From first principles wouldn't a model trained on the written law and court rulings be capable of supplanting an "expert" that is in essence creating derivative works based off of such documents?
I mean, maybe AGI will be capable in the future, but I don't see a LLM doing that.
Experts are required, at least for now, to advance the body of knowledge.
Anyway, the thinking is that humans take so long to learn that we can't master two or more disciplines, and there's useful stuff in the gaps between disciplines that's actually quite easy to discover for even a simple AI model if it can be taught enough.
It's not quite the same as extrapolating General Relativity, but it could be a case where AI advances the body of knowledge.