Are there tools for that or is it done from scratch?

Why use a personal harness?

You have to pay API pricing, which is far more costly.

I'd either switch to GLM wholesale or just continue to use Opus within Claude Code as the blessed, subsidized path.

Re: Claude Code is steganographically marking requests

#86

Post by ipsod »

wolttam wrote: post #69
tonmoy wrote: post #43

Earlier quoted context omitted.

What models are you using? Aren’t you still dealing with some provider even if you are not using their binary

I self-host DeepSeek V4 Flash on 2 DGX Sparks (approx. $10k) I expect DeepSeek V4 Flash (or an equivalently sized model) to reach parity with GLM 5.2 some time this year (this based on DeepSeek V4 Flash launching at GLM 5.0 parity[0], and GLM 5.2 being freely available to distill from) GLM 5.2 is within spitting distance of Opus 4.8 and is at least as good as Opus 4.6[1] which some devs were willing to spend hundreds…

How fast is it?

Re: Claude Code is steganographically marking requests

#87

Post by krupan »

wolttam wrote: post #36

I used Claude Code for a month because my boss gifted me a sub and wanted me to try it. I used that month to complete a work project and then beef up my personal harness so I'd never have to deal with Anthropic (and these sorts of shenanigans) again.

Given the Anthropic shenanigans, do you trust the personal harness code it wrote for you?

Re: Claude Code is steganographically marking requests

#88

Post by axutio »

palmotea wrote: post #48
axutio wrote: post #30

Earlier quoted context omitted.

Would you also say that "someone who wants to use an IDE / LSP features to code and not give credit to the IDE / LSP is the worst kind of person"? If not, what is the difference between the two for you?

> Would you also say that "someone who wants to use an IDE / LSP features to code and not give credit to the IDE / LSP is the worst kind of person"? That's a false equivalency. > If not, what is the difference between the two for you? Let's start this out right: if they're equivalent, first you explain to us why you think so.

I just don't agree that it's a false equivalency. I see them both as "tools I use to get the job done". For me, the job is not "writing code" - it is "deliver feature", "fix bug", and the accountability, responsibility, and communication that comes with it.

Re: Claude Code is steganographically marking requests

#89

Post by thewebguyd »

Maken wrote: post #37
atonse wrote: post #14

[flagged]

If scrapping content is legal, model distillation should be legal too.

I suppose model distillation is technically legal, in terms of copyright, because LLM output is automatically public domain.

It's only "illegal" from a standpoint of breach of contract given its against the terms of use/service, which is to say its not illegal at all, there's no criminality there.

Re: Claude Code is steganographically marking requests

#90

Post by selfhoster11 »

conception wrote: post #17
theplumber wrote: post #6

The more I learn about Anthropic the more they disgust me. Finger crossed for all the companies from their “ban list”

Which AI company have you learned more about where you liked them more as more details came out?

Moonshot.
Post reply on HN