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

All public GitHub code was used in training Copilot

twitter.com

Post reply on HN
Print view View original

341–350 of 734 posts

Re: All public GitHub code was used in training Copilot

#341

Post by hedora »

keonix wrote: post #185

If the training set contains verbatim (A)GPL code does this mean that Copilot also should be distributed by Microsoft under GPL? Because without it Copilot (as it is distributed by Microsoft) couldn't be built, wouldn't it make it a derivative work of GPL'd code (and obviously every other license)? I see a lot of people comparing human learning to machine learning in the comments, but there is a huge difference - we…

More interestingly, if we can trick it into regurgitating a leaked copy of the windows source code, Microsoft apparently says that’s fair use.

Re: All public GitHub code was used in training Copilot

#342

Post by feanaro »

AaronFriel wrote: post #193
keonix wrote: post #185

If the training set contains verbatim (A)GPL code does this mean that Copilot also should be distributed by Microsoft under GPL? Because without it Copilot (as it is distributed by Microsoft) couldn't be built, wouldn't it make it a derivative work of GPL'd code (and obviously every other license)? I see a lot of people comparing human learning to machine learning in the comments, but there is a huge difference - we…

No, see Authors Guild v. Google. Even without a license or permission, fair use permits the mass scanning of books, the storage of the content of those books, and rendering verbatim snippets of those books. The Google Books site is not a derivative work of the millions of authors they copied from, and if they did copy any coincidentally GPL, AGPL, or creative commons copyleft work, the fair use exception applies befo…

When the recent Github v. youtube-dl fiasco happened, I remember reading similarly strongly-worded but dismissive comments regarding fair use, stating how it is quite obvious that youtube-dl's test code could never be fair use and how fair use itself is a vague, shaky, underspecified provision of the copyright law which cannot ever be relied on.

To me, seeing youtube-dl's case as fair use is so much easier than using hundreds of thousands source code files without permission in order to build a proprietary product.

Re: All public GitHub code was used in training Copilot

#343

Post by rcfox »

spywaregorilla wrote: post #152
agilob wrote: post #142

Earlier quoted context omitted.

>copilot is not a human so it can't commit crimes I can setup my drone to detect me and attempt to crash into me. AI would be quite poor, probably would attempt to crash at any human. Would it be my fault it didn't crash into me and someone lost eyes? Can I setup torrent box that automatically downloads and seeds all detected links from public trackers? Would I be responsible for it?

Both of these examples include you creating something and then using it. I don't know how copilot works, but using the second example, if you wrote a script to download and seed trackers, and someone else used it, I don't think you would be held under any liability, especially if you don't profit off of it. Not a lawyer or even particularly well informed edit: I am reminded of the monkey selfie, in which it was ruled…

It sounds like you're arguing that Github isn't liable for people using copyrighted code through Copilot.

I think most people are more concerned about whether the user of Copilot would be liable for using copyrighted code generated by Copilot.

Re: All public GitHub code was used in training Copilot

#344

Post by nh2 »

glogla wrote: post #249
blibble wrote: post #203

Earlier quoted context omitted.

it would be a win for Microsoft that don't distribute their source code whereas for open source it's a disaster

Which seems very much align with what has Microsoft been trying to do for decades now.

It would be quite impressive if this was a long-time planned "Embrace, extend, extinguish" move against Copyleft, with a casual acquisition of Github to make it work.

Finally, it beat the "cancer that attaches itself in an intellectual property sense to everything it touches" after all those years, with its own tools!

Now it's safe to touch.

Re: All public GitHub code was used in training Copilot

#345

Post by mrh0057 »

Luker88 wrote: post #335

Like it or not, it seems like: * most people here are unhappy * most laywers will say it's fine (it very probably passed MS ones) I can understand that. Copyright was not created with AI/ML in mind, even as a random stray thought. Those were not even words at the time. So the question is: If we change the law and require trained algorithms to only work on licenses that permit this, and to output the "minimum common l…

There is nothing intelligent about this. What they did is a context aware search and trying to claim that not what this is. If it was just used as a search engine and people weren’t using the results or following the license of the original source, then it would fine. There has been so much of a hype of machine learning people likely have a false impression of what it is.

