I agree that we won’t create a human intelligence. If we create an intelligence it’ll at best be a warped “something else.” However I’ll make the trite point t

I agree that we won’t create a human intelligence. If we create an intelligence it’ll at best be a warped “something else.” However I’ll make the trite point that I don’t think we understand what makes human intelligence manifest. There’s no reason to believe its not at its basis a statistical optimization based on reinforced gradient descent. In fact it’s fairly likely to be the case. My assertion is as we mix other AI techniques in a feedback cycle we will see something more than this simple single model exploration can do and it’ll converge to what we would consider intelligence. But I agree fundamentally it won’t map to a human intelligence modality.