If a machine or an AI program matches or surpasses human intelligence, does that mean it can simulate humans perfectly? If yes, then what about reasoning—our ability to apply logic and think ...

However, there's a problem with this, which I can't resolve. Anyone who's done a computing machinery course will have heard of the Halting Problem, or Self Applicability, or some variant thereof.