Comment by foota

I can't see it being superhuman, that's for sure. Chess AI are superhuman because they do vast searches, and I can't see that being replicated by an LLM architecture.