684. He Helped Clean Up the Last Crash. Does He See Another One Coming?
with Gary Gensler
7 Aug 20265 min read38m
TL;DR
Gary Gensler warns that AI spending — up nearly fivefold to $750 billion in three years — looks like every historical tech investment bubble, with revenues of only $150-200 billion to show for it. The real danger is a 'parlay bet': the capital expenditure must generate both sufficient revenues AND near-term productivity gains to avoid a serious economic washout. Meanwhile, US federal debt at 100% of GDP and a yield curve that has structurally narrowed over 20 years make the system unusually fragile if the AI boom plateaus.
Key Moments
Gary Gensler
“We have a parlay bet right now. It's that the model companies like Open AI and Anthropic and so forth and the hyperscalers like Microsoft and Google that their capital expenditures will lead to enough revenues.”
Gensler explains why AI represents a compounded financial risk — two bets that both have to win simultaneously.
“When I was born in the 1950s and you said, 'What was finance's total aggregate part of the US economy?' It was maybe 3%. When I went to Wall Street in 1979, it was maybe 5%. And now it's 8%.”
Gensler traces the growing dominance of finance in the US economy and questions whether that growth made things better.
“Right now, we have give or take 750 billion of spend. And we might have native revenues if we're being generous this year 150 maybe 200 billion. So we know something. It's not in equilibrium right now.”
Gensler quantifies the gap between AI capital expenditure and actual revenues to show the boom is structurally unbalanced.
“All of a sudden it's a little bit like that cartoon. You're watching the character run and all of a sudden they ran off the cliff and their feet are still moving and there's not the revenues to support where they are.”
Gensler uses a Wile E. Coyote metaphor to describe how AI investment booms eventually disconnect from underlying economic reality.
Gary Gensler spent 18 years at Goldman Sachs before entering public service, where he chaired the CFTC from 2009 to 2014 and later the SEC under the Biden administration. At the CFTC he led the post-2008 cleanup of derivatives markets, overseeing 67 rules under Dodd-Frank. He is now a professor of the practice at MIT, specializing in financial booms and busts.
Takeaways
1
AI capex is a $600B unproven bet AI-related capital spending has grown nearly fivefold in three years to $750 billion, but revenues are only $150-200 billion. Every major technology wave — railroads, electrification, the internet — saw this investment-revenue gap, and most ended in significant economic washouts before the technology paid off.
2
Yield curve compression hid US borrowing costs The gap between 2-year and 10-year Treasury yields narrowed from 80-100 basis points when Gensler was at Goldman Sachs to just 40-50 basis points in recent decades, enabled by global central banks and China's savings. Gensler says this subsidy is ending — a reckoning in real interest rates is coming.
3
Hyperscalers' 'own cash' defense has limits A common argument is that this AI boom is safer because Google, Microsoft, and Meta are funding it from cash flows, not debt. Gensler pushes back: they are tapping debt markets, and off-balance-sheet financing through NeoCloud companies like CoreWeave creates hidden interconnections that could amplify a downturn.
4
Finance's share of GDP doubled — at a cost Finance grew from roughly 3% of the US economy in the 1950s to 8% today. Gensler argues this expansion correlates directly with rising wealth and income inequality and political polarization — more finance has not straightforwardly meant a better economy.
5
Think tasks, not jobs, for AI disruption Gensler argues that AI analysis at the job level is too coarse — what matters is which tasks within jobs get automated, then which full processes get transformed (as Ford transformed the factory floor). This distinction affects both how quickly disruption hits and which roles are actually at risk.
6
Information asymmetry still defines Wall Street winners Being at the center of a market — as Goldman Sachs and Morgan Stanley once were, and as Citadel and Jane Street now are — confers superior information flow. Gensler's 'fast deer vs. slow deer' framing captures how structural position, not just skill, determines who extracts value from financial markets.
7
US debt trajectory has no political solution Federal deficits run at 6% of GDP with debt at 100% of GDP ($31 trillion), yet both parties avoid addressing it because the bulk of spending — Social Security, Medicare, Medicaid — is politically untouchable. Gensler says only about 4% of GDP is discretionary spending, making arithmetic fixes nearly impossible without entitlement reform.