Not sure about LLMs but I yearn for someone to train neural nets to play the NPCs in a shooter game well but without having aimbots or seeing through walls - basically a “true” version of F.E.A.R. AI where they play simulated squads that overwhelm the player with various tactics and adapt to their playstyle. Or even just make increasingly complex state machines and goal oriented systems which might be more performant but still encode the main rules.

I expect the area of fun between painfully easy to beat and infuriatingnly overpowered with superhuman skill level to be quite narrow.