Orbital, a startup founded in 2026, has secured funding to build artificial intelligence data centers in orbit. The company is part of a growing wave of ventures trying to solve the AI infrastructure problem from a direction that would have seemed far-fetched five years ago.
The Logic Behind Space-Based Compute
AI data centers are among the most power-hungry facilities humans build. Large installations can consume as much electricity as a small city, and cooling alone accounts for 30-40% of that energy use. In orbit, both problems get simpler: solar panels run in near-continuous sunlight with no atmosphere to scatter it, and excess heat radiates directly into the cold vacuum of space.
The pitch to investors is that as AI compute demand keeps growing, Earth-based infrastructure will hit real physical limits - grid capacity, cooling water availability, land near fiber networks. Space sidesteps all three at once.
The counterarguments are real. Launch costs, even with reusable rockets, remain high per kilogram. On-orbit maintenance is practically impossible with current technology. Any data center in space adds latency - the signal travel time between orbit and ground - which matters for real-time applications.
Orbital is at the groundwork stage. This funding supports planning and preliminary development, not a facility already in orbit. Real space-based AI compute from any company pursuing this vision is likely several years out. The investor interest says less about when it will exist and more about how seriously the long-term infrastructure bottleneck is being taken.