Artificial intelligence (AI) and machine learning (ML) in the form of conversational expert systems can be programmed to follow — or reject — first principles. Systems can be “liberal” or “conservative” depending on how they’re programmed. We should be thinking seriously about ethics and AI.