(This doesn't disprove your point because it seems like it should be trivial to make but interestingly)
I don't think there's any computer system which autonomously come up with new directions in openings. As far as I know; a GM looks at stockfish's evaluation of the top x moves and analyzes one that hasn't been played a lot etc.
Comments
Sure, but if we continue that analogy it does mean that there will be no human contributions to frontier mathematics.
(This doesn't disprove your point because it seems like it should be trivial to make but interestingly)
I don't think there's any computer system which autonomously come up with new directions in openings. As far as I know; a GM looks at stockfish's evaluation of the top x moves and analyzes one that hasn't been played a lot etc.
If the output is better who cares?