I’ve watched plenty of hyped IPOs march to market with a great story and a shaky balance sheet. Few have arrived with as much noise as SpaceX.
Elon Musk‘s company filed for an IPO confidentially this spring, and Wall Street is already treating it like the listing of the decade.
CNN joked that the prospectus reads like a beach-read thriller, and in a way, it does. Over more than 270 pages, it lays out the ambitions of a company that builds rockets, beams internet from space, and now runs AI data centers.
The hype is real. So are the warning signs. Here are five reasons the SpaceX IPO could land with a thud instead of a bang.
Inside the biggest IPO Wall Street has ever seen
SpaceX is no longer just a rocket company. After Musk merged his AI startup xAI into SpaceX earlier this year, the combined business was valued at roughly $1.25 trillion, according to CNBC.
The company now spans rockets, the Starlink satellite-internet service, the social network X, and a fast-growing AI unit.
That matters to everyday investors because this will likely be the largest public listing ever, and many will be tempted to buy in.
Before you do, it helps to know what you are actually paying for. SpaceX brought in $18.7 billion in revenue in 2025, per its filing.
That sounds massive, until you see how much it spends to keep the lights on.
Where the SpaceX IPO math starts to wobble
Reports suggest SpaceX is aiming for a valuation near $1.77 trillion.
However, research firm Morningstar pegs the company closer to $780 billion, which is a discount of 50% to its IPO price.
Most of the lofty number rides on what SpaceX might become and not on what it earns today.
The company claims a total addressable market of $28.5 trillion, most of which (about 90%) is tied to the AI segment.
Related: Dan Ives spills the beans on SpaceX future
Strip out the dream, and you are left with rockets and Starlink.
The second problem is cash burn. SpaceX lost almost $5 billion in 2025, and the losses grew by another $4.3 billion in the first quarter of this year, CNN reported.
The bleeding traces largely to the AI division, which lost $6.4 billion last year while bringing in just $3.2 billion. You can check the full math in the company’s S-1 filing with the Securities and Exchange Commission.
The third problem is trust.
Days before the roadshow, Musk posted on X that the company’s big AI deal with rival Anthropic was really a short-term lease that either side could cancel on 90 days’ notice, a detail the 300-plus-page prospectus never spelled out, CNBC reported.
The filing instead described Anthropic paying $1.25 billion a month through 2029. One of those versions is incomplete.
Referring to Anthropic in a post on X, Musk stated:
“We won’t leave them hanging and will provide a reasonable off-ramp. But if compute gets super tight I said we might need it back at some point.”
According to CNBC:
- Columbia Law School professor Eric Talley summed up the bind for investors: either the filing misleads, or Musk is up to his usual antics.
- PitchBook analyst Franco Granda also flagged missing details on subscriber churn, rocket unit economics, and how much of that AI compute is actually being used.
AFP/Getty Images
What history says about mega-IPOs like SpaceX
The fourth reason is the cruelest, because it is just math.
A long-running study of more than 9,200 IPOs by University of Florida finance professor Jay Ritter found that the biggest companies tend to disappoint.
Firms with more than $1 billion in sales before going public posted an average three-year return of -2.1% versus the broader market, per Ritter’s 2026 data.
Tech deals priced at more than 40 times sales fared even worse as 12 of 14 trailed the market over three years.
In plain terms, when everyone already loves a giant, pricey stock, there is little room left for it to climb.
More AI:
- Micron sits at the center of a red-hot chip rally
- IBM CEO sends blunt message on AI and quantum computing
- Anthropic CEO makes shocking admission about AI
The fifth reason is control. The filing shows Musk holds super-voting Class B shares that give him about 85% of the shareholder vote, per CNN. Translation: Musk would have to vote to fire himself.
Public investors take on the risk but get almost no say. Ritter’s research notes that nearly half of all tech IPOs in 2025 used these dual-class setups.
To be fair, there is also a bull case.
Ark Invest’s Cathie Wood argues the Anthropic deal turns those costly data centers into a profit engine.
Starlink is profitable, earning $4.4 billion in operating profit last year.
And if the AI bet pays off, today’s price could look cheap in hindsight.
But “if” is doing a lot of heavy lifting. SpaceX itself admits in the filing that some of its plans rely on technology that does not yet exist.
For an IPO this size, that is a lot of faith to ask of investors.
Related: The SpaceX–OpenAI IPO boom: What investors must do now
