GitHub Copilot as open source code laundering? • Hacker News Board

Re: GitHub Copilot as open source code laundering?

#131

Post by mysterydip » Wed, Jun 30, 2021, 2:16 PM UTC

MaxBarraclough wrote: post #103

pc86 wrote: post #53

Earlier quoted context omitted.

Every decade was supposed to see fewer hours working for higher pay and quality of life. It didn't happen, as business owners (not just 1% fat cats, the owners of mom and pop shops are at least as guilty as anyone, they just sucked at scaling their avarice). So the claim that this technological revolution will be different and that it will result in a broad social safety net, universal basic income, and substantive,…

> those left behind working more hours for less Doing what? Isn't the concern here that automation will push many people out of the workforce entirely?

Who debugs the software when there's a problem?

Re: GitHub Copilot as open source code laundering?

#132

Post by pc86 » Wed, Jun 30, 2021, 2:17 PM UTC

belter wrote: post #101

One interesting aspect, that I thing will make it difficult for GitHub to argue and justify its not a a license violation would be the answer to the following question: Was Copilot trained using Microsoft internal source code or will it be in the future ? As GitHub is a Microsoft company and OpenAI although a non-profit just got a massive one billion investment from Microsoft (presumably not for free), will it start…

The "because" in your last bit is a huge leap.

It wasn't trained on internal Microsoft code because the training set is publicly available code. It has nothing to do with whether or not it suggests exactly identical, functionally identical, or similar code. MS internal isn't publicly available. Copilot is trained on publicly available code.

Re: GitHub Copilot as open source code laundering?

#133

Post by akerl_ » Wed, Jun 30, 2021, 2:18 PM UTC

belter wrote: post #101

One interesting aspect, that I thing will make it difficult for GitHub to argue and justify its not a a license violation would be the answer to the following question: Was Copilot trained using Microsoft internal source code or will it be in the future ? As GitHub is a Microsoft company and OpenAI although a non-profit just got a massive one billion investment from Microsoft (presumably not for free), will it start…

Without weighing in on the overall question of “is this a license violation”, you’ve created a false dichotomy.

“GitHub included Microsoft proprietary code in the training set because they view the results as non-derivative” and “GitHub didn’t include Microsoft proprietary code because they view the results as derivative” are clearly not the only options. They could have not included Microsoft internal code because it was way easier to just use the entire open source corpus, for example.

Re: GitHub Copilot as open source code laundering?

#134

Post by shawnz » Wed, Jun 30, 2021, 2:18 PM UTC

cycomanic wrote: post #46

In the discussion yesterday I pointed to the case of some students suing turnitin for using their works in the turnitin database and the studemts lost [1]. I think an individual suing will not go anywhere. The way to create a precedent is someone feeding all the Harry Potter books and some additional popular books (twilight?) to GPT 3 and letting them write about some kids at a sorcerer school. The outcomes of that c…

Disney's intellectual property would be a good choice for this exercise

Re: GitHub Copilot as open source code laundering?

#135

Post by emodendroket » Wed, Jun 30, 2021, 2:19 PM UTC

belter wrote: post #101

One interesting aspect, that I thing will make it difficult for GitHub to argue and justify its not a a license violation would be the answer to the following question: Was Copilot trained using Microsoft internal source code or will it be in the future ? As GitHub is a Microsoft company and OpenAI although a non-profit just got a massive one billion investment from Microsoft (presumably not for free), will it start…

Since quite a lot of Microsoft code is on GitHub, I'd say yes.

Re: GitHub Copilot as open source code laundering?

#136

Post by MaxBarraclough » Wed, Jun 30, 2021, 2:20 PM UTC

mysterydip wrote: post #131

MaxBarraclough wrote: post #103

Earlier quoted context omitted.

> those left behind working more hours for less Doing what? Isn't the concern here that automation will push many people out of the workforce entirely?

Who debugs the software when there's a problem?

Professional software developers, i.e. members of one of the well-paid professions that is not under immediate threat from automation.

Re: GitHub Copilot as open source code laundering?

#137

Post by gus_massa » Wed, Jun 30, 2021, 2:23 PM UTC

I think copyright is a problem for GPL-like licenses. They should have restricted the training data to MIT/BSD-like.

Anyway, there is another problem that is patents and is huger, much huger. I think the Apache license has a provision about patents, but most of other licenses may have code that has patents and if the AI generate something similar it may be included in the patent.

Re: GitHub Copilot as open source code laundering?

#138

Post by WillDaSilva » Wed, Jun 30, 2021, 2:23 PM UTC

amelius wrote: post #127

WillDaSilva wrote: post #87

Earlier quoted context omitted.

It is very similar to the current situation, but intensified. Technology tends to be an intensifier for existing power structures.

Except some random nobody can become a disruptor.

I was debating bringing up disruptors when I made the grandparent comment. My 2 cents: they can shift the balance of power at the very small scale (e.g. "some random nobody" getting rich, or some rich person going bankrupt), but the large scale power structures almost always remain largely intact. For instance, that "random nobody" may well get rich through the sale of shares in their company - now the company is owned by the owner class, who were previously at the top of the power hierarchy.

Re: GitHub Copilot as open source code laundering?

#139

Post by pjerem » Wed, Jun 30, 2021, 2:28 PM UTC

Ygg2 wrote: post #124

jackbeck wrote: post #68

Earlier quoted context omitted.

He mentioned that the Linux-like kernel will be closed source which violates GPL

Does it, if code was written by a bot that trained on Linux kernel?

You know, that's precisely what the topic here is about.

Re: GitHub Copilot as open source code laundering?

#140

Post by 6gvONxR4sf7o » Wed, Jun 30, 2021, 2:29 PM UTC

> previous """AI""" generation has been trained on public text and photos, which are harder to make copyright claims on, but this is drawn from large bodies of work with very explicit court-tested licenses

This seems pretty backwards to me. A GPL licensed data point is more permissive than an unlicensed data point.

That said, I’m glad that these data points do have explicit licenses that say “if you use this, you must do XYZ” so that it’s clear that our large ML projects are going counter to creators intent when they made it open.

I’d love to start seeing licenses about use as training data. Then maybe we’d see more open access to these models that benefit from the openness of the web. I’d personally use licenses that say if you want to train on my work, you must publish the model. That goes for my code, my writing, and my photography.

Anyways GitHub is arguing that any use of publicly available data for training is fair use, but they also admit that it’s all new and unprecedented, regarding training data.