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

We've filed a lawsuit against GitHub Copilot

githubcopilotlitigation.com

Post reply on HN
Print view View original

441–450 of 824 posts

Re: We've filed a lawsuit against GitHub Copilot

#441

Post by varajelle »

Aeolun wrote: post #414
xtracto wrote: post #337

Earlier quoted context omitted.

Say you publish a song and copyright it. Then I record it and save it in a .xz format. It's not an MP3, it is not an audio file. Say I split it into N several chunks and I share it with N different people. Or with the same people, but I share it at N different dates. Say I charge them $10 a month for doing that, and I don't pay you anything. Am I violating your copyright? Are you entitled to do that? To make it funni…

What you are actually giving people is a set of chords that happen to show up in your song, the machine can suggest an appropriate next chord. It’s also smart enough to rebuild your song from the chords _if you ask it to_.

I take your code and I compress it in a tar.gz file. Il call that file "the model". Then I ask an algorithm (Gzip) to infer some code using "the model". The algorithm (gzip) just learned how to code by reading your code. It just happened to have it memorized in its model.

Re: We've filed a lawsuit against GitHub Copilot

#442

Post by deanjones »

klabb3 wrote: post #420
deanjones wrote: post #322

This will fail very quickly. The licence that project owners publish with their code on Github applies to third parties who wish to use the code, but does not apply to Github. Authors who publish their code on Github grant Github a licence under the Github Terms: https://docs.github.com/en/site-policy/github-terms/github-t... Specifically, sections D.4 to D.7 grant Github the right to "to store, archive, parse, and d…

> Authors who publish their code on Github grant Github a licence under the Github Terms: https://docs.github.com/en/site-policy/github-terms/github-t ... This sounds unenforceable in the general case. How could github know whether someone pushes their own code or not? Is it a license violation to push someone's FOSS code to github because the author didn't sign up with GH?

> Is it a license violation to push someone's FOSS code to github because the author didn't sign up with GH?

It depends on the licence.

It's very much enforceable that companies who provide content publishing platforms will indemnify themselves against people publishing content to which they do not have an appropriate licence.

Re: We've filed a lawsuit against GitHub Copilot

#443

Post by an1sotropy »

bugfix-66 wrote: post #377
an1sotropy wrote: post #357

Earlier quoted context omitted.

Licenses get to set terms of redistribution. But training of the ML model -- the thing described by your #3 -- is not redistribution (imho). So maybe it's as unenforceable as saying left-handed people can't read your code. The redistribution happens later, either when copilot blurps out some of your code, or when the copilot user then distributes something using that code (I'm curious which). At that point, whether s…

The BSD license also controls "use", not just "redistribution": Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: That's word-for-word BSD license. The only change I made is adding clause 3: 3. Use in source or binary forms for the construction or operation of predictive software generation systems is prohibited.

Okay; thanks for clarifying. I actually hadn't noticed that use of "use" in the BSD license before. I think I need an IP lawyer explain what that "use" means.

Re: We've filed a lawsuit against GitHub Copilot

#444

Post by BizarroLand »

Aeolun wrote: post #414
xtracto wrote: post #337

Earlier quoted context omitted.

Say you publish a song and copyright it. Then I record it and save it in a .xz format. It's not an MP3, it is not an audio file. Say I split it into N several chunks and I share it with N different people. Or with the same people, but I share it at N different dates. Say I charge them $10 a month for doing that, and I don't pay you anything. Am I violating your copyright? Are you entitled to do that? To make it funni…

What you are actually giving people is a set of chords that happen to show up in your song, the machine can suggest an appropriate next chord. It’s also smart enough to rebuild your song from the chords _if you ask it to_.

With the exception that there are infinite types of chords in this case, and even though many musicians follow familiar chord structures the underlying melodies and rhythms are unique enough for any familiar person to be able to differentiate "Red Hot Chill Peppers" from "All-American Rejects", and now there is a system where All-American Rejects hit a few buttons and a song is generated (using audio samples of "Under the Bridge") that sounds like "Under the Bridge pt 2, All-American Rejects Boogaloo".

That's why it's actionable and why there is meat on the bone for this case. The real issue is going to be if they can convince a jury that this software is just stealing code and whether its wrong if a robot does it.

Re: We've filed a lawsuit against GitHub Copilot

#445

Post by S201 »

albertzeyer wrote: post #407

I really don't understand how there can be a problem with how Copilot works. Any human just works in the same way. A human is trained on lots and lots of of copyrighted material. Still, what a human produces in the end is not automatically derived work from all the human has seen in his life before. So, why should an AI be treated different here? I don't understand the argument for this. I actually see quite some dan…

I think there's a parallel in surveillance systems. For example, it's perfectly reasonable for a police officer conducting an investigation to follow a suspect as they drive around town. After all, it's happening in public and it's not illegal to watch what someone does in public (caveat being taking it to the level of stalking).

