Major technology companies are exploring orbital data centers to power artificial intelligence, with Elon Musk claiming costs could drop below Earth-based systems within two to three years. However, experts at leading universities say significant technical and financial hurdles remain before space-based computing becomes practical .

What's Driving the Push for Data Centers in Space?

The motivation is straightforward: artificial intelligence is consuming electricity at an unprecedented rate. Global data-center power consumption is expected to roughly double to nearly 1,000 terawatt-hours by the end of the decade, according to the International Energy Agency . On Earth, companies are running out of places to build new power plants to feed this demand.

Space offers a compelling alternative. Unlike terrestrial data centers, satellites orbiting Earth experience continuous sunlight, providing free, renewable energy. Musk laid out this vision in March, announcing that SpaceX, which recently merged with his artificial intelligence company xAI, would deploy data-crunching satellites into orbit. "You're power constrained on Earth," Musk explained. "Space has the advantage that it's always sunny" .

Space

"We're very quickly running up on constraints on where you can build new energy projects terrestrially. Within six months, they'll just be leaving chips in warehouses because they don't have power for turning them on," said Philip Johnston, CEO and co-founder of Starcloud, a company seeking to build orbital data centers.

Philip Johnston, CEO and co-founder of Starcloud

Starcloud already launched its first spacecraft last fall with an Nvidia H100 chip on board and demonstrated the ability to run a version of Google's Gemini artificial intelligence from space. The company plans to launch a second spacecraft in October with 100 times the power generation of the first, though it will still produce only around 8 kilowatts of power .

How Realistic Is Musk's Two-to-Three-Year Timeline?

Skepticism abounds among experts. Brandon Lucia, a professor of electrical and computer engineering at Carnegie Mellon University who specializes in putting computers on satellites, called Musk's timeline "an optimistic interpretation" . The napkin math looks appealing, but the engineering challenges are formidable.

The power requirements alone are staggering. The International Space Station, the largest power-producing facility in space, has solar panels roughly half the size of a football field and produces around 100 kilowatts of average power, according to Olivier de Weck, a professor of astronautics at the Massachusetts Institute of Technology . To replicate a 100-megawatt data center in space would require a facility 500 to 1,000 times larger, depending on the orbit.

"Is that feasible? Yeah, I think it's feasible, but not next year and certainly not in three years," said Olivier de Weck, professor of astronautics at MIT.

Olivier de Weck, Professor of Astronautics at MIT

Google is pursuing the idea through a project called Suncatcher, envisioning an 81-satellite cluster built in partnership with Planet, a satellite-imagery company. Two prototype satellites will launch in early 2027 . Google believes launch costs must drop by at least a factor of five, from around $1,000 per kilogram to $200 per kilogram, before orbital data centers make financial sense.

What Are the Major Technical Obstacles?

Beyond raw power generation, orbital data centers face multiple interconnected challenges:

• Heat Dissipation: While space is cold, it's also a vacuum, meaning heat cannot escape naturally. Satellites must use radiators to move liquids to giant panels where heat can be dissipated, adding significant mass and complexity to spacecraft.
• Cooling Infrastructure: Combining massive radiators with massive solar arrays creates extremely large satellites or very large satellite constellations, according to Rebekah Reed, a former NASA official now at Harvard University's Belfer Center for Science and International Affairs.
• Data Transmission Latency: Smaller satellites flying in preset formations called constellations would need to send huge amounts of data between units using lasers. Even at the speed of light, this introduces delays long enough to slow down computing.
• Maintenance and Upgrades: Terrestrial data centers require constant maintenance, with vendors and workers on-site daily installing new servers and fixing equipment. Replicating this in orbit presents unprecedented logistical challenges.

Musk has proposed launching upward of a million satellites and placing them in orbit around Earth's poles. He recently unveiled the first generation "AI Sat Mini" spacecraft with solar arrays spanning roughly 180 meters, or about 600 feet .

How Can Companies Make Space Data Centers Economically Viable?

The path forward depends heavily on reducing launch costs and solving the engineering puzzles. Musk believes his Starship rocket, still in development, can dramatically lower launch expenses. Starcloud's Johnston told investors: "If you don't think Starship's going to work, don't invest in us, that's totally fine" .

Musk

Google's Project Suncatcher proposes flying satellite groupings in extremely tight clusters to reduce latency between units. This approach allows heat and power problems to be distributed across multiple spacecraft rather than concentrated in a single massive satellite .

"Orbital data centers are an idea whose time has come. When exactly it will be more cost efficient than terrestrial ones is debatable but now is the time to be working on this," wrote Will Marshall, Planet's CEO, to NPR.

Will Marshall, CEO of Planet

The companies pursuing this vision acknowledge the timeline is longer than Musk suggests. Bloomberg reported this week that SpaceX had filed documents confidentially with the Securities and Exchange Commission with the goal of listing an initial public offering this summer, with orbital data centers as part of its growth narrative .

For now, the race is on. Starcloud has demonstrated proof of concept. Google is building prototypes. And SpaceX is positioning itself as the launch provider that could make it all possible. But experts agree that transforming orbital data centers from an intriguing idea into a practical, cost-effective alternative to Earth-based computing will take years, not months.