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

Relicensing with AI-Assisted Rewrite

tuananh.net

Post reply on HN
Print view View original

351–360 of 408 posts

Re: Relicensing with AI-Assisted Rewrite

#351

Post by popalchemist »

calny wrote: post #260

The maintainer's response: https://github.com/chardet/chardet/issues/327#issuecomment-4... The second part here is problematic, but fascinating: "I then started in an empty repository with no access to the old source tree, and explicitly instructed Claude not to base anything on LGPL/GPL-licensed code." Problem - Claude almost certainly was trained on the LGPL/GPL original code. It knows that is how to solve the prob…

Copyright does not cover ideas. Only specific executions of ideas. So unless it's a line-by-line copy (unlikely) there is no recourse for someone to sue for a re-execution/reimplementation of an idea.

Re: Relicensing with AI-Assisted Rewrite

#352

Post by foota »

msdz wrote: post #76
foota wrote: post #57

Earlier quoted context omitted.

I don't know how far it would get, but I imagine that a FAANG will be able to get the farthest here by virtue of having mountains of corporate data that they have complete ownership over.

They’d probably get the farthest, but they won’t pursue that because they don’t want to end up leaking the original data from training. It is possible in regular language/text subsets of models to reconstruct massive consecutive parts of the training data [1], so it ought to be possible for their internal code, too. [1] https://arxiv.org/abs/2601.02671

Copyright for me not for thee? :) That's a good point though. Maybe they could round trip things? E.g., use the model trained only on internal content to generate training data (which you could probably do some kind of screening to remove anything you don't want leaking) and then train a new model off just that?

Re: Relicensing with AI-Assisted Rewrite

#353

Post by postalrat »

DannyBee wrote: post #333
DrammBA wrote: post #6

I like the idea of AI-generated ~code~ anything being public domain. Public data in, public domain out.

Lawyer here. Its not. This article is highly confused. The case was about whether an AI could be considered an author for copyright purposes. Mainly as a way of arguing for robot rights, not copyright. The person listed the AI as the sole author: On the application, Dr. Thaler listed the Creativity Machine as the work’s sole author and himself as just the work’s owner. This is not the first time someone tried to say…

"the person who created, operated, or used artificial intelligence" so which one is it? because there the person(s) who created the ai is almost always different that the person who used it.

Re: Relicensing with AI-Assisted Rewrite

#354

Post by josephg »

amelius wrote: post #192
josephg wrote: post #183

Earlier quoted context omitted.

> It is also not fair to claim that an AI-copy is fundamentally different from photocopying. If you clean-room copy it, I think it is different. Eg, first get one agent to make a complete spec of what the program does. And a list of all the correctness guarantees it meets. Then feed that spec into another AI model to generate a program which meets that spec. The second program will not be based on any of the code in…

The output will still be dependent on the input. And it is still copying even if you first lift the input to a different abstraction level. Finally, even if your rationale is 99% correct, then there is still that 1% that makes the result a mechanistic copy. And I see no way in which most people would 100% agree with your view.

If you write program A that does something, and I look at what your program does and write program B that does the same thing, have I copied your program? So long as I didn’t copy any of the lines of code in program A directly, no. At least, not according to copyright law. A copyright on Netscape navigator doesn’t apply to internet explorer or chrome. They’re all “copies” of Netscape navigator. But copyright applies to the work. New work? New copyright. I really don’t see how an LLM being involved changes any of that.

If you want to protect the idea or the design, get a patent. A patent on one h264 encoder applies to all h264 encoders.

There is a chance the courts or the legislature will decide differently. But until then, we should assume the existing law of the land holds.

Re: Relicensing with AI-Assisted Rewrite

#355

Post by fc417fc802 »

larodi wrote: post #337
jcranmer wrote: post #335

Earlier quoted context omitted.

SCOTUS hasn't ruled on any AI copyright cases yet. But they've said in Feist v Rural (1991) that copyright requires a minimum creative spark. The US Copyright Office maintains that human authorship is required for copyright, and the 9th Circuit in 2019 explicitly agreed with the law that a non-human animal cannot hold any copyright. Functionally speaking, AI is viewed as any machine tool. Using, say, Photoshop to dra…

The lack thereof (of human use). Prompts are not copyrightable thus the output also - not. Besides retelling a story is fair use, right? Otherwise we should ban all generative AI and prepare for Dune/Foundation future. But we not there, and we perhaps never going to be. So the LLM training first needs to be settled, then we talk whether retelling a whole software package infringes anyone's right. And even if it does,…

> Prompts are not copyrightable

Surely that varies on a case by case basis? With agentic coding the instructions fed in are often incredibly detailed.

