Scientists Warn SpaceX's Orbital Data Center Plans Would Devastate Astronomy

On January 30, 2026, SpaceX filed an application with the Federal Communications Commission to launch up to one million satellites into low Earth orbit—not for internet service, but to create a distributed network of solar-powered data centers designed to run artificial intelligence workloads in space [1]. The filing, designated SAT-LOA-20260108-00016, arrived at the FCC and was processed in days rather than the typical weeks or months [2]. Within weeks, more than 1,000 public comments flooded in, the majority opposed [2].

For astronomers who have spent years negotiating with satellite operators over the brightness of existing constellations, the proposal represents a qualitative escalation. "This just blows right past that," said Aaron Boley, co-director of the Outer Space Institute and a physics professor at the University of British Columbia. "And by almost all metrics that we can think of, this is just a bad idea" [2].

The Proposal: AI Computing in Orbit

The orbital data center concept emerged from the formal merger of SpaceX and xAI, Elon Musk's artificial intelligence company [3]. SpaceX's filing proposes satellites at altitudes between 500 and 2,000 kilometers, in both 30-degree and sun-synchronous orbits [1]. The sun-synchronous orbits would keep satellites in sunlight more than 99 percent of the time, maximizing solar power generation for continuous computing [1].

SpaceX's core argument is economic: as terrestrial data center costs and power demands rise, and launch costs fall, space-based computing will become cheaper. The company claims that once Starship achieves full reusability, launching one million tonnes of satellites per year—each tonne generating 100 kilowatts of computing power—would add 100 gigawatts of AI compute capacity annually [1]. "By directly harnessing near-constant solar power with little operating or maintenance cost, these satellites will achieve transformative cost and energy efficiency," SpaceX wrote in the filing [1].

The company contends that "within a few years the lowest cost to generate AI compute will be in space" [1]. The satellites would communicate through intersatellite optical links and relay data to the ground via the existing Starlink network [1].

Not everyone finds the timeline credible. The European Space Policy Institute calculated that a competitive orbital data center is "at least 20 years away" [4]. Thermal management in the vacuum of space remains a fundamental engineering challenge—"there is no fluid to dissipate heat," as one researcher noted, meaning massive radiator infrastructure would be required [4]. Orbital hardware typically lasts only five years due to radiation damage, raising questions about maintenance and replacement at scale [4].

The Scale of Impact on Astronomy

To understand what one million additional satellites would mean, consider the current baseline. As of March 2026, SpaceX operates approximately 10,087 active Starlink satellites, comprising 65 percent of all active satellites in orbit [5]. The proposed data center constellation would multiply the number of SpaceX spacecraft by roughly a hundredfold.

The astronomical consequences have been studied with increasing urgency. In December 2025, Alejandro Borlaff at NASA's Ames Research Center and colleagues published findings in Nature showing that adding 500,000 satellites to Earth orbit would mean "almost every single telescope image taken from Earth or space will be contaminated by satellites" [6]. Their simulations of four space telescopes found that 96 percent of all images from NASA's SPHEREx, the European Space Agency's ARRAKIHS, and China's Xuntian telescope would be affected, while a third of Hubble Space Telescope images would be tainted [6]. A single image could contain as many as 92 satellite streaks [6].

One million data center satellites—double the number Borlaff modeled—would amplify these effects substantially. John Barentine, an astronomer and dark sky consultant, calculated that the constellation "will be in high-inclination orbits and will be fully illuminated by sunlight even as seen from the ground at midnight," producing "possibly tens of thousands of moving objects as bright as stars that are visible in the night sky at any given moment, even to the naked eye" [2]. He called this "a challenge unlike any we have encountered thus far" [2].

Critically, orbital data center satellites would likely be brighter than existing Starlink spacecraft. They would require not only large reflective solar panels but also large radiators to expel waste heat—structures similar to those on the International Space Station [2].

Radio Astronomy Under Threat

The damage extends beyond optical interference. Scientists have cataloged more than 112,000 unintended radio emissions from existing Starlink satellites [7]. Over a four-month observation period, researchers detected radio frequency interference in up to 30 percent of images from a prototype radio telescope station, often in frequencies that are supposed to be protected for scientific use [7]. The average interference level was approximately 100,000 times higher than limits considered acceptable for detecting signals from the early universe [7].

No regulations currently extend radio-quiet zone protections against downlinks from space [8]. Since all transmitters above the horizon are detectable by radio telescopes, every additional satellite degrades the radio environment. Low-frequency transmissions from unshielded electronics and motors aboard satellites are detectable from the ground even when they are not intentionally broadcasting [8].

The Royal Astronomical Society has predicted that satellite trails from an expanded constellation could obscure up to 10 percent of data captured by the European Southern Observatory's Very Large Telescope [7].

Billions in Observatory Infrastructure at Risk

The facilities threatened by satellite proliferation represent decades of investment. The Vera C. Rubin Observatory in Chile, which began operations in 2025, cost more than $800 million in U.S. funding alone [9]. It captures approximately 1,000 wide-field images per night across a planned 10-year survey. Conservative studies of constellations comprising 26,000 to 48,000 satellites—a fraction of the proposed million—predict that 20 percent of midnight images and 30 to 80 percent of twilight exposures would show satellite trails [10].

