Key Points

SpaceX reports ongoing losses, pinning its strategy on expected future expansion.
The firm aims to quickly ramp up manufacturing and launch of AI processing satellites.
Placing these units in sun-synchronous paths may cause intense light pollution.
10 stocks we prefer over Space Exploration Technologies

Where to place $1,000 today? Our research group recently shared the 10 top stocks to acquire now, when you join Stock Advisor. View the stocks.

Within five trading days after its public debut on June 12, Space Exploration Technologies (NASDAQ: SPCX) briefly topped the valuations of Microsoft and Amazon, yet it has since fallen 18 percent from its peak.

Following a gain in 2024, the Elon Musk-led enterprise posted an operating loss of $2.59 billion in 2025 along with a $1.94 billion operating shortfall for the quarter ended March 31, 2026. Certain backers accept these shortfalls due to SpaceX's extended path to growth via its Starlink satellite array, AI initiatives at xAI, the X platform, and especially its aim to place potentially millions of AI processing satellites into orbital clusters.

Here is why SpaceX holds strong promise, plus a notable obstacle that could hinder its orbital data center plans.

Image source: Getty Images.

A striking obstacle

SpaceX finished 2025 with 9,600 Starlink broadband and mobile satellites circling in low-Earth orbit. That figure represents roughly 75 percent of active steerable satellites as of March 31, 2026. As Starlink has grown, a key hurdle has been curbing light pollution from sunlight bouncing off satellite bodies and solar panels. The company has tackled this through low-reflectivity materials, reflective surfaces to cut glare, and by angling solar panels during key twilight periods to limit bright flashes. Because they sit in low Earth orbit, Starlink units also enter Earth's shadow, hiding them from view at night.

Yet SpaceX's regulatory filing specifically notes plans to use sun-synchronous orbit for its AI processing satellites. These units would circle near the poles to maintain steady solar input. This setup differs sharply from Starlink's low-Earth paths. With constant sunlight feeding power-hungry data centers, the higher AI satellites would stay visible at night and prove hard to dim from a dark-sky standpoint.

Large-scale placement of AI satellites in sun-synchronous paths would spoil nighttime viewing and pose the biggest threat to astronomy yet seen. Given pushback against offshore wind projects from locals and firms, a sky lit by satellites would likely spark strong opposition.

Mitigating AI satellite light issues

Two main paths exist to address SpaceX's sky-impact problem. One involves developing advanced coatings for sun-synchronous satellites that lower visibility without cutting solar output. Starlink has already advanced this area with low-reflectivity parts, dark foam, and dark coatings. Still, those steps pair with tilting panels away from Earth and using Earth's shadow, options unavailable to sun-synchronous units.

The second path relies on energy storage, possibly from Tesla, to skip sun-synchronous placement. Storage would add weight and launch cost while possibly shortening satellite life from battery wear. Yet it would ease the need for peak solar generation. Data center groups could also align orbits to capture solar power during peak AI compute loads, much like grid peaker plants handle demand spikes at morning or evening hours.

Ready for deployment

Orbital data centers stand as the most compelling part of SpaceX's case for backers who see AI growth still early, yet they confront a growing energy crunch. Free solar power and no water cooling would cut costs and environmental load. The idea still meets hurdles around light pollution and upkeep.

Starlink satellites carry thrusters, avoidance systems, remote updates, and retirement commands to de-orbit units. SpaceX may apply similar methods to AI satellites.

Since AI units skip Starlink's antennas, SpaceX expects simpler builds focused on solar arrays plus liquid radiators from a central compute section. The initial version spans 70 meters wide and 20 meters tall, with 150 kW peak output and 120 kW average load. That equals roughly one-fourth a football field's area, sizable for production yet modest in orbit.

SpaceX targets an annual rate of 1 GW of orbital AI compute by late 2027, then multiplies that tenfold each year to reach 10 GW by late 2028, 100 GW by late 2029, and 1 TW yearly by late 2030. For scale, U.S. yearly power use totals about 4 TW.

Monitor SpaceX closely

SpaceX is preparing to mass-produce AI compute satellites. The plan still carries major risks if it depends on sun-synchronous paths.

Investors can best handle this visible concern by tracking SpaceX and watching how the firm tackles these issues plus any updates from earnings discussions.

From Starlink, SpaceX's launch role, xAI, and X alone, the company could deliver solid near-term gains, reach profitability, and perhaps support a $2 trillion valuation. Yet for new buyers at current levels, the growth case needs successful orbital data centers that spare the night sky.

SpaceX's goal is to create systems for multiplanetary life, explore the universe, and spread awareness to the stars. That goal loses meaning if people on Earth lose sight of the stars themselves.

Should you acquire shares in Space Exploration Technologies today?

Before purchasing shares in Space Exploration Technologies, review this:

The Motley Fool Stock Advisor team recently named what they view as the 10 best stocks for buyers to own now, and Space Exploration Technologies did not make the list. The 10 selected stocks could deliver strong gains ahead.

Recall when Netflix appeared on the list December 17, 2004. A $1,000 investment then would now equal $417,305. Or when Nvidia joined April 15, 2005. The same $1,000 would now total $1,293,148.

Note that Stock Advisor's overall average return stands at 936 percent, far ahead of the S&P 500's 207 percent. Do not overlook the latest top 10 list, available via Stock Advisor, and join an investor community built by individuals for individuals.

View the 10 stocks.

Stock Advisor returns as of June 19, 2026.

Daniel Foelber holds no positions in the mentioned stocks. The Motley Fool owns stakes in and recommends Amazon, Microsoft, and Tesla. The Motley Fool maintains a disclosure policy.

The views and opinions expressed here belong to the author and may not reflect those of Nasdaq, Inc.