You know how when you give an AI shallow directives in machine learning, it'll start doing weird things? Like that Roomba that got stuck permanently driving bac

You know how when you give an AI shallow directives in machine learning, it'll start doing weird things? Like that Roomba that got stuck permanently driving backwards, or the game AIs that figure out that standing still is the best way not to die. Well it works like that for capitalism too. People optimize for the goal (money), not necessarily the hopeful outcome (useful things).