Comment by Andrex
3 years ago
Reproduced to what fidelity? 100%?
If OpenAI's output reproduces a copyrighted image with one pixel changed, is that valid in your view? Where does the line end?
Copyrighted material should never be used for nonacademic language models. "Garbage in, garbage out." All results are tainted.
"But being forced to use non-copyrighted works will only slow things down!"
Maybe that's a good thing, too. Copyright is something every industry has to accept and deal with -- LLMs don't get a "cool tech, do whatever" get-out-of-jail free card.
> Reproduced to what fidelity? 100%?
Whatever the standard is for humans doing the exact same thing.
Copyright is decided by the courts, it's a legal thing not some a biological. If the courts decide it's legal it will be.
I'm totally down for the courts handling this AI/copyright mess, but I don't think technologists are going to like the results.
By virtue of the fact that it is "fuzzy" and open to interpretation, we're going to see lawsuits, the resulting chilling effects of those lawsuits will blunt US tech firms from the practice of ingesting large amounts of copywritten material without a second thought. US tech firms will be giving it a second, third, fourth, etc. thought once the lawsuits start.
It's gonna be like submarine patents on steroids.
Like I said, I'm down for letting the courts decide. But AI supporters should probably avoid kicking the hornets' nests regarding copyright.