Paul Bloom: Can AI be Moral?

Why can’t AI bots be made to be good, to be moral, so they’ll help us and not do harmful or terrible things? But just whose moral values would we want them to have? And what if they become too moral?