An interesting paper, and also a direction worth examining further. I am actually surprised that both a group of humans and GPT-4 output rather diverse ideas. R

An interesting paper, and also a direction worth examining further. I am actually surprised that both a group of humans and GPT-4 output rather diverse ideas. Regarding the latter, however, I do wonder not only the feasibility aspect they discuss but also the radicality of ideas outputted, i.e. whether it can yield true novelty, which I still doubt.