Musings on Markets (Aswath Damodaran) ·
The AI Trade Deconstructed: AI Builders, AI Suppliers and AI Product firms!
By Aswath Damodaran
ChorusBring the market's conversation to your AI. Learn more
Summary
Aswath Damodaran breaks the AI trade into three groups: AI suppliers (chipmakers and infrastructure sellers who book capex as current revenue), AI builders (hyperscalers like Meta, Microsoft, Alphabet and Amazon, plus LLM developers, who expense spending over time and face negative free cash flow), and AI product firms, which turn profitable last. He argues collective AI product and service revenue of roughly $250 billion is small against the money invested, so the premise depends on customers growing into trillions of dollars of spending. He values Anthropic at about $1.25 trillion in a sensitivity test, showing that higher compute needs and a 7.5% regulatory margin drag cut value even with $1 trillion in target revenue, and he remains bearish on its IPO pricing as market mood has soured. He also flags election-driven risks to US data center growth, especially in Texas, and possible tighter AI rules from Congress, while describing overinvestment and overpricing as a feature of big market change that will reward those who spot overvalued AI names.
Key takeaways
- Aswath Damodaran (uncertain on Anthropic): Anthropic's equity value falls to $1.25T if compute costs and regulatory drag rise.
- Aswath Damodaran (bearish on Anthropic): Anthropic's IPO offer price faces growing pressure to fall as market mood darkens.
- Aswath Damodaran (neutral on Space Exploration Technologies): SpaceX's AI arm is more an AI builder than a product firm, leasing compute.
- Aswath Damodaran on AI bubble & tech valuations: AI has almost certainly led to overinvestment, overcommitment to compute and aggregate overpricing.
- Aswath Damodaran on AI infrastructure & data centers: Hyperscalers' AI capex is hitting free cash flow hard, with legacy businesses providing a buffer.
What was said, by company and theme
10 views from 1 speaker. Every quote is checked against the post's text (how).
Anthropic
Across the shows: 104 speakers in other episodes, 140 bullish and 91 bearish calls, past 30 days
Where everyone stands on Anthropic
Aswath Damodaran · NYU Stern School of BusinessUncertain
Anthropic's equity value falls to $1.25T if compute costs and regulatory drag rise.
Damodaran tests how sensitive his Anthropic valuation is to post-election changes. Raising the compute investment needed for future growth (sales-to-capital ratio cut from 3 to 2.5) and raising the regulatory margin drag (from 5% to 7.5%) lowers estimated equity value to $1.25 trillion, even with target revenues held at $1 trillion. He also notes Anthropic's rush to lock in compute commitments before the rules changed may give it a moat against competitors, so the net effect is uncertain.
“Just to give you a measure of how sensitive the Anthropic valuation is to these changes, increasing how much Anthropic will have to invest to add compute power in future years (by lowering the sales to cap ratio from 3 to 2.5) and increasing the drag on margins from regulation (from 5% to 7.5%) reduces the estimated equity value to $1.25 trillion”
Aswath Damodaran · NYU Stern School of BusinessBearish
Anthropic's IPO offer price faces growing pressure to fall as market mood darkens.
IPO pricing is a mood and momentum game, and the mood has darkened for AI and for the broader market over the last two weeks. Bankers worried about overpricing a trillion-dollar-plus offering and then being left with the shares have strong reasons to push the offer price lower, so the pressure to cut the price only grows.
“The truth is that pricing is a mood and momentum game, and it is undeniable that the mood has darkened not just for AI, but also for the market in the last two weeks. If you consider bankers who are terrified of over pricing a trillion-dollar plus offering and picking up the pieces, the pressures to lower the offer price only get stronger.”
Space Exploration Technologies
Across the shows: 52 speakers in other episodes, 68 bullish and 24 bearish calls, past 30 days
Where everyone stands on Space Exploration Technologies
Aswath Damodaran · NYU Stern School of BusinessNeutral
SpaceX's AI arm is more an AI builder than a product firm, leasing compute.
Damodaran included SpaceX in the AI product group, but almost all of its AI-arm revenue currently comes from leasing compute power to Anthropic and Alphabet. That makes it more an AI builder than an AI product company, so he treats it that way.
“I have included SpaceX as an AI product company, but at the moment, almost all of its revenues on its AI arm come from leasing out its compute power to Anthropic and Alphabet, making it more of an AI builder than AI product company.”
AI infrastructure & data centers
Where everyone stands on AI infrastructure & data centers
Aswath Damodaran · NYU Stern School of Business
Hyperscalers' AI capex is hitting free cash flow hard, with legacy businesses providing a buffer.
AI builders such as Meta, Microsoft, Alphabet and Amazon see the immediate effect of their infrastructure spending as negative for both earnings and cash flow, however justified the build-up is over time. Their legacy and cloud businesses have muted the earnings impact, but the free cash flow hit has been real. The market has been patient so far, but it may run out of patience while waiting for payoff.
“For AI builders, no matter how merited the build-up (in terms of delivering value over time), the immediate effect on both earnings and cashflows will be negative ... but the free cash flow impact has been real.”
