‘Big Short’ investor Steve Eisman: I'm 'not a fan' of the upcoming SpaceX IPO

Key Concepts

• Capital Intensity: The shift from "asset-light" software models to "asset-intensive" infrastructure models requiring massive capital expenditure (CapEx).
• Commoditization of AI: The argument that Large Language Models (LLMs) lack "moats" (competitive advantages) and are becoming interchangeable commodities.
• Total Addressable Market (TAM): The theoretical maximum revenue opportunity for a product or service.
• Token Economics: The shift from subsidized AI usage to charging users the actual cost of compute tokens.
• Monopoly Rents: The ability of a company with a dominant market position (like SpaceX in space launch) to charge premium prices due to lack of competition.

---

1. The SpaceX IPO and Financial Analysis

Steve Eisman expresses skepticism regarding the upcoming SpaceX IPO, primarily based on his analysis of the company’s S-1 prospectus.

• Capital Expenditure (CapEx) Surge: Eisman highlights a dramatic shift in the company's financial structure. In fiscal year 2023, CapEx was 42% of revenue; by the first quarter of the current year, it had ballooned to 215% of revenue.
• Strategic Pivot: He argues that SpaceX is no longer just a space and satellite (Starlink) company but has pivoted heavily into AI. He notes that 85% of the company’s stated $28.5 trillion Total Addressable Market (TAM) is tied to AI, which he views as a highly capital-intensive and risky endeavor.
• Critique of "Grok": Eisman dismisses Elon Musk’s AI model, "Grok," as not being at the "leading edge" of the industry.

2. The Transformation of the Tech Sector

Eisman posits that the entire technology sector is undergoing a fundamental transformation from "asset-light" to "asset-intensive" business models.

• Capital Raising: He points to Google’s recent $80 billion capital raise as evidence that even cash-rich tech giants can no longer fund their AI ambitions through internal cash flow alone. He suggests that other major players like Meta and Microsoft will likely follow suit.
• Lack of Competitive Moats: A central argument is that current AI outputs are not "differentiable." Because users switch between models like ChatGPT frequently, there is no brand loyalty or structural barrier to entry, leading to the commoditization of AI services.
• Token Pricing: Historically, companies have subsidized AI usage by charging below the cost of compute tokens. Eisman notes that this is changing (e.g., Microsoft’s GitHub), and he predicts that once users are forced to pay the "actual cost," demand may drop significantly, questioning whether the business model can "thrive."

3. The "Elon Musk" Factor and Potential M&A

The discussion touches on the risks and rewards of betting against Elon Musk.

• Monopoly Power: The interviewer (Andrew) argues that SpaceX holds a monopoly in space launch, which should allow for "monopoly rents." Eisman acknowledges this but maintains that the company's current valuation is predicated on its AI ambitions rather than its space-launch success.
• Potential Tesla Acquisition: Eisman speculates that Musk might use SpaceX’s stock to acquire Tesla, creating a single entity ("X"). He views this as a negative for SpaceX shareholders, citing Tesla’s declining earnings over the last four years and the intense, low-margin competition from Chinese EV manufacturers.

4. Notable Quotes

• "The whole tech sector is being transformed because it's so... from an asset-light business to an asset-intense business." — Steve Eisman
• "For all the money that's being spent, what's being created, it seems to me at this point is a commodity." — Steve Eisman
• "What I love about the S-1 is that it reads like a science fiction novel." — Steve Eisman (referring to the $28.5 trillion TAM projection).

---

Synthesis and Conclusion

Steve Eisman’s perspective is one of extreme caution regarding the current AI investment frenzy. He argues that the market is ignoring the reality of "asset-intensive" business models, where the cost of compute is rising while the product itself is becoming a commodity. While he acknowledges the historical difficulty of betting against Elon Musk, he concludes that the valuations presented in the SpaceX prospectus are disconnected from reality, relying on a massive, speculative TAM that is largely unproven. He chooses to avoid the investment entirely rather than shorting it, suggesting that the "good news" is already over-baked into current market expectations.