Generally doing "the right thing" based on a rough definition of "what we have and what we want" is not mind-reading, but it is essentially what Prolog does. Th

Generally doing "the right thing" based on a rough definition of "what we have and what we want" is not mind-reading, but it is essentially what Prolog does. The author seems to conflate AI with declarative programming here, and I think that was the point.