Hence, GitHub is effectively using the work others made—for personal or non-commercial use, without having GitHub in mind, and without any way to say ‘no’ - to sell a product back to them, for their own profit.
How different is this from a human being learning how to code from reading public code on the internet, and then selling their own software? I suspect Copilot directly copies code less often than humans do! GitHub claims “about 1% of the time, a suggestion may contain some code snippets longer than ~150 characters that matches the training set”, and that their research shows this is mostly for very common patterns.
I’ve used Copilot a fair bit myself, and the code I actually use from it is typically very tailored to my specific use case, as only part of some bigger function I’m writing. (often unit tests)
The same principle applies to an AI like DALL-E—it’s not copying artwork, it’s simply learned how to make art from publicly available art.
That all said, I’m not actually opposed to an opt-out, nor do I disagree that there’s something unjust in the idea of replacing human work with AIs that learned from them for free. But I think the moral intuition has more to do with the sheer scalability and potential impact of these AIs than the act of learning from public code/art itself?
Good point. This exposes the true opposing argument, which is that people fear the possibility of the strong (i.e. Microsoft) becoming even stronger, via profits, influence, etc… But they don’t mind people like themselves, presumably quite weak in comparison to Microsoft, becoming stronger. Same as with punching up versus punching down and so on.
That’s not the true opposing argument. At least not “the” true one. I’ve no doubt that many people oppose it for that reason (in the case of Microsoft—totally understandable). But it’s neither the only or even main reason for many people.
There’s also the general free software idea, that you get to use my code however you like (yes, totally including earning billions from it without me getting anything), but only if you share any changes to the code with everyone else. It’s about fairness and freedom, not power.
There may be a group of people unconcerned about Microsoft profiting from their code yet concerned about the lack of sharing of the downstream work, but I can’t see how that could be the majority of people on GitHub who have complaints.
In every culture I’m aware of a third party profiting without prior consent is seen as more taboo than refusing to share a downstream product, without any attached profits.
How is it different? Obviously a person wanting to opt-out has a way—keep their source closed. Copilot and all the hundreds of AIs that will come after it can only see source they can reach.
Simplicity, Precedent—I don’t see an argument here. Anyone trying to make a more competitive coding AI is going to ignore any of these flags not enforced by law.
Ethics—Big companies obviously are in favor of them becoming even bigger, they didn’t get that way from holding back opportunities to win due to their scale.
Risk/Risk Compensation—these don’t apply to the current level of capabilities for coding agents.
How different is this from a human being learning how to code from reading public code on the internet, and then selling their own software? I suspect Copilot directly copies code less often than humans do! GitHub claims “about 1% of the time, a suggestion may contain some code snippets longer than ~150 characters that matches the training set”, and that their research shows this is mostly for very common patterns.
I’ve used Copilot a fair bit myself, and the code I actually use from it is typically very tailored to my specific use case, as only part of some bigger function I’m writing. (often unit tests)
The same principle applies to an AI like DALL-E—it’s not copying artwork, it’s simply learned how to make art from publicly available art.
That all said, I’m not actually opposed to an opt-out, nor do I disagree that there’s something unjust in the idea of replacing human work with AIs that learned from them for free. But I think the moral intuition has more to do with the sheer scalability and potential impact of these AIs than the act of learning from public code/art itself?
Good point. This exposes the true opposing argument, which is that people fear the possibility of the strong (i.e. Microsoft) becoming even stronger, via profits, influence, etc… But they don’t mind people like themselves, presumably quite weak in comparison to Microsoft, becoming stronger. Same as with punching up versus punching down and so on.
That’s not the true opposing argument. At least not “the” true one. I’ve no doubt that many people oppose it for that reason (in the case of Microsoft—totally understandable). But it’s neither the only or even main reason for many people.
There’s also the general free software idea, that you get to use my code however you like (yes, totally including earning billions from it without me getting anything), but only if you share any changes to the code with everyone else. It’s about fairness and freedom, not power.
There may be a group of people unconcerned about Microsoft profiting from their code yet concerned about the lack of sharing of the downstream work, but I can’t see how that could be the majority of people on GitHub who have complaints.
In every culture I’m aware of a third party profiting without prior consent is seen as more taboo than refusing to share a downstream product, without any attached profits.
How is it different? Obviously a person wanting to opt-out has a way—keep their source closed. Copilot and all the hundreds of AIs that will come after it can only see source they can reach.
Simplicity, Precedent—I don’t see an argument here. Anyone trying to make a more competitive coding AI is going to ignore any of these flags not enforced by law.
Ethics—Big companies obviously are in favor of them becoming even bigger, they didn’t get that way from holding back opportunities to win due to their scale.
Risk/Risk Compensation—these don’t apply to the current level of capabilities for coding agents.