We actually know so little about human cognition that that assumption seems wildly naive to me. Having said that, I assume that at some point in the future AI'

We actually know so little about human cognition that that assumption seems wildly naive to me. Having said that, I assume that at some point in the future AI's problem-solving abilities will surpass those of humans. But there's still other issues to reckon with even then. I think we should all be much more worried about what humans will do in response to AI than AI itself. Framing this all in terms of AI rather than AI-related policy distracts from the real issues in my opinion. I'm not worried about AI, I'm worried about people: what people will do with it and what rationalizations they will have for how they treat other people as a consequence.