SpaceX IPO filing lays out a $1.75 trillion bet on Mars, AI and Musk control

SpaceX is not really selling rockets. At a proposed $1.75 trillion valuation, with Elon Musk locking in just over 85% of the voting power before the doors open, what the company is actually offering public investors is a permission structure — a way to price a future economy that does not yet exist.

That is the strange thing about reading the prospectus. The numbers describe one company; the pitch describes another. SpaceX generated $18.7 billion in revenue in 2025 but still lost $4.9 billion, according to The Guardian’s summary of the prospectus. The business is now framed as three parts: the original rocket-launch operation, the Starlink satellite-broadband unit, and an AI arm built around xAI and X.

And sitting on top of all three, by design, is a governance structure built to keep ordinary shareholders at arm’s length from any decision that actually matters.

That structure matters because the investment case is no longer simply that SpaceX can launch rockets more cheaply than its rivals. The more ambitious claim is that the company can turn its space infrastructure into the foundation for markets that are either early, unproven or not yet commercially real.

A profitable Starlink story wrapped inside a much bigger bet

The most solid business inside SpaceX appears to be connectivity. The Guardian reported that Starlink was the only profitable segment in the first three months of 2026, while the broader company remained loss-making. The AI division, by contrast, lost $6.4 billion last year, largely because of the costs of running and building the systems behind Musk’s Grok tool.

There is a pattern here that anyone who watched the late-stage growth wave of the last decade will recognise: a profitable core unit subsidising a constellation of bets that may or may not pay off, all bundled into a single equity story. Investors are not only buying a satellite-internet company or a launch-services company. They are being asked to buy a combined space, connectivity and AI empire whose biggest upside depends on technologies and markets that are still being built.

The future-markets section is the real story

The prospectus reportedly points to future markets that include space tourism, manufacturing and energy production on the moon and Mars, and asteroid mining. The same Guardian report notes that the prospectus acknowledges these markets “do not exist today.”

That is the tension at the heart of the filing. SpaceX already has real revenue, real customers and one of the most important launch businesses in the world. But a $1.75 trillion valuation depends on a much larger argument: that the company can expand from launch and broadband into entire industries that do not yet generate meaningful commercial revenue.

One example is point-to-point rocket travel on Earth. Musk floated that idea years ago, when The Verge reported in 2017 that he had proposed using SpaceX’s interplanetary rocket system for long-distance travel between cities on Earth. It was a striking idea then. In an IPO context, it becomes part of a much bigger question: how much public-market value should be assigned to businesses that still require enormous technical, regulatory and consumer-behaviour leaps?

The AI business adds scale, and risk

The filing also brings Musk’s AI ambitions directly into the SpaceX story. Business Insider reported that SpaceX and its recently acquired xAI business bought hundreds of millions of dollars of Tesla products in 2024 and 2025, including Megapack batteries and Cybertrucks, in transactions listed as related-party dealings. Those ties show how tightly Musk’s companies are now intertwined, and they may make operational sense in some cases. But for public investors, they raise a familiar governance question: when the same founder sits at the centre of multiple companies doing business with one another, who decides whether the terms are best for the company being bought into?

Musk keeps control

Picture an investor sitting with the prospectus open on a screen, scrolling toward the share-class section. The Guardian reported that Musk will control just over 85% of SpaceX’s voting power through Class B shares, which carry 10 votes per share compared with the Class A stock available to other shareholders. The arithmetic is unforgiving: even if every outside investor in the float voted in unison, they would still not move the needle on any decision Musk did not want moved. The public float, on those terms, offers economic exposure without much practical influence. Shareholders can buy into the SpaceX story, but they are unlikely to have meaningful power over board control, executive accountability or the strategic direction of the company. The compensation structure underlines the point. The Guardian reported that Musk has been granted 1 billion Class B shares that vest if SpaceX achieves the establishment of a permanent human colony on Mars with at least 1 million inhabitants, alongside market-cap targets that stretch as high as $7.5 trillion. Another tranche is tied to goals including space-based data centres.

The pitch is huge because the story has to be huge

SpaceX is not a conventional aerospace IPO. It is not being valued like a defence contractor, a broadband provider or a launch-services company. It is being presented as an infrastructure company for a future economy that may include orbital computing, AI, Mars settlements, off-world manufacturing and resources extracted beyond Earth.

That is what makes the filing so consequential. SpaceX has already changed the economics of launch and built a global satellite-internet business. But the proposed valuation asks investors to believe that those achievements are only the foundation for something much larger.

The risk is not that SpaceX lacks ambition. The risk is that the public market is being asked to price ambition as if the markets it points toward already exist.