Aswath Damodaran · NYU Stern School of Business
Election results signal voter aversion to data centers that will slow US data center growth.
Damodaran reads the election results as a clear voter message against data centers. Texas hosts about a third of the new data centers coming online in the near term. Even if construction there does not stop, the restrictions will slow data center growth, at least in the United States.
“The most critical of these were the election results, where voters sent a clear message about their aversion to data centers. In Texas, which is home to about a third of the new data centers coming online in the near term, it may not stop existing data center construction, but it will slow the growth of data centers, at least in the United States.”
Aswath Damodaran · NYU Stern School of Business
AI buildout has lifted aggregate US corporate earnings and supported stock prices despite rising rates.
Suppliers report the infrastructure capex as current-year revenue, which, net of their operating costs, raises their earnings. Builders expense the capex over time, so their earnings drag is spread out. In Damodaran's earlier analysis, the net effect across US companies was higher aggregate earnings over the last year, which supported stock prices even as interest rates soared.
“The suppliers report these same capital expenditures as revenues in the current year, and when netted out against their operating costs, increase earnings at these companies. I wrote about how the net effect of the infrastructure buildout has been an increase in aggregate earnings at US companies in the last year, and has supported stock prices, even as interest rates have soared.”
AI: overall outlook
Where everyone stands on AI: overall outlook
Aswath Damodaran · NYU Stern School of Business
The easy AI trade is mostly done; AI groups' interests and prices will increasingly diverge.
Suppliers see revenue first, builders face a negative cash flow effect, and product companies become profitable last. Overbuilding helps suppliers but hurts builders and their shareholders, and long leases raise default risk for product companies. As these competing interests overwhelm shared ones, the market will differentiate winners and losers. In 2026, strength in some AI groups has masked weakness in others.
“I do believe that the easy AI trade, where you could buy any company that was part of the AI ecosystem ... and make money from a general optimism about AI's potential is mostly done”
Aswath Damodaran · NYU Stern School of Business
AI product revenues of about $250B are tiny against the trillions invested, with little yet to show.
Even using leaked ARR figures from the lead players, collective AI product and service revenue is about $250 billion. That is small relative to the money invested in the space. The ecosystem's value depends on customers becoming much larger in reach and spending, in the trillions of dollars, so the gap between current revenue and the premise is wide.
“Even if you take the ARR numbers that are being leaked by the lead players, the collective revenues from AI products and services right now is about $250 billion, a small number relative to what has been invested in the space.”
AI bubble & tech valuations
Where everyone stands on AI bubble & tech valuations
Aswath Damodaran · NYU Stern School of Business
AI has almost certainly led to overinvestment, overcommitment to compute and aggregate overpricing.
Damodaran's 'big market delusion' argues that a big market's potential, combined with overconfidence from founders, VCs and CEOs, leads to overinvestment and then correction. He believes AI has produced overinvestment by builders, overcommitment to compute by LLM companies, and overpricing by the market in aggregate. The correction will hurt unevenly, but it will create opportunities for those who find AI companies overpriced. He sees this as a feature of big change rather than a bug.
“I believe that AI, because of its reach and potential, almost certainly has led to over investment, by AI builders, over commitments to compute power by LLM companies and over pricing, by the market, at least in the aggregate.”
AI regulation & safety
Where everyone stands on AI regulation & safety
Aswath Damodaran · NYU Stern School of Business
A newly elected Congress may tighten AI rules, adding a bigger drag on margins.
Damodaran says it is possible the newly elected Congress will write tighter AI restrictions. He argues these would be a larger drag on AI companies' margins. He presents this as a possibility rather than a forecast, and it feeds into his Anthropic valuation sensitivity.
“It is also possible that the newly elected congress will write in tighter restrictions on AI, creating a bigger drag on margins.”
Across the shows
What other episodes Chorus covered said about the same companies in the 30 days up to this one.
- Anthropic104 speakers · 140 bullish, 91 bearish calls
- Space Exploration Technologies52 speakers · 68 bullish, 24 bearish calls
Topics
- AI capex and aggregate US earnings
- AI compute leasing
- AI overinvestment and overpricing
- AI regulation
- AI revenues versus investment
- AI value chain divergence
- IPO pricing
- data center siting and permits
- hyperscaler capex and free cash flow
- valuation sensitivity to compute and regulation
More from Musings on Markets (Aswath Damodaran)
- The Cost of Human Frailty: Nvidia's Jensen Huang Test! · November 19, 2026
- The Rocky Road from Aggregate to Per Share Value: Anthropic's IPO update · November 19, 2026
- The Anthropic Feedback Loop: The Cost of Capital Delusion and Diversion! · November 3, 2026
- The Anthropic Prospectus Arrives: An IPO valuation! · October 28, 2026
Ask your AI about this post
Chorus gives Claude, ChatGPT and other assistants every view from this post and every other show Chorus follows. Ask who agrees, who changed their mind, and what else was said about the same companies.
Chorus reports what speakers said. It isn't investment advice. How Chorus works · About Chorus