> One thing I find strange is how much we emphasize the artificial nature of the intelligence. We really don't know what intelligence (sans qualifications) is.

> One thing I find strange is how much we emphasize the artificial nature of the intelligence. We really don't know what intelligence (sans qualifications) is. AI has been a term for effort emulate what we roughly think of as "intelligent" behavior. It's far from successful so far and the lack of a "theory of intelligence" is probably part of that. But it's pretty clear what "AI" researchers and systems are doing now is far from intelligence. > AI and automation always occurs in the context of human processes. Nothing is truly autonomous, so why design it as if human involvement is a failure? This argument makes as much sense as "we'll never exceed the speed of light, why act like faster transportation matters". An automated factory still requires some maintenance but it's creation certainly is significant. > We can easily design artifacts to enhance human intelligence or team intelligence. Why the focus on the machine part and not the overall system that functionally accomplishes the desired work? Both approaches matter and since there's really nothing keeping people from doing both of these, people pursue each separately. Moreover, I'd say AI research could do well to cross-pollinate with human-computer interaction theory. But overall, you seem to just not understand why automation matters - automation has brought vast productivity in a variety of fields. It may or may not be possible other further fields but if it is, it will transform the world equivalently.