The $1.5 Trillion Rocket: Inside the SpaceX IPO Valuation

01The premise, checked

A common claim about SpaceX is that it makes no money. The headline number behind that claim is real: SpaceX reported a GAAP net loss of $4.94 billion for 2025, and a single quarter, Q1 2026, added another $4.28 billion in losses.^[1] Read alone, those figures describe a company burning cash at industrial scale. Read with context, they describe something more specific, and the context changes the conclusion.

02The profit picture and the xAI twist

In February 2026, Elon Musk merged xAI, his artificial-intelligence company, into SpaceX at a combined valuation of $1.25 trillion, with xAI carried at roughly $80 billion.^[2] That merger is where the losses live. xAI lost more than $6 billion in 2025 and burned another $2.5 billion in the first quarter of 2026, training models and buying compute.^[1] The consolidated entity reports those AI losses on the same statement as the rockets, so a profitable launch-and-internet business and a deeply unprofitable AI business net out to a loss.

03The valuation in context

SpaceX filed its S-1 on May 20, 2026, targeting a $1.75 trillion valuation and a raise of about $75 billion, which would rank among the largest public offerings ever attempted.^[3] At that price the company would be worth more than the traditional listed U.S. aerospace-and-defense primes combined, a private rocket maker valued above the entire incumbent industry it competes with. A number that size is a claim about the future of launch, satellite internet, and space-based computing, not a multiple of current earnings.

04The multiple only works as a bet

Against roughly $11 billion in Starlink revenue plus launch income, $1.75 trillion implies a multiple in the range of 80 times sales, an order of magnitude above what a mature industrial company commands.^[4] Measured against the profit the space business actually throws off, the multiple runs into the thousands. By the standards of an established company those numbers are indefensible. They are defensible only as venture pricing applied at trillion-dollar scale: investors are paying for Starship reaching routine flight, Starlink compounding its subscriber base, and a plausible monopoly on heavy launch for a decade.

For comparison, the most richly valued large public companies trade at perhaps 10 to 30 times sales at the top of a bull market, and they do it with proven, growing profits. SpaceX is asking for several times that on a consolidated statement that currently shows a loss. The gap is the size of the bet. Buyers are not pricing the company that exists; they are pricing the company they expect to exist in 2035, and discounting almost none of the risk that it does not arrive.

05What the company actually provides

The case for the price rests on assets that already exist. SpaceX flies the only reusable orbital-class rockets in regular service and launches more mass to orbit than the rest of the world combined, which gives it pricing power over commercial, civil, and military customers who have nowhere comparable to go.^[6] Starlink is the largest satellite constellation ever built, serving roughly 10 million subscribers at about $81 a month and generating $11.4 billion in 2025 revenue, a business that did not exist at scale five years ago.^[5] Starship, if it reaches routine operation, lowers the cost of putting mass in orbit by another order of magnitude and resets the economics of everything downstream of launch. Those are real, hard-to-replicate advantages, which is why serious investors entertain the number at all rather than dismissing it.

06Bull case, bear case

The bull case is that SpaceX owns the road to orbit during the decade space becomes infrastructure, that Starlink becomes a global utility, and that xAI's losses convert into a defensible AI position with captive launch and power. The bear case is that the $1.75 trillion price front-runs all of that, that xAI is a $6-billion-a-year hole bolted onto a healthy business to flatter one valuation with another, and that a single founder's control concentrates execution risk in one person. Both cases read the same balance sheet. The disagreement is entirely about the future.

07Is the price rational?

The honest answer is that it is rational only on the company's most optimistic timeline. The "no profit" criticism misreads the accounting: the space business earns money, and the loss is an xAI artifact. The valuation criticism lands harder. At $1.75 trillion the offering prices in a future that has not happened yet, and pays in advance for Starship, Starlink saturation, and an AI franchise that is currently a cash sink. For a buyer who believes that future arrives on schedule, the price is a bet worth making. For everyone else, it is a very expensive way to own a very good rocket company.