4 ms·

Speaking about this, there is something I don't understand about AI-produced work : given how you need to train AI on an existing set of existing code (or artwo

Speaking about this, there is something I don't understand about AI-produced work : given how you need to train AI on an existing set of existing code (or artworks, or novels, or whatever), I don't even understand how it would be possible to patent/copyright the output as it is, to some remote degree, derivative by nature ? Humans don't need AIs to produce code, but the opposite is not true.

Filligree 4y ago
Humans also read code, though. I don’t think we want a world where programmers are required never to read someone else’s code, or otherwise be barred from the profession.
- Dalewyn 4y ago
  That is the world we live in, though. For a prominent example, WINE goes to extreme ends to ensure their code isn't derived from any prior direct knowledge of Microsoft code. Replicating functions of released binaries with original code is fair game, but reverse engineering or copying Microsoft code is a hard no-go. As far as AI is concerned in all this, it's probably even easier to distinguish the unacceptable. Whereas humans can argue for plausible deniability, we clearly know what data an AI is fed to generate subsequent output. So unless an AI is certified to never have eaten licensed materials, literally everything it produces will be license infringements.
- Ialdaboth 4y ago
  Feels fallacious ? Most good developers I personally know, when doing code, tend to think about the problem at hand, establish the logic of how they want to solve it, and then implement the solution. They rarely try to do this by doing pattern-hunting & merge on a mountain of code from similar projects.
  - Filligree 4y ago
    Why do you assume copilot is doing the latter? The training data doesn’t exist at runtime, and if the developers are competent then it only learned a bit or two from each input snippet. So, though even though it’s certainly a highly capable compression algorithm, verbatim copies shouldn’t be… possible. My best guess for this bug is that a lot of already plagiarised the code. AI has a habit of holding up a mirror to humanity. Sometimes we don’t like what we see.
- AlexandrB 4y ago
  I hate this comparison because it implies that humans are somehow legally the equals of an AI algorithm. I don't think that's the case at all. For example, watching a movie or listening to a song has never been considered copyright infringement even thought on some level you're creating a copy of the copyrighted work in your head.
  - Timwi 4y ago
    Nobody would complain about an AI that does that, either. The problem isn't that the AI read some code and learnt from it. The problem is that it spits it back out, but without attribution. It's like listening to a song and then reproducing and publishing a strikingly similar version of it, but claiming that you wrote it on your own.
saynay 4y ago
You can copyright derivative, transformative work though? I don't think current legal frameworks are equipped to handle these types of generative works. There will need to be either new laws, or at least some legal precedence established. For example, it would make total sense for stuff posted publicly to optionally have a "not to be used to train models" license. But models are being trained on public data that pre-dates anyone thinking that is something that they should worry about.

Filligree 4y ago

Humans also read code, though. I don’t think we want a world where programmers are required never to read someone else’s code, or otherwise be barred from the profession.

Dalewyn 4y ago
That is the world we live in, though. For a prominent example, WINE goes to extreme ends to ensure their code isn't derived from any prior direct knowledge of Microsoft code. Replicating functions of released binaries with original code is fair game, but reverse engineering or copying Microsoft code is a hard no-go. As far as AI is concerned in all this, it's probably even easier to distinguish the unacceptable. Whereas humans can argue for plausible deniability, we clearly know what data an AI is fed to generate subsequent output. So unless an AI is certified to never have eaten licensed materials, literally everything it produces will be license infringements.
Ialdaboth 4y ago
Feels fallacious ? Most good developers I personally know, when doing code, tend to think about the problem at hand, establish the logic of how they want to solve it, and then implement the solution. They rarely try to do this by doing pattern-hunting & merge on a mountain of code from similar projects.
- Filligree 4y ago
  Why do you assume copilot is doing the latter? The training data doesn’t exist at runtime, and if the developers are competent then it only learned a bit or two from each input snippet. So, though even though it’s certainly a highly capable compression algorithm, verbatim copies shouldn’t be… possible. My best guess for this bug is that a lot of already plagiarised the code. AI has a habit of holding up a mirror to humanity. Sometimes we don’t like what we see.
AlexandrB 4y ago
I hate this comparison because it implies that humans are somehow legally the equals of an AI algorithm. I don't think that's the case at all. For example, watching a movie or listening to a song has never been considered copyright infringement even thought on some level you're creating a copy of the copyrighted work in your head.
- Timwi 4y ago
  Nobody would complain about an AI that does that, either. The problem isn't that the AI read some code and learnt from it. The problem is that it spits it back out, but without attribution. It's like listening to a song and then reproducing and publishing a strikingly similar version of it, but claiming that you wrote it on your own.

saynay 4y ago

You can copyright derivative, transformative work though? I don't think current legal frameworks are equipped to handle these types of generative works. There will need to be either new laws, or at least some legal precedence established. For example, it would make total sense for stuff posted publicly to optionally have a "not to be used to train models" license. But models are being trained on public data that pre-dates anyone thinking that is something that they should worry about.