SpaceX on the Stock Market, Data Centers in Orbit, and Humanity on Mars

BY EMILIO COZZI AND MATTEO MARINI

The listing sought by Elon Musk could happen in 2026 or 2027. A move to finance computing centers in orbit using space solar power. A new technological continent to dominate in order to give shape to the Martian utopia.

How much is SpaceX worth?
When it comes to Elon Musk, announcements are often gargantuan and promise new records.
It will likely be the same on the day the world’s richest man decides to list SpaceX, his space company, on the stock market. When it happens, it will probably be the wealthiest initial public offering ever.
According to sources consulted by Bloomberg and Reuters, estimates range between 25 and 30 billion dollars or more, numbers that could push the company’s total value beyond 1.5 trillion.
The operation is considered likely between the second half of 2026 and all of 2027; if the forecasts are confirmed, SpaceX would approach the market value established by Saudi Aramco during what remains the record listing: the oil giant raised 29 billion dollars in 2019.
In Musk’s mind, the operation is about money only in the initial phase; the long-term goal will be the conquest of a new technological continent with the Martian utopia on the horizon.

The rising business of SpaceX

2024 is the year when, according to analysis, the revenues from the Starlink constellation surpassed SpaceX’s other businesses, particularly the space transport service provided to NASA, the Department of Defense, and other American entities, as well as private companies and foreign countries, including Italy, which expects the launch of the third second-generation Cosmo-SkyMed satellite by the end of the year. Revenues, rising, have far exceeded 10 billion dollars. The range varies depending on estimates because, as mentioned, SpaceX’s accounts are not yet public.

When the company lands on the New York Stock Exchange and becomes listed, the books will have to be transparent. And the stock price, as has already happened for years with Tesla, will be at the mercy of the market. Meanwhile, no one doubts that Musk’s ventures are thriving. Indeed, their growth seems unstoppable: as announced by SpaceX itself at the beginning of November, Starlink has over eight million subscribers in 150 countries. Meanwhile, the company continues the evolution of the direct-to-device service, allowing internet connectivity with portable devices without the need for an antenna.

SpaceX has also carried out periodic share buybacks to provide liquidity to employees and investors, according to Musk’s summary on X. The purchase value projected the company’s virtual value at 800 billion dollars. Although impressive, Musk called the number inaccurate, later specifying that valuation increases depend on progress with Starship and Starlink and the acquisition of the global direct-to-cell spectrum, which greatly expands the potential market. “And,” he added, “on another thing that may be by far the most significant.”

Data centers where there is energy

Musk refers to data centers in orbit, a type of satellite infrastructure that, instead of providing communication, connection, Earth observation, or broadcast services, hosts computing centers.
Of course, transferring computing power into space is not an exclusive idea. Ongoing projects are diverse and scattered worldwide: in Italy, for example, D-Orbit’s acquisition of Planetek last April aimed precisely at entering a market still largely unexplored.

The problem of data centers has been well known for years, but it is since ChatGpt debuted online on November 30, 2022, that a potential crisis has begun to emerge: an energy crisis.
In 2024, to handle billions of search, text, calculation, and even more image and video requests, data centers around the world consumed 415 terawatt-hours, about 1.5 percent of the global electricity consumption that year. According to an analysis by the International Energy Agency, demand has grown by 12 percent per year since 2020.
Moreover, it is not only about energy consumption to support growing computing power; terrestrial data centers are strictly thirsty, as cooling requires huge amounts of water.

For this reason, just as the collection and transfer of solar energy from space so-called space-based solar power could represent the future of energy production overall, moving computing centers and supercomputers into orbit could, if not eliminate, at least relieve the burden of an energy-hungry humanity. Certainly, not indefinitely, as Starlink’s expansion suggests.
Data centers can also function simply as archives and be placed further from Earth American company Lonestar Data Holdings is planning one on the Moon but computing centers for systematic and strategic use require low latency, meaning closer orbits to the planet, where traffic is increasing at a worrying pace.

A new market to dominate

Musk has clearly stated that Starlink satellites, thanks to laser connectivity, which allows very high bitrate communication both inter-satellite and with Earth, will form the foundation for this. Advanced space data center projects worldwide are still counted on two hands: in Italy, Thales Alenia Space produced a feasibility study for Ascend for the European Commission; Texas-based Axiom is developing a private orbital space station; American startup Starcloud works with Nvidia; Google is considering Suncatcher, a name evoking one of the crucial issues, having the Sun always within panel reach; another U.S. company, Aetherflux, aims to use space-based solar power for AI computing centers in orbit. China is also part of this list, having already launched 12 satellites of the Three-body Computing Constellation, a name deliberately taken from a best-selling Chinese sci-fi novel.

Technical challenges abound as well because space data centers will be massive and must be built directly in space, requiring very heavy launches. A perspective that, at this moment, Musk can handle better than anyone else in the world. The 30 billion to raise on the market will finance what promises to become yet another sector dominated by the multi-billionaire obsessed with Mars.

As always, he shared his thoughts on X: “Satellites with localized AI processing, where only results are retransmitted at low latency from a sun-synchronous orbit, will be the most cost-effective way to generate AI data streams in less than three years. And by far the fastest way to scale within four years, because easily accessible electric power sources are already scarce on Earth. One million satellites per year with 100 kilowatts each produces 100 gigawatts of additional AI per year without operational or maintenance costs, connecting via high-bandwidth lasers to the Starlink constellation.
The next level involves building satellite factories on the Moon and using a mass driver to accelerate AI satellites to lunar escape velocity without rockets. This allows scaling to more than 100 terawatts per year of AI and enables significant progress toward a Kardashev II civilization,” one that, according to Russian astronomer Nikolaj Kardashev’s classification, can capture all the energy of its star.

Conquering Mars with AI

As Eric Berger noted on Arstechnica, this is part of a long-term strategy, which has evolved over time.
Until now, Musk has refused to take his creation, SpaceX, into the public stock market. The decision to proceed differently stems from fundraising possibilities to accelerate what the 54-year-old prophet of long-termism sees as an urgent need, which he aims to achieve during his lifetime, especially as new conflicts, including nuclear, loom: saving humanity by taking it to Mars.
To achieve this goal, Musk considers supercomputing and consequently AI power essential.

It is worth remembering that the third level of the Kardashev scale evoked by Musk is a civilization capable of harnessing all the energy of its galaxy. To date, we have not even reached the first type, a civilization able to exploit all the energy available on its home planet.
The problem is precisely energy. And Elon Musk has, so far, used it better than anyone else.