This is explained better in the Stanford Encyclopedia of Philosophy: https://plato.stanford.edu/entries/artificial-intelligence/#...