SpaceX is betting its $75 billion IPO on three engineering moonshots that have never been attempted at scale: building a fully reusable rocket, manufacturing AI satellites at twice its current production rate, and constructing a chip foundry in the United States. Financial analysts from Morningstar and New York University professor Aswath Damodaran both value the company significantly lower than its bankers' $1.8 trillion assessment, with the gap largely reflecting skepticism about whether SpaceX can deliver on these ambitious plans.

The core of SpaceX's business case centers on orbital data centers, a vision that emerged over the last 18 months as Elon Musk sought to unite his conglomerate before going public. Unlike traditional data centers on Earth, space-based facilities would theoretically offer unlimited solar power and avoid terrestrial real estate constraints. But pulling this off requires solving problems that have stumped the aerospace and semiconductor industries for decades.

What Are the Three Engineering Challenges SpaceX Must Overcome?

Musk laid out his vision in a recent video interview, arguing that SpaceX is uniquely positioned to tackle all three challenges simultaneously. However, each represents a formidable obstacle on its own.

• Fully Reusable Starship: SpaceX's Starship rocket is the linchpin of the entire plan, enabling the company to put massive amounts of mass into orbit cheaply. A recent test flight showed progress, but the booster stage failed to make a controlled reentry as planned, triggering a Federal Aviation Administration mishap investigation. The company currently expects to begin launching Starlink satellites with Starship by the end of 2026, though NASA, which has a nearly $4 billion contract with SpaceX to use Starship as a moon lander, still isn't ready to commit to a test mission scheduled for late 2027.
• AI Satellite Production at Scale: Musk claims SpaceX can reach an annualized production rate of roughly one gigawatt of computing power per year by the end of 2027, based on satellites delivering 150 kilowatts each. That translates to manufacturing approximately 6,666 AI satellites annually, or about 556 per month. For context, SpaceX's current Starlink production rate is roughly 70 satellites per week. While Musk says the AI satellites have simpler architecture, this production facility hasn't been built yet.
• Domestic Chip Foundry (Terafab): SpaceX plans to build Terafab, a chip manufacturing facility intended to produce the processors for its orbital data centers. Chip fabrication plants are among the most complex industrial projects ever attempted, typically costing billions of dollars and taking a decade or longer to construct. Even dedicated semiconductor companies have grown reluctant to undertake such projects in the United States.

Why Are Financial Analysts Skeptical of SpaceX's Valuation?

Morningstar assigns SpaceX a fair value of approximately $825 billion, with a share price of $63, compared to the company's offering price of $135. Damodaran, the NYU finance professor, suggests a valuation of $1.2 trillion. Both analyses reveal a stark divide in how they value different parts of SpaceX's business.

The company's space launch business and Starlink satellite internet network represent the most attractive and profitable segments. These operations have demonstrated revenue streams and relatively predictable margins. The AI business, by contrast, is where valuations diverge most dramatically. SpaceX's market analysis claims a $22.7 trillion opportunity in enterprise artificial intelligence, compared to just $2.4 trillion for AI infrastructure and under $2 trillion for its space efforts. Yet the company is simultaneously selling significant compute capacity to Anthropic and Google, competitors in the AI model business, which raises questions about whether SpaceX can dominate both the compute provider and model-builder roles simultaneously.

"This is not a promise of what we'll do. This is what we are going to try to do, and think we probably can do, which is to get to roughly an annualized rate of a gigawatt per year by the end of next year, in terms of space AI compute," said Elon Musk in a video interview released by SpaceX.

Elon Musk, Chief Executive Officer, SpaceX

How to Evaluate SpaceX's Investment Potential

• Compare Valuations Across Analysts: Morningstar's $825 billion valuation and Damodaran's $1.2 trillion assessment both fall well short of the $1.8 trillion implied by the IPO price. Investors should understand that the $72 per share difference between Morningstar's fair value and the offering price represents a bet on SpaceX's ability to execute its orbital data center vision, not a reflection of proven business fundamentals.
• Assess Execution Risk on Timelines: SpaceX claims it will reach one gigawatt of annual AI compute capacity by the end of 2027, roughly 18 months away. The company took a decade to build its Starlink manufacturing capability. Investors should weigh whether SpaceX can compress that timeline by half while simultaneously building a chip foundry and perfecting Starship reusability.
• Monitor Starship Reusability Progress: The entire financial model depends on Starship achieving rapid, full reusability. Recent test flights have not yet demonstrated that capability. Investors should track FAA mishap investigations, booster reentry success rates, and NASA's confidence in the vehicle before committing capital.
• Distinguish Between Space and AI Valuations: SpaceX's space business, including launch services and Starlink, is relatively mature and profitable. The AI data center business is speculative. Investors should consider whether they're comfortable paying a premium valuation for a company whose upside depends on an unproven technology that industry experts say is a decade away from viability at scale.

SpaceX's IPO arrives at a moment of genuine technological ambition, but also genuine uncertainty. The company is the only entity positioned to attempt these three engineering challenges simultaneously, which speaks both to its capabilities and to the magnitude of what it's attempting. Musk once said he wouldn't take SpaceX public until the company reached Mars, fearing that fickle investors might lose faith in long-term projects. The orbital data center vision may prove just as difficult to achieve, even if the timeline is shorter.

For public investors, the question is whether SpaceX's track record of delivering on seemingly impossible rocket engineering translates to success in satellite manufacturing, chip fabrication, and AI infrastructure. History suggests that betting against Musk has often been unwise, but so has paying premium valuations for companies dependent on executing multiple simultaneous moonshots. The IPO is reportedly deeply oversubscribed, with institutional investors committing billions, but the financial analysis suggests that enthusiasm may not align with the underlying business fundamentals.