However, is it reasonable to write an AI system that monitors the time and location of all license plates seen around town, puts them into a database, and then that same officer can simply put in the suspect's license plate instead of actually following them around? Maybe, maybe not, that's not my point here. But the creation of that functionality can easily lead to its abuse.

Is this exactly the same case as Copilot? Of course not, these are two wildly different systems. But I think it's an interesting parallel to consider when discussing the point of "it's okay when a human does it" because humans and algorithms operate at two very different levels of scale. The potential for abuse of the latter being far higher and far easier than something a human has to do manually.

Re: We've filed a lawsuit against GitHub Copilot

#446

Post by nodogoto »

rowanG077 wrote: post #320
Supermancho wrote: post #303

Earlier quoted context omitted.

> The brain is also just a "complex math program". This is not a fact.

Explain yourself. There is not a understood natural phenomenon which we could not capture in math. If you argue behavior of the brain cannot be modeled using a complex math program you are claiming the brain is qualitative different then any mechanism known to man since the dawn of time. The physics that gives rise to the brain is pretty much known. We can model all the protons, electrons and photons incredibly accur…

> There is not a understood natural phenomenon which we could not capture in math.

You are saying "If we know how something works, we can explain how it works using math."

But we know almost nothing about how the brain works.

> The physics that gives rise to the brain is pretty much known.

...no it is not! No physicist would describe any physical phenomenon as being "pretty much known". Let alone cognition. We don't even have a complete atomic model.

Re: We've filed a lawsuit against GitHub Copilot

#447

Post by layer8 »

rowanG077 wrote: post #326
lisper wrote: post #304

Earlier quoted context omitted.

Somewhere in the complex math is the origin of whatever it is in intellectual property that we deem worthy of protection. Because we are humans, we take the complex math done by human brains as worthy of protection by fiat . When a painter paints a tree, we assign the property interest in the painting to the human painter, not the tree, notwithstanding that the tree made an essential contribution to the content. The…

Creations by AI should obviously be protected by fiat as well. Anything else is a ridiculous double standard that will stifle progress.

As long as AIs are incapable of recognizing when they are plagiarizing, as humans are generally capable of, the double standard seems entirely warranted.

Re: We've filed a lawsuit against GitHub Copilot

#448

Post by willhslade »

rowanG077 wrote: post #297
Cort3z wrote: post #270

I’m not a lawyer, but here is why I believe a class action lawsuit is correct; “AI” is just fancy speak for “complex math program”. If I make a program that’s simply given an arbitrary input then, thought math operations, outputs Microsoft copyright code, am I in the clear just because it’s “AI”? I think they would sue the heck out of me if I did that, and I believe the opposite should be true as well. I’m sure my ow…

The brain is also just a "complex math program". Since math is just the language we use to describe the world. I don't feel this argument has any weight at all.

So how and why do we solve the halting problem?

Re: We've filed a lawsuit against GitHub Copilot

#449

Post by willhslade »

rowanG077 wrote: post #297
Cort3z wrote: post #270

I’m not a lawyer, but here is why I believe a class action lawsuit is correct; “AI” is just fancy speak for “complex math program”. If I make a program that’s simply given an arbitrary input then, thought math operations, outputs Microsoft copyright code, am I in the clear just because it’s “AI”? I think they would sue the heck out of me if I did that, and I believe the opposite should be true as well. I’m sure my ow…

The brain is also just a "complex math program". Since math is just the language we use to describe the world. I don't feel this argument has any weight at all.

Also, to pile on, if I use my brain to read GPL source code and then type it in again on my own, I'm pretty sure I'm guilty of violating copyright.

Re: We've filed a lawsuit against GitHub Copilot

#450

Post by OJFord »

heavyset_go wrote: post #338
smoldesu wrote: post #313

Earlier quoted context omitted.

> “AI” is just fancy speak for “complex math program” Not really? It's less about arithmetic and more about inferencing data in higher dimensions than we can understand. Comparing it to traditional computation is a trap, same as treating it like a human mind. They've very different, under the surface. IMO, if this is a data problem then we should treat it like one. Simple fix - find a legal basis for which licenses a…

I've been saying AI is computational statistics on steroids for a while, and I think that's an apt generalization of what ML is.

That oughtn't be controversial, in fact I wouldn't even bother with 'on steroids', implying it's a slightly different/morphed thing. The way I learnt it (very slightly, at university, not a particular focus) it was abundantly clear it was just stats.
Post reply on HN