Even with Google Maps, we still need human drivers because current AI systems aren’t so great at driving and/or are too expensive to be widely adopted at this point. Once AI figures out driving too, what do we need the drivers for?

And I think that’s the point he was making, it’s hard to imagine any task where humans are still required when AI can do it better and cheaper. So I don’t think the Uber scenario is realistic.

I think the only value humans can provide in that future is “the human factor”: knowing that something is done by an actual human and not a machine can be valuable.

People want to watch humans playing chess, even though AI is better at it. They want to consume art made by humans. They want a human therapist or doctor, even if they heavily rely on AI for the technical stuff. We want the perspective of other humans even if they aren’t as smart as AI. We want someone that “gets” us, that experiences life the same way we do.

In the future, more jobs might revolve around that, and in industries where previously we didn’t even consider it. I think work is going to be mostly about engaging with each other (even more meetings!)

The problem is, in a world that is that increasingly remote, how do you actually know it’s a human on the other end? I think this is something we’ll need to solve, and it’s going to be hard with AI that’s able to imitate humans perfectly.

The spinning jenny put seamstresses out of work. But the history of automation is the history of exponentially expanding the workforce and population.

8 billion people wake up every morning determined to spend the whole day working to improve their lives. we're gonna be ok.