Why can\u2019t AI bots be made to be good, to be moral, so they\u2019ll 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?