Fleuret argues AGI will trigger a sharp, noticeable surge.
Andrej Karpathy thinks AGI's impact on the economy will just be folded into the existing rate of growth. AI will be barely noticeable in GDP statistics.
When he came on the show, I pushed back, saying AGI will cause a massive jump in productivity and growth.
Watch our back-and-forth on this:
Some users praised Karpathy as insightful and agreed AGI will blend into normal economic growth like past tech, while others dismissed the claims as delusional nonsense and criticized Dwarkesh's interview style.
@dwarkesh_sp I wonder if he still believes this
@stanislavfort @karpathy @mioana this brookings article we co-wrote: https://www.brookings.edu/articles/artificial-intelligence-saturation-and-the-future-of-work/
Sorry. I got to say this publicly. I really agree with @karpathy point here. My wife @mioana is a leading economist and we discuss this all the time. The singularity think of the AI community is rather misguided.
Andrej Karpathy thinks AGI's impact on the economy will just be folded into the existing rate of growth. AI will be barely noticeable in GDP statistics.
When he came on the show, I pushed back, saying AGI will cause a massive jump in productivity and growth.
Watch our back-and-forth on this:
@SOPHONTSIMP @karpathy @mioana It may be surprising but exactly yes. Most tasks, once you give them infinite intelligence, are bottlenecked by something else. Maybe ad placement would be infinitely good though ;)
@KordingLab @karpathy @mioana So we’ll have AGI and it will just be business as usual, the machine minds are here and it’s still just kind of boring? I don’t think that necessarily crazy but I think that’s funny.
@tszzl @dwarkesh_sp I think all his points still apply; what do you think has changed?
@Raamana_ @EliSennesh @karpathy @mioana https://www.brookings.edu/articles/artificial-intelligence-saturation-and-the-future-of-work/
@dwarkesh_sp Weird how obviously and trivially wrong Andrej is here. Dwarkesh has the obvious correct sense here. If you double the population overnight you’ll get much more than 2% GDP growth.
@tszzl @dwarkesh_sp GDP = C + I + G + (X − M)
Perhaps with AGI the biggest impact will be in export/import because factories can be fully automated locally.
@KordingLab @karpathy @mioana Say more?
"The Economy" is just relative popularity of things that are scarce. In the 1700 spices and photos were scare.
Today good healthcare is scarce, when that becomes abundant it drops out of GDP and something else then becomes more scarce relatively that is why GDP doesn't go up as much as you expect.
@tszzl @dwarkesh_sp we might be past the time where we get to hear publicly what he believes in private
@Navier_Tokes @tszzl @binarybits @dwarkesh_sp And adding more agents or IQ points doesn't really change the story - it's about now economies work not how AGI works.
@binarybits @besttrousers @dwarkesh_sp at the time Karpathy seemed not AGI-pilled. understandably, I mean the models just weren’t as good in the middle of 2025. he may no longer believe it’s “normal technology”
Adoption of these technologies throughout the Fortune 493 takes a long time. It will be adopted, but it’ll be slower than you’d think. That’s why you won’t see a big change in GDP.
Btw, opposing AI’s productivity growth are state governments that are doing everything they can to wreck their own economies through useless or even harmful spending.
@Navier_Tokes @tszzl @binarybits @dwarkesh_sp So to some extent AGI policy is sort of like "a free trade agreement with a new civilization of a trillion people each of which has an IQ of 500".
To the extent the metaphor works (maybe it doesn't!) I'd expect the economic effects to be somewhat limited.
@tszzl @binarybits @dwarkesh_sp I think an interesting question for this is whether any given prediction is about a stylized fact about "the economy" or a stylized fact about "AI capacity/future AI capacity".
@besttrousers @tszzl @dwarkesh_sp Yeah, I'm also confused about this.
@dwarkesh_sp This prof discusses both point of views you guys mentioned : https://youtu.be/xBpGn3BDcOY?si=IUfCgryZZg7ePz3o
the part you're both circling: if AGI lands and GDP barely moves, that's not AI failing. it's the number failing.
economists said the same about computers for two decades everywhere except the productivity stats.
karpathy's right, just not the way it sounds. 'invisible in GDP' isn't a ceiling on AI. it's a blind spot in the ruler.
Fleuret argues AGI will trigger a sharp, noticeable surge.
Andrej Karpathy thinks AGI's impact on the economy will just be folded into the existing rate of growth. AI will be barely noticeable in GDP statistics.
When he came on the show, I pushed back, saying AGI will cause a massive jump in productivity and growth.
Watch our back-and-forth on this:
