Jack is one of my favorite follows in AI so I always wallow in doubt when I disagree, but having built enterprise AI at scale, I disagree here.

Yes, we might see human-light and AI-heavy companies and we will see labor augmented on a massive scale.

But 1) humans and economy will evolve quickly in response, moving our labor to more effective uses, just like we’ve done over the last 100 years.

And, what I think a lot of AI research overlooks: 2) even in the existing economy, there’s enough muck layer in most jobs - the interhuman messiness - that they will be resistant to major changes within our lifetime. I wrote about it here: loadbearingtech.substac…

Footnote: in BlueDot’s AGI Strategy course, the facilitator pointed out to me that my argument assumes humans stay relevant as the “labor” part of the economy, but capital and technology might detach from the equation, leaving labor behind (economy = labor + capital + technology). There’s some convenience to the sweepiness of this argument, so let’s think through the minutiae of it. Can we pick out the parts of the economy and how they’ll be automated? Medicine? Law? Politics? Childcare? Entertainment? Which parts in what order? I acknowledge that this is tedium but it’s in the tedium that the devil of non-total-automation sits. Because as soon as we have one human involved, we need a host of humans to support their various human flaws, both executive and emotional. That said: I am curious about engaging in this argument, because many people smarter than me have expressed it. I would love to understand that POV.