“Peak Insanity”? What You Need to Know About SpaceX’s Orbital Data Centers

In February, Elon Musk’s SpaceX asked the Federal Communications Commission (FCC) for permission to launch 1 million solar‑powered satellites, engineered as orbital data centers to support the development of artificial intelligence (AI). 

“In the long term, space-based AI is obviously the only way to scale,” Musk said. “To harness even a millionth of our Sun’s energy would require over a million times more energy than our civilization currently uses! The only logical solution therefore is to transport these resource-intensive efforts to a location with vast power and space.”

As with much that Musk does, the proposal has sparked a range of reactions, from incredulity to admiration. Let’s look at the broader context, then boil down some analysis and try to establish the viability of a space-based data center.

An Old Idea Acquires New Urgency

The notion of using solar power in orbit has been around since the 1970s, when the US Department of Energy worked with NASA on some concepts, before deciding that the idea was prohibitively expensive.

More recently, in the context of the new space race, both Alphabet [GOOGL] and Jeff Bezos’ Blue Origin have considered the idea.

The recent surge in AI demand has fundamentally altered the economics of data infrastructure. As companies like OpenAI, Google and Meta [META] scale up large language models and generative systems, the strain on terrestrial data centers has intensified. Training and running these models requires vast amounts of electricity and generates significant heat, pushing existing power grids, cooling systems and suitable real estate to their limits. 

In that context, the previously outlandish notion of orbital data centers has suddenly become very interesting indeed. Space offers continuous solar energy, fewer physical constraints and more efficient heat dissipation through radiation.

At the same time, the feasibility of deploying infrastructure in orbit has improved markedly due to falling launch costs. The progress of SpaceX, particularly through reusable rocket technology, has reshaped the cost curve for sending payloads into space. With the development of heavy-lift vehicles such as Starship, the idea of placing large-scale computing infrastructure in orbit is no longer purely theoretical. What was once prohibitively expensive seems now to be edging into the realm of commercial viability.

Why Musk? Why Now?

The plan came on the heels of the merger of SpaceX and xAI, and as SpaceX began gearing up for a potential $1.5trn IPO.

The merger concentrated investor attention on how Elon Musk could tackle major technical and economic constraints by integrating rockets, satellites and AI into a unified ecosystem — effectively extending AI infrastructure beyond Earth. 

SpaceX filed the plan on January 30. The FCC opened the application for public comment just five days later, although The Register wondered whether the alacrity of their response was “just so [the FCC] can collect enough public comments to shoot the ridiculous idea down before it gets beyond the Muskian brainstorm stage”.

Indeed, at the time of writing, some 1500 comments had been made, far in excess of what such filings usually garner. 

Nay-Sayers

Experts caution that significant commercial returns remain a long way off. 

The concept faces a host of formidable technical and financial obstacles, including exposure to radiation, orbital debris, heat dissipation challenges, communication latency and steep operating costs, all of which make profitability uncertain for years to come.

Perhaps the most stringent critic of the SpaceX plan has been Amazon [AMZN]. In a filing dated March 6, its Amazon Leo telecoms wing (formerly known as Project Kuiper) notes that the SpaceX proposal “lacks basic details like satellite design (still under development), orbital altitudes (anywhere between 500 km and 2,000 km), radiofrequency characteristics (deemed too complex to supply, so instead provided for only three representative satellites), or any credible plan for managing conjunctions or interference at million-satellite scale (offering only generalized assertions instead of specific assessment or demonstration).”

In short, Amazon says, the SpaceX plan is speculative, incomplete and unrealistic. The tech firm urged the FCC to “fully address the impacts and risks it poses to competition and the commission’s broader space agenda”.

Elsewhere, and even more pungently, analyst firm Gartner has described the idea as “peak insanity”.

“Companies are wasting money by pouring funds into the orbital data center ‘bubble’ because the economics do not work,” wrote analyst Bill Ray. “This is due to the prohibitive costs of launching hardware and the immense technical challenges of cooling these orbital datacenters in the vacuum that is space.” 

Another set of concerns was voiced by astronomers, who told Space.com that the satellite streaks caused by the 1 million satellites would severely impair observations, as well as lead to large amounts of space debris.

Yay-Sayers

Nevertheless, a number of voices have expressed cautious optimism around the project. 

“Compute in space isn’t sci-fi anymore,” David Ariosto, founder of space intelligence firm The Space Agency, told Reuters. “And Elon Musk has already proven himself capable across multiple domains.”

More concretely, Kathleen Curlee, a researcher at Georgetown University, noted that “SpaceX has structural advantages that few others can match. It controls the world's most active launch fleet, has demonstrated mass production of spacecraft through Starlink and has access to substantial private capital.” 

Meanwhile, SpaceX has responded to criticisms in a recent filing to the FCC. 

“SpaceX understands commenters’ concerns regarding frequent launches and satellite re-entries potentially impacting Earth’s atmosphere,” it wrote. 

To address these concerns, it said it would work with “relevant federal agencies and stakeholders to study how the Orbital Data Center system interacts with Earth’s atmosphere."

“Consistent with SpaceX’s commitment to environmental stewardship and data-driven decision making, SpaceX also plans a phased deployment approach for this system to monitor actual atmospheric effects, validate models with real-world data and implement any necessary adjustments as it scales.”

The company also accused Amazon of “fearmongering”. 

The Chairman of the FCC, Brendan Carr, has also upbraided Amazon for its critique of the proposal, saying on X that it “should focus on the fact that it will fall roughly 1,000 satellites short of meeting its upcoming deployment milestone, rather than spending their time and resources filing petitions against companies that are putting thousands of satellites in orbit”.

What Investors Need to Know

Whether they think Musk is a transcendent genius or an arrant fantasist, investors should view orbital data centers as a long-term, high-risk play rather than a near-term revenue generator. 

While the AI boom and falling launch costs make space-based computing increasingly plausible, technical challenges remain significant, and operating expenses are steep. Early-stage ventures may attract funding and strategic partnerships, but widespread adoption is likely years away. 

Exposure to companies pursuing this frontier is essentially a bet on visionary infrastructure, integrated ecosystems and regulatory breakthroughs. 

For now, investors must balance the promise of unprecedented scale with the reality of formidable execution hurdles.