Re: All public GitHub code was used in training Copilot

#346

Post by wpietri »

treesprite82 wrote: post #338
wpietri wrote: post #127

Earlier quoted context omitted.

I think you're right. Especially given that Copilot can reproduce significant blocks of code: https://twitter.com/mitsuhiko/status/1410886329924194309 Famous code: https://en.wikipedia.org/wiki/Fast_inverse_square_root#Overv...

I see this held up as an example a lot, but the fast inverse square root algorithm didn't originate from Quake and is in hundreds of repositories - many with permissive licenses like WTFPL and many including the same comments. GitHub claims they didn't find any "recitations" that appeared fewer than 10 times in the training data. That doesn't mean it's a completely solved issue (some code may be repeated in many repo…

That may be true, although even GitHub doesn't know for sure. But the problem remains: they're reproducing other people's code without regard to license status.

Re: All public GitHub code was used in training Copilot

#347

Post by slownews45 »

CyberRabbi wrote: post #118
ssivark wrote: post #64

Calling it “public” code feels like doublespeak. It’s most definitely NOT public domain code — it only happens to be hosted on GitHub and browsable ( but not copyable ) by people. “Source available for viewing” is very different from “public property” as the phrase is commonly understood: https://en.m.wikipedia.org/wiki/Public_property

Movies are “public” too. That does not mean you are allowed to use them for any purpose. The term “Public” does not have specific legal consequences in copyright law outside of something being “public domain” as you say.

You are allowed to watch them. Many moves take ideas from other movies, which took ideas from myths and earlier stories. In fact, I find modern movies highly highly derivative.

Re: All public GitHub code was used in training Copilot

#348

Post by zoomablemind »

rapind wrote: post #329
keonix wrote: post #185

If the training set contains verbatim (A)GPL code does this mean that Copilot also should be distributed by Microsoft under GPL? Because without it Copilot (as it is distributed by Microsoft) couldn't be built, wouldn't it make it a derivative work of GPL'd code (and obviously every other license)? I see a lot of people comparing human learning to machine learning in the comments, but there is a huge difference - we…

This is pretty interesting for AI in general. Should you be able to train with material you don't own? Can your training benefit from material that has specific usage licenses attached to it? What about stuff like GameGAN?

> ...Should you be able to train with material you don't own?

If relating this to how humans learn, books and other sources are used to inform understanding and human knowledge. One can purchase or borrow a book without actually owning the copyright to it. Indeed, a given passage may be later quoted verbatim, provided it is accompanied with a reference to its source.

Otherwise, a verbatim use without attribution in authored context is considered plagiarism.

So, sure one can use a multitude of material for the training. Yet, once it gets to the use of the acquired "knowledge" - proper attribution is due for any "authentic enough" pieces.

What is authentic enough in this case is not easy to define, however.

Re: All public GitHub code was used in training Copilot

#349

Post by slownews45 »

mehdix wrote: post #170

GitHub's Copilot looks like a "code laundering" machine to me.

Developers have lost the plot here. The number of people browsing stack exchange and copying code is huge. The number of people who have read GPL'ed code to learn from (from the kernel to others) is huge. The number of people who learned from code they had to maintain -> huge.

This idea that a snippet of a code is a work seems crazy to me. I thought we went through this with SCO already.

Re: All public GitHub code was used in training Copilot

#350

Post by 0-_-0 »

feanaro wrote: post #342
AaronFriel wrote: post #193

Earlier quoted context omitted.

No, see Authors Guild v. Google. Even without a license or permission, fair use permits the mass scanning of books, the storage of the content of those books, and rendering verbatim snippets of those books. The Google Books site is not a derivative work of the millions of authors they copied from, and if they did copy any coincidentally GPL, AGPL, or creative commons copyleft work, the fair use exception applies befo…

When the recent Github v. youtube-dl fiasco happened, I remember reading similarly strongly-worded but dismissive comments regarding fair use, stating how it is quite obvious that youtube-dl's test code could never be fair use and how fair use itself is a vague, shaky, underspecified provision of the copyright law which cannot ever be relied on. To me, seeing youtube-dl's case as fair use is so much easier than using…

How would you feel about a paid-for search engine using hundreds of millios of web pages without permission in order to build a proprietary product?
Post reply on HN