"AI companies claim their tools couldn't exist without training on copyrighted material. It turns out, they could — it's just really hard. To prove it, AI researchers trained a new model that's less powerful but much more ethical. That's because the LLM's dataset uses only public domain and openly licensed material."

tl;dr: If you use public domain data (i.e. you don't steal from authors and creators) you can train a LLM just as good as what was cutting edge a couple of years ago. What makes it difficult is curating the data, but once the data has been curated once, in principle everyone can use it without having to go through the painful part.
So the whole "we have to violate copyright and steal intellectual property" is (as everybody already knew) total BS.

engadget.com/ai/it-turns-out-y…

It turns out you can train AI models without copyrighted material

It's just a pain in the ass.

^{Will Shanklin (Engadget)}