A company has just filed to go public at a valuation of roughly $1.75 trillion. For context, that is larger than the entire GDP of Canada. The company lost $4.3 billion in a single quarter. It has accumulated $41.3 billion in losses since it was founded. Its hottest new business segment - the one investors are most excited about - lost $2.5 billion in the same quarter it is using to sell itself to the public.
That company is SpaceX. And today it filed the documents to pull off what could be the largest initial public offering - the process by which a private company sells shares to the public for the first time - in the history of financial markets.
The prospectus, the thick document companies are required to file with the US Securities and Exchange Commission before listing publicly, arrived this morning. It revealed for the first time the actual financial guts of a company that has spent 24 years being famously secretive. Investors who have been waiting years for this peek got one. It is not entirely reassuring.
The background
SpaceX was founded in 2002 by Elon Musk with the stated goal of making humanity a multi-planetary species. That is not the kind of sentence that usually precedes a serious financial pitch. But for two decades, the company somehow managed to keep private investors - venture funds, sovereign wealth funds, and pension managers - writing ever-larger checks, even as the losses mounted.
The business that actually emerged is a strange hybrid. SpaceX builds and launches rockets, including the Falcon 9, which has become the workhorse of commercial space launch. It operates Starlink, a constellation of roughly 10,000 satellites in low Earth orbit that beams internet service to users in remote areas, ships, and airplanes. It is developing Starship, a massive next-generation rocket intended to eventually carry people to the Moon and Mars. And now, following a merger completed in January 2026, it also owns xAI - Musk's artificial intelligence company, which makes the Grok chatbot and owns X, the social media platform formerly known as Twitter.
That merger, valued at a combined $1.25 trillion with SpaceX at $1 trillion and xAI at $250 billion, was completed just months before the IPO filing. The stated purpose was to pursue "orbital data centers" - putting AI computing infrastructure in space. The financial effect was to fold a heavily loss-making AI business into a rocket company and then take the whole thing public.
An IPO is essentially a sale. The company sells a portion of itself to public investors, who hand over cash in exchange for shares. The company gets the money it needs to keep operating and growing. The investors get a piece of the company's future profits - assuming there are any.
EBITDA - earnings before interest, taxes, depreciation, and amortisation - is the financial metric companies love to use when they want to make their numbers look better than they are. It strips out certain costs that are perfectly real. For a rocket company, that distinction matters more than almost anywhere else.
What is actually happening
Today, SpaceX filed its S-1 prospectus with the SEC, confirming plans to list on the Nasdaq under the ticker SPCX and targeting a June 12 debut. The roadshow - the process by which the company pitches itself to big institutional investors to set a final price - is reportedly scheduled to begin June 5.
The headline numbers, according to the filing: revenue of $4.694 billion in the first quarter of 2026, up 15% from the same period last year. Net loss of $4.27 billion in the same quarter, up from $528 million in Q1 2025. Operating loss of $1.943 billion. Adjusted EBITDA of $1.127 billion - a figure that required adding back nearly $2.5 billion in depreciation and amortisation costs to achieve.
For context, a net loss of $4.3 billion in a single quarter means SpaceX is losing money at a rate of roughly $17 billion a year. Apple, for comparison, generates about $100 billion in annual profit. The $37 billion in total losses cited in the filing - now revised to $41.3 billion on the balance sheet as of March 2026 - represents decades of burning through investor cash.
The filing breaks the business into three segments. Space - rockets, launches, NASA contracts - lost $662 million in Q1. Connectivity - Starlink internet - made $1.1 billion in profit. AI - xAI, Grok, X - lost $2.5 billion. The only part of the business that generates real cash is the satellite internet service. The rest is a drain.
Patrick Boyle, a professor at King's College London and former hedge fund manager, described the EBITDA figure as essentially meaningless for a company like SpaceX. Depreciation is not an accounting trick at a rocket company - it is the actual cost of the hardware. Starlink satellites last roughly five years before falling out of orbit. SpaceX has to constantly build and launch replacements just to keep the network running. Stripping out that cost to produce a cheerier number is, at minimum, misleading.
The filing also raised governance concerns. Musk is set to retain approximately 85% of the voting shares after the IPO, meaning public investors will own a financial stake in the company but almost no ability to influence its direction. Shareholder lawsuits, traditionally a mechanism for holding executives accountable, must go to arbitration rather than open court - a clause that limits investor recourse.
The prospectus itself is remarkable in tone. The phrase "light of consciousness" appears 10 times. The word "AI" appears more than 1,200 times. Future businesses mentioned as potential revenue sources include point-to-point passenger travel by rocket, in-orbit manufacturing, and asteroid mining. None of these are currently operational.
The money trail
The timing of this IPO is not accidental. SpaceX has been burning cash at extraordinary rates - the company lost $37 billion since its founding, and those losses accelerated sharply after the xAI merger. The venture capital funding model - where private investors absorb losses in exchange for the chance to cash out eventually - has limits. At a certain scale, even the deepest-pocketed private backers want liquidity. An IPO provides exactly that: it lets early investors sell their stakes to the public at what may be the peak of enthusiasm.
There is also a competitive timing dimension. OpenAI is also preparing to go public, and Anthropic has been in conversations about a potential offering. The combined capital demand from these companies is enormous. Investors who buy into SpaceX today may have less cash available to put into OpenAI six months from now. Getting to market first, while the sentiment around AI and space is still running hot, is a rational strategy - even if it means rushing out a prospectus with unflattering numbers.