Source: Nature, NPR, Rubin Observatory, ESO, GMTO, SKA Observatory

Data as of Mar 23, 2026CSV

The United States invests more than $2 billion annually in astronomy and astrophysics [11]. Major ground-based facilities under construction or recently completed include the Giant Magellan Telescope and the Extremely Large Telescope, each costing in the billions. Nearly 70 percent of the world's astronomical infrastructure is concentrated in Chile, where dark-sky conditions are now directly threatened [11].

If ground-based astronomy were severely compromised, replacing lost capabilities with space-based telescopes would be extraordinarily expensive and slow. The James Webb Space Telescope cost $9.7 billion and took over two decades from conception to deployment [12]. JWST observes primarily in infrared wavelengths—it cannot replicate the wide-field optical survey work of facilities like Rubin. Building space-based equivalents for the full range of ground-based astronomy would cost tens of billions of dollars and take decades, with no guarantee that satellite contamination wouldn't affect those space telescopes too, as Borlaff's study demonstrated [6].

SpaceX's Mitigation Track Record

SpaceX has previously attempted to reduce the brightness of its Starlink satellites. The DarkSat prototype, featuring black antireflective coating, was roughly half as bright as standard Starlinks—an improvement, but researcher Jonathan McDowell described it as "essentially a dead end" [13]. The subsequent VisorSat design, using a black sunshade to block reflected sunlight, achieved a brightness reduction of approximately one magnitude [8].

The transition from Generation 1 to Generation 2 Starlinks made satellites fainter despite a larger physical cross-section [8]. But astronomer Meredith Rawls warned that these gains are insufficient as constellations grow, and the problem worsens as additional companies deploy their own mega-constellations [13].

The SATCON1 workshop, a major collaborative assessment of satellite interference, produced 10 recommendations. Only one addressed satellite darkening. The others emphasized lower orbital deployment, orientation controls, sharing orbital data with observatories, and developing software to remove satellite trails from images [13]. None of these approaches scales to a million-satellite constellation.

For the proposed data center satellites, mitigation faces additional obstacles. The large solar arrays and thermal radiators required for computing and cooling are inherently reflective surfaces that cannot easily be darkened without compromising their function [2]. SpaceX's FCC filing mentioned minimizing "atmospheric impacts" but offered no specifics on optical or radio frequency mitigation [2].

Regulatory Gaps

The regulatory landscape offers limited protection for astronomy. The FCC licenses satellite communications and has adopted a five-year deorbiting rule for defunct satellites [14]. In recent approvals for other constellations, the FCC has begun requiring operators to work with the National Science Foundation to minimize impacts on astronomy—for example, AST SpaceMobile's license includes conditions requiring brightness targets, annual reporting, and preference for lower orbits [14].

But these conditions are ad hoc, not systematic. No U.S. law grants any agency authority to deny a satellite license based on astronomical impact alone. The FCC's jurisdiction covers radio spectrum management and orbital debris, not optical brightness or scientific observatory protection [14].

Internationally, the International Astronomical Union established the Centre for the Protection of the Dark and Quiet Sky from Satellite Constellation Interference, which has convened at least 15 satellite operators to discuss best practices [8]. The UN Committee for the Peaceful Use of Outer Space has taken up the issue, but no binding international treaty addresses the protection of astronomical observation from satellite interference [14].

Amazon, SpaceX's competitor in the satellite internet market through its Project Kuiper, formally requested that the FCC reject SpaceX's application [2]. The competitive dynamics add complexity: Amazon's opposition may be commercially motivated, but the underlying questions about orbital capacity and environmental impact apply to all operators.

Monopoly and Sovereignty Concerns

Beyond astronomy, the orbital data center concept raises questions about market concentration. If solar energy in orbit becomes effectively free, whoever controls space-based computing infrastructure could dominate AI services globally [4]. With the SpaceX-xAI merger consolidating launch capability, communications networks, and AI development under one corporate umbrella, experts have flagged the risk of monopolistic control.

"When the same small group controls multiple choke points, like launch and comms, plus AI, plus robotics, plus consumer platforms, you get something closer to an oligarchy that's hard to regulate," one expert warned [4]. European analysts noted that the continent lacks any comparable space data center plans and risks the kind of dependency that already characterizes its reliance on American cloud providers [4].

Environmental Costs

The environmental implications extend beyond the night sky. Barentine calculated that with a million satellites having limited operational lifespans, a decommissioned spacecraft would reenter Earth's atmosphere approximately every three minutes [2]. Each reentry releases aluminum oxides and lithium into the upper atmosphere—compounds that may impede the recovery of Earth's ozone layer [2].

Rocket launches required to deploy and replace the constellation would deposit black carbon into the stratosphere at unprecedented rates. These emissions are roughly 500 times more effective at warming the atmosphere than the same particles released at the surface [7].

What Comes Next

The FCC's public comment period for SpaceX's application closed on March 6, 2026, with the majority of submissions opposing the project [2]. The commission faces a decision without clear precedent: no regulatory framework was designed for a constellation of this magnitude.

Astronomers are not calling for a halt to commercial space activity. They are asking for regulatory structures that account for the cumulative impact of satellite deployments on scientific infrastructure worth billions of dollars. The question is whether those structures can be built before the satellites are launched.

"It really feels like it's undermining what we have achieved in the last few years," Barentine said of the mitigation progress made with existing constellations [2]. For an endeavor that promises to bring AI computing to orbit, the cost may be measured not just in dollars but in humanity's ability to study the universe from its home planet.