2. GPL does not allow you to take the code, compress it in your latent space, and then sell that to consumers without open sourcing your code.
Sure, that's what the paper says. Most people don't care what that says until some ramifications actually occur. E.g. a cease and desist letter. Maybe people should care, but companies have been stealing IP from individuals long before GPL, and they still do.
If AI training is found to be fair use, then that fact supercedes any license language.
If there is some copyrighted art in the background in a scene from a movie, maybe that's fair use. If you take a high resolution copy of the movie, extract only the art from the background and want to start distributing that on its own, what do you expect then?
However if I memorise that code and write it down that's not fair use. If I copy the encyclopedia that's bad.
The problem then comes into "how trivial can a line be before it's copyrighted"
def main():
print("This is copyrighted")
main()
No one goes to prison for this. They might get sued, but even that is doubtful.
We're talking about the users getting copyright-laundered code here. That's a pretty equal playing field. It's about the output of the AI, not the AI itself, and there are many models to choose from.
There don’t seem to be any usable open-source models.