Comment by jacquesm
5 months ago
> which is about AI using code as input to produce similar code as output
> not about AI being trained on code
The two are very directly connected.
The LLM would not be able to do what it does without being trained, and it was trained on copyrighted works of others. Giving it a piece of code for a rewrite is a clear case of transformation, no matter what, but now it also rests on a mountain of other copyrighted code.
So now you're doubly in the wrong, you are willfully using AI to violate copyright. AI does not create original works, period.
Every programmer is trained on the copyrighted works of others. there a vanishingly few modern programs with available source code in the public domain.
it isn't clear how/if llm is different from the brain but we all have training by looking at copywrited source code at some time.
> it isn't clear how/if llm is different from the brain
It's very clear: the one is a box full of electronics, the other is part of the central nervous system of a human being.
> but we all have training by looking at copywrited source code at some time.
That may be so, but not usually the copyrighted source code that we are trying to reproduce. And that's the bit that matters.
You can attempt to whitewash it but at its core it is copyright infringement and the creation of derived works.
> but we all have training by looking at copywrited[sic] source code at some time.
The single word "training" is here being used to describe two very different processes; what an LLM does with text during training is at basically every step fundamentally distinct from what a human does with text.
Word embedding and gradient descent just aren't anything at all like reading text!
1 reply →
Programmer training and AI training are legally distinct processes.