Oct 31, 2026 · 26 min
China’s AI correction exposes the bubble’s weakest assumptions
China’s AI crash is warning sign for the AI bubble | Dean Baker
The episode tests whether AI valuations can survive falling prices, uncertain demand, and infrastructure spending that may outrun revenue.
- 1China’s AI sell-off reflects a broader problem: falling token prices make extraordinary profits harder to sustain.
- 2AI companies may face commodity-like economics as competition erodes pricing power and customer loyalty.
- 3A large share of AI spending circulates within the sector, leaving outside demand to justify its enormous investment.
Don't miss
Dean Baker argues that AI spending may be circular because many major buyers are themselves AI companies, leaving outside demand unproven.
The brief
Dean Baker opens by questioning how AI companies can generate the revenues needed to justify their valuations, using China’s sharp stock-market correction as the test case.
China’s AI boom combined speculative enthusiasm, government involvement, and intense competition; Baker compares its collapse with the dynamics of the U.S. technology bubble.
Falling token prices expose the central business problem: companies need unusually high prices to support lofty valuations, but competition is pushing those prices down.
The conversation compares AI with airlines: potentially useful, widely adopted, and still a low-margin commodity, with uncertain applications and little customer loyalty.
Baker questions whether China and the United States are building more data-center capacity than demand can support, especially amid energy constraints and local opposition.
The sharpest warning comes when Baker notes that major AI spenders are often AI companies themselves, creating a circular flow that still needs outside revenue.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.