Every orbital data center pitch deck leads with the physics. The economics is where most of them stall.
Dr. Oguz Karasu is a Postdoctoral Research Fellow at the University of Oxford’s Saïd Business School, where he researches the economic impact of the space industry and co-convenes the Smart Space Class for MBA students. He completed his PhD in Economics in 2023 with a dissertation analyzing the economic impact of the U.S. space industry through quantitative and qualitative methods. His research spans space-based solar power constellations, Earth observation business models, launch industry economics, and the economic viability of orbital data centers. He also serves as a Strategy and Investment Consultant for space companies, providing economic impact analysis and helping startups refine their strategic plans for investment rounds. He founded the Space and Economy Alliance, a UK-based NGO that facilitates workshops connecting space companies with adjacent sectors.
Karasu’s session at SmallSat Europe arrives at a moment when the orbital data center concept has shifted from conference speculation to active competition. SpaceX filed an FCC application in January 2026 for up to one million orbital data center satellites. Blue Origin announced its own orbital data center program in late 2025. Starcloud launched the first NVIDIA H100 GPU into orbit in November 2025 and trained the first large language model in space a month later. SatNews examined the technical and commercial architecture emerging around these systems, and separately assessed the thermal engineering constraints that make orbital cooling a defining bottleneck.
The economics remain contested. Ground-based data center power costs roughly $570 to $3,000 per kilowatt-year, while orbital solar power runs closer to $14,700 per kilowatt-year at current launch prices. The strongest near-term argument for space-based compute is not cost per watt but time to deployment: Northern Virginia grid connections now take up to seven years, and primary data center markets are effectively closed to new megawatt-scale builds. Whether that bottleneck justifies a three-times cost premium is the question Karasu’s research is built to answer.
At SmallSat Europe, Karasu delivers a Market Brief titled “Orbital Data Centers — Economics.”
The satellites, the GPUs, and the cooling systems are engineering problems. The spreadsheet is the hard part.
