CROSSPOST: NOAH SMITH: The AI Bust Scenario That No One Is Talking About
Deep Dives
Explore related topics with these Wikipedia articles, rewritten for enjoyable reading:
-
Panic of 1873
12 min read
Linked in the article (16 min read)
-
United States v. Microsoft Corp.
13 min read
The article references Microsoft's actions against Netscape as a template for how tech giants might crush AI competitors. This landmark antitrust case provides essential historical context for understanding platform monopoly tactics and 'embrace, extend, extinguish' strategies that could repeat in AI markets.
-
Contestable market
9 min read
The Airline Scenario hinges on the economic concept that 'markets are contestable and increasing returns largely absent,' leading to prices being pushed to marginal cost. Understanding contestable market theory explains why profitable monopolies may be impossible even for useful technologies.
Think of airlines: very useful, very high-tech, immense user surplus, next to no profits. Even if AI works, and even if it gets adopted very fast, it might not make much in the way of profits for anyone. The part of Noah Smith’s post above his paywall...
Very much worth reading, because, I think, highly likely to be correct.
Noah sees three ways that the current AI-bubble and MAMLM-capex buildout could spectacularly crash:
(1) The Virtual Reality Scenario is if “AI” as we know it is just a useful enough technology to justify the capital-expenditure build-out, given that it is not write-once run-everywhere for-all-time, but rather train and infer, train and infer, train and infer again and again and again without mammoth economies of scale.
(2) The Railroad Scenario is if the financing seizes up as the profits do not appear in time before those who have been shorting the build-out and those who own the leveraged debt decide it is time to take their profits and cash-in their chips, respectively. The historical model is the Panic of 1873 that took down Jay Cooke & Co. In this case the sector and the technology are fine, but only after a very disruptive five-year final AI-Winter.
(3) The Airline Scenario is one of a very useful high-tech technology deployed at scale, but because markets are contestable and increasing returns largely absent, there are next to no profits as prices keep getting pushed down to short-run marginal cost.
This last is, I think, the most likely. And Noah thinks so too: The technology works. It diffuses. It becomes infrastructural. Yet the rents flow not to the operators, and in the long run not that many rents flow to the toolmakers and the suppliers. Instead, what rents there are flow to the downstream complementors. And the overwhelming bulk of value flows to customers in the form of user surplus. It is glorious (provided our AI can preserve our attention from being hacked to our detriment by Zuckerberg’s AI)! But it is not profitable for investors! And it is profitable for lucky entrepreneurs who keep their wits about them and understand that their trees will not grow to the sky. And it is profitable for only those VCs who understand that they are basically rerunning their crypto grift, and prioritize exit while the getting is still good and the crash has not yet come
There ...
This excerpt is provided for preview purposes. Full article content is available on the original publication.