Elon Musk's healthiest company, SpaceX, has spent years being called a space exploration pioneer. But after the first quarterly earnings report as a public company, it's clear the name is a vestige of identity. What we have here, by revenue, is primarily a telecom company and a company that rents compute. The space sector didn't break a billion dollars this quarter, contributing only a touch over 10 percent of total revenue. SpaceX remains its own biggest customer, because there just aren't enough outside customers who want its rockets.

One might think that if the rockets were the primary focus of the company, SpaceX wouldn't be in danger of blasting a new crater into the Moon with its space trash. But the numbers paint a very different picture. The lion's share of spending and hype is around what's graciously being referred to as AI, and the revenue streams now resemble a neocloud provider more than a traditional aerospace company.

Starlink is the only profitable business

The telecom segment, which SpaceX calls connectivity, is Starlink. The satellite internet service generated $4.2 billion in revenue and was the only part of SpaceX that did not have a loss from operations. Gwynne Shotwell, speaking on the earnings call, outlined plans for a phone service meant to compete with the likes of AT&T, Verizon, and T-Mobile. But even with that ambition, the scale of AI-related investment dwarfs connectivity.

Analyst Alexander Potter expects spending on the neocloud business, which leases data center capacity to AI companies, to climb to $65 billion next year. That's $17 billion more than previously estimated, according to Bloomberg. The AI story isn't just about future plans; it's already moving money through the company today.

Data centers are the real moneymaker

It was hard to miss how important leasing data center space was to SpaceX in the quarter. Not only did the AI business make more money than SpaceX's rockets, it also drove more spending. The company poured $15.8 billion into AI alone in the second quarter. Meanwhile, spending on space and connectivity was a touch over a billion each. That business makes SpaceX a direct competitor with neocloud companies like CoreWeave and Nebius.

Except that wasn't originally the plan. Musk built SpaceX's Colossus 1 data center in Memphis for Grok, his AI model that has had a rocky history. But xAI had trouble running the complex, encountering latency issues that made it hard to train in-house models. The center employed a mix of newer and older chips that created bottlenecks. So xAI decided to rent the compute out instead. On the earnings call, Musk said that only 10 percent of the compute SpaceX builds will go to Grok.

Now SpaceX has deals with Google, Anthropic, Reflection AI, and Cursor, an AI company Musk eventually chose to acquire. On the second quarter earnings call, SpaceX chief financial officer Bret Johnsen said those deals put the company on a trajectory, including contribution from Cursor, to reach $100 billion of ARR, or annualized revenue run rate. That financial measurement estimates how much money a company will receive in a year based on a shorter period. Musk was even more bullish, saying that the $100 billion ARR in December is not a question mark and that the actual ARR might be higher.

Revenue isn't profit

That's very nice. But revenue isn't profit, and building data centers is expensive. There are some problems with being in the bare-metal business of renting out compute. Those include inevitable obsolescence, the vagaries of construction, and the fact that compute is basically a commodity. Companies compete on cost. The more data centers get built, the more compute is available. With more compute available, companies can't charge as much for their chips.

Musk has claimed that he took SpaceX public because he wanted to build data centers in space. That's a strange rationale from someone who had trouble running a data center on the ground. One would think the logical step is to get better at terrestrial operations before trying something an order of magnitude harder, something that had never been done before. But that is not how Musk thinks, apparently.

Orbital data centers and a lunar mass accelerator

SpaceX has even proposed an orbital data center consisting of as many as 1 million satellites to the Federal Communications Commission. The application is light on technical details. There is no clear information about satellite size or deployment schedule. That suggests the filing may be more for public relations than engineering reality. Musk has also released a few drawings of the proposed satellites. In this vision, a Musk-owned chip producer called Terafab will manufacture one terawatt of chips every year. A billion Optimus robots will do the work, at least once Musk figures out hands.

This all sounds expensive. Musk claims the end goal is to build a mass accelerator on the Moon. Such visions are straight out of mid-20th-century science fiction. You can argue about the feasibility of space data centers if you want. Remind me, did the Hyperloop ever get built? It's all vaporware till it ships, babe.

What's left without the sci-fi big talk?

If we take away the pseudoscientific big talk, we are left with a company that launches rockets mostly for itself, has a fairly successful satellite internet business, and is basically running the risky and capital-intensive bare-metal business we've seen from a bunch of the neoclouds. That's not quite as dreamy as getting to Mars in six years.

What it does do, however, is provide Tesla with an important customer. SpaceX buys $295 million in Tesla Megapack battery storage. SpaceX has also been buying a bunch of Cybertrucks, Musk's historically poor-performing automobile. As of this writing, Tesla stock is down 25 percent since January of this year.

There's also the political angle. Musk can perhaps charge a premium on his compute because of his political connections. He is reportedly planning to splash out $100 million on the midterms, and that likely isn't for nothing. Political connections could make it easier for Musk to build terrestrial data centers, though those seem to be wildly unpopular across the political spectrum. The game plan might even be predatory: that's a nice AI business you have there. Would be a real shame if something happened to it. Perhaps you should buy my compute, so nothing happens.

Lockups, short-sellers, and Nasdaq rule changes

But maybe the long-term reasoning is too generous. After all, SpaceX's insider lockups start expiring on August 6th, which is tomorrow. If those insiders sell, as short-sellers expect, the already slumping stock will be even worse off. So reminding everyone of the mass accelerator on the Moon might be a way to rally the troops. Good thing Nasdaq changed its rules for SpaceX, so those losses will hit everyone with an index fund.

The reality of SpaceX today is a far cry from the romantic vision of a multiplanetary future. It's a company powered by Starlink subscriptions, AI compute rental, and a government-adjacent ability to move fast and break things. The rockets are real, but they're no longer the main event. The main event is data centers and connectivity, with the occasional exploding rocket as a front for the moneymakers. That may be enough for investors, but it's not the same dream that once captured the public's imagination.

What remains is a heavily capitalized machine that is trying to bridge two very different industries: space infrastructure and AI cloud services. The former has high margins for specialized customers, but the latter requires constant reinvestment to stay relevant. As compute becomes commoditized, the AI rental market will only get more competitive. SpaceX may have the brand and the balance sheet to survive, but the question of what it actually is, or wants to be, becomes more uncertain with each quarterly report.

Source:The Verge News