The xAI acquisition also serves a second purpose. By folding a $250 billion AI company into SpaceX before the IPO, Musk created a combined entity where the loss-making AI segment is obscured within a larger balance sheet. The Space and Connectivity businesses together are, by some measures, viable. Presenting all three as a single company allows the profitable satellite internet business to carry the AI losses. Whether public investors notice this structure depends on how carefully they read the filing.
There is a structural mechanism that will help the stock regardless of the fundamentals. Under Nasdaq's revised listing rules, SpaceX will qualify for automatic inclusion in the Nasdaq-100 index after just 15 days of trading. The Nasdaq-100 is tracked by trillions of dollars in passive index funds - investment vehicles that automatically buy every stock in the index, whether or not their managers think it is a good idea. The moment SpaceX enters the index, those funds are obligated to buy shares. That forced demand provides a built-in buyer for early investors who want to exit quickly.
According to Fortune, Starlink generated more than $11 billion in revenue in 2025, nearly half of total consolidated revenue, and $1.2 billion in profit in Q1 2026 alone. That single business - satellites and internet - is the financial core of everything else. Without it, SpaceX is purely a capital-intensive science project.
What people are doing about it
For most retail investors - ordinary people rather than professional fund managers - the SpaceX IPO will be accessible through standard brokerage accounts once the stock begins trading on June 12. Some brokerages have already indicated they will allow clients to participate in the IPO itself, though allocation tends to favour institutional buyers.
Pension funds and index-tracking funds face a different kind of choice. Under the Nasdaq-100 fast-entry rule, any fund that tracks that index will automatically become a SpaceX shareholder within two weeks of listing, regardless of any fund manager's view of the valuation. For millions of people with retirement accounts in index funds, SpaceX exposure will simply arrive.
Professional investors have been more cautious. The filing, as noted by Morningstar, shows losses widening significantly year-over-year, with the AI segment driving most of the deterioration. Hedge funds and long-only equity managers who focus on fundamentals - cash flows, margins, return on capital - are working through numbers that do not obviously support a $1.75 trillion valuation.
The comparison made most frequently by sceptics is with WeWork, the co-working company that filed a similarly grandiose prospectus in 2019, full of philosophical language about elevating consciousness and transforming how people work. That IPO collapsed under scrutiny of its financials, and the company later went bankrupt. SpaceX's business is more substantial than WeWork's ever was, but the structural parallel - extraordinary valuation, founder control, losses obscured by adjusted metrics, and a mission statement that sounds more like a manifesto than a business plan - has not been lost on analysts watching the filing.
Governments are a separate category. SpaceX's relationship with NASA and the US Department of Defense means that a significant portion of its launch revenue is effectively guaranteed by federal contracts. Those contracts do not disappear if the stock falls. But they also do not scale fast enough to offset the AI losses at current burn rates.
The bottom line
SpaceX is asking the public to pay $1.75 trillion for a company where the only profitable division is satellites, the AI unit is bleeding $2.5 billion a quarter, and the founder controls 85% of the votes. The satellite business is genuinely valuable. Starlink has real customers, real revenue, and real margins. Everything else - rockets, AI, asteroid mining, point-to-point passenger transport - is either structurally unprofitable today or entirely notional. The IPO is arriving now not because the business is ready, but because the window is open, the market is hot, and the private investors want out. That is not necessarily a reason to panic, but it is a reason to read the filing very carefully before doing anything else.
Timeline
- 2002 - Elon Musk founds SpaceX in Hawthorne, California, with the stated goal of making humanity multi-planetary
- 2011 - NASA ends its Space Shuttle program; SpaceX becomes the agency's primary commercial launch partner
- March 2025 - xAI acquires X (formerly Twitter) in an all-stock transaction; Sullivan and Cromwell values the combined entity at $113 billion at the time
- November 2025 - xAI raises $20 billion in a funding round valuing it at $230 billion; xAI's quarterly losses reach $1.46 billion
- January 30, 2026 - SpaceX completes acquisition of xAI in an all-stock deal valued at $1.25 trillion, described as the largest private merger in history
- February 2026 - Following the merger, xAI co-founders begin departing and the company restructures into four development teams
- April 2026 - SpaceX confidentially files its S-1 with the SEC for initial review
- May 20, 2026 - SpaceX publicly files its IPO prospectus with the SEC, revealing $4.694 billion in Q1 revenue, a $4.27 billion net loss, and $41.3 billion in accumulated losses
- June 5, 2026 - Roadshow with institutional investors expected to begin
- June 12, 2026 - SpaceX expected to begin trading on the Nasdaq under ticker SPCX
Summary
Who: SpaceX, controlled by Elon Musk, alongside early private investors seeking liquidity
What: A public filing for an IPO targeting up to $75 billion in new capital, at a valuation of approximately $1.75 trillion - which would make it the largest IPO in history - despite a net loss of $4.27 billion in Q1 2026
When: The prospectus was filed today, May 21, 2026; trading is expected to begin June 12, 2026
Where: The company will list on the Nasdaq under the ticker SPCX; SpaceX is headquartered in Starbase, Texas
Why: Private venture capital funding for a company that has lost $41.3 billion since inception is finite; the IPO lets early investors convert paper gains into cash while public demand for AI and space stories remains high, and before competing IPOs from OpenAI and Anthropic absorb available investor capital