Re: Relicensing with AI-Assisted Rewrite

#356

Post by 0x457 »

tzs wrote: post #340
galaxyLogic wrote: post #300

Earlier quoted context omitted.

"... If AI-generated code cannot be copyrighted (as the courts suggest) ". So, Supreme Court has said that. AI-produced code can not be copyrighted. (Am I right?). Then who's to blame if AI produces code large portions of which already exist coded and copyrigted by humans (or corporations). I assume it goes something like this: A) If you distribute code produced by AI, YOU cannot claim copyright to it. B) If you dist…

The Supreme Court has not ruled on this issue. An appeal of a lower court's ruling on this issue was appealed to the Supreme Court but the Supreme Court declined to accept the case. The Supreme Court has "original jurisdiction" over some types of cases, which means if someone brings such a case to them they have to accept it and rule on it, and they have "discretionary jurisdiction" over many more types of cases, whi…

But this means code generated by snippet expanders or any sort of templates is non-copyrightable.

Re: Relicensing with AI-Assisted Rewrite

#357

Post by pmarreck »

ostacke wrote: post #166
pmarreck wrote: post #159

Earlier quoted context omitted.

> With LLMs, the probability is much higher (since in truth they are very much not a "clean room" at all). I beg to differ. Please examine any of my recent codebases on github (same username); I have cleanroom-reimplemented par2 (par2z), bzip2 (bzip2z), rar (rarz), 7zip (z7z), so maybe I am a good test case for this (I haven't announced this anywhere until now, right here, so here we go...) https://github.com/pmarrec…

Bu that's not really what danlitt said, right? They did not claim that it's impossible for an LLM to generate something different, merely that it's not a clean room implementation since the LLM, one must assume, is trained on the code it's re-implementing.

Well, if you’re coding it in Zig, and it’s barely seen any Zig, then how exactly would that argument hold up in that case?

Re: Relicensing with AI-Assisted Rewrite

#358

Post by pmarreck »

sarchertech wrote: post #339
galaxyLogic wrote: post #304

Earlier quoted context omitted.

BUt LLM has seen millions (?) of other code-bases too. If you give it a functional spec it has no reason to prefer any one of those code-bases in particular. Except perhaps if it has seen the original spec (if such can be read from public sources) associated with the old implementation, and the new spec is a copy of the old spec.

Yes if you are solving the exact problem that the original code solved and that original code was labeled as solving that exact problem then that’s very good reason for the LLM to produce that code. Researchers have shown that an LLM was able to reproduce the verbatim text of the first 4 Harry Potter books with 96% accuracy.

This is not an argument against coding in a different language, though. It would be like having it restate Harry Potter in a different language with different main character names, and reshuffled plot points.

Re: Relicensing with AI-Assisted Rewrite

#359

Post by fc417fc802 »

zabzonk wrote: post #182
bandrami wrote: post #171

Earlier quoted context omitted.

Different but still derivative

Well, I am not exactly a hotshot 8086 programmer (though I do alright) but if I was asked to reproduce the IBM BIOS (which I have seen) I think I would come up with something very similar but not identical - it is really not rocket science code, so the LLM replacing me would have rather few alternatives to choose from.

I believe those are actually separate matters. A proper clean room implementation on the one hand, and the question of whether or not a particular outcome was a foregone conclusion on the other. I don't recall where I saw the latter but it might have come up during Google v Oracle?

Re: Relicensing with AI-Assisted Rewrite

#360

Post by pmarreck »

airza wrote: post #163
pmarreck wrote: post #159

Earlier quoted context omitted.

> With LLMs, the probability is much higher (since in truth they are very much not a "clean room" at all). I beg to differ. Please examine any of my recent codebases on github (same username); I have cleanroom-reimplemented par2 (par2z), bzip2 (bzip2z), rar (rarz), 7zip (z7z), so maybe I am a good test case for this (I haven't announced this anywhere until now, right here, so here we go...) https://github.com/pmarrec…

By what means did you make sure your LLM was not trained with data from the original source code?

Because it’s written in an entirely different language, which makes this whole point moot
Post reply on HN