[YeGoblynQueenne]

Their AI is "alpha-beta search agents" which, unless that's a very misleading description, sounds like a bunch of instances of minimax with alpha-beta pruning [1]. Which indeed should be the simplest, cheapest and most sensible option for that kind of game. I mean, no need to run an LLM/ agent harness on such a simple board game, that'd be overkill. And, well, tbh, it wouldn't work as well.

So they could have used wear patterns as a component in an evaluation function used in minimax but I get the feeling they most likely did a wear analysis afterwards to weed out some of the possible rulesets indicated by the minimax play patterns.

______________

[1] https://en.wikipedia.org/wiki/Alpha%E2%80%93beta_pruning