Best of BiggerPockets Money: Aswath Damodaran on the $10 Trillion AI Bet

The Unprecedented AI Factory Build [0:00]

• The collective investment in AI infrastructure, including chips, power, and data centers, is estimated to exceed $2 to $2.5 trillion.
• This massive buildup, dating back to November 30, 2022, the launch of ChatGPT, is the largest in business history, surpassing railroad and dot-com investments.

  "So there's this architecture companies that make their money in building the factory."

Separating Architecture from Products [1:15]

• It's crucial to distinguish between companies building the AI architecture (e.g., Nvidia) and those aiming to profit from AI products and services (e.g., Meta, Alphabet, Amazon).
• Adding market caps of all involved companies contaminates the analysis, as some profit from building the infrastructure while others hope for returns from its output.

  "The first thing to do is step back instead of adding up all of those market caps because some of these companies make their money from building the architecture and some hope to make their money from the products and services that come out of the architecture is to separate the companies and look at it differently."

The Revenue Gap [3:20]

• While the AI "factory" cost $2 trillion, current collective revenues from AI products and services are only about $250 billion over the last 12 months.
• If revenues plateau at $250 billion, the $2 trillion investment would largely be written off.

  "You have 250 billion in revenues. Let's play the worst case scenario. If revenues level off at 250 billion, that 2 trillion is almost entirely going to be written off."

The $10 Trillion Revenue Target [4:20]

• To justify the $2 trillion upfront investment, AI business revenues would need to climb to an estimated $8 to $10 trillion in a steady state, considering expenses, taxes, and profit margins.
• The central question becomes whether the AI product and service market can realistically generate $10 trillion in revenue with decent profit margins.

  "The revenues will have to be maybe $8 trillion, $10 trillion. Because if you think about the profits and, I mean, the expenses to produce the revenues and the profits and the tax and the profits, you're very quickly going to start to scale the ladder up to $10 trillion."

Plausibility vs. Probability [5:30]

• Achieving $10 trillion in AI revenue is "possible" but likely pushes the limits of "plausible" and is currently "improbable."
• Damodaran cautions against declaring it a bubble without sufficient data, emphasizing the need for a "3P test": Possible, Plausible, Probable.

  "Right now it's possible. You can have a $10 trillion revenue. And I'll explain why that possibility exists. Is it plausible? You're probably pushing the limits of what's plausible with $10 trillion. And is it probable? Right now I'd attach a low probability to it."

Investor Exposure and Risk Containment [7:00]

• Profitable existing businesses (like Alphabet's advertising) can cushion AI investment write-offs, with losses primarily impacting shareholders.
• Companies without such buffers (like CoreWeave, which has significant debt) face greater risk of financial distress if AI investments fail.

  "The damage will be limited there. With CoreWeave, it gets messier. The reason it gets messier is if the factory has to be written down, CoreWeave is enough debt that you worry about not being able to make debt payments."

Investment Strategy Amidst AI Uncertainty [10:00]

• Investors face a choice: exit AI-related stocks, potentially missing future gains, or ride the wave, accepting potential losses during a correction.
• Market timing is perilous; staying out of the market for too long after exiting can lead to greater losses than enduring a correction.

  "The good news for you then is when the correction comes, you're going to be hurt less. But the bad news is it could take three years for the correction to happen. And what you lose while you wait might vastly exceed what you benefit."

Personal Investment Approaches [15:00]

• Damodaran advocates for diversification and avoiding over-concentration, using a 15% cap on any single investment.
• He sold Tesla due to its political entanglement and NVIDIA despite its massive gains, as it became excessively priced.

  "I sold Tesla right after the election because I don't like to own companies that become political as well as business plays."

The AI Revenue Ceiling: Tool vs. Displacer [23:00]

• If AI primarily acts as a productivity tool, its revenue potential is limited, perhaps around $2 trillion.
• If AI displaces a significant portion of the global workforce (currently earning $26 trillion in salaries), it creates macroeconomic instability, as displaced workers lose consumption power, making a $10 trillion+ market improbable.

  "So when people talk about total addressable markets, 26 trillion is the absolute limit. But it's got to be much lower than that because you have to have a sustaining economy where people really have jobs."