AIBRIEF
Space based solar, satellites that collect sunlight in orbit and beam it to ground stations, is one candidate answer to AI's electricity crunch, with per kilowatt cost tied to Starship's launch cadence.
The people building the AI stack now say power is the binding constraint. At a Sept. 29 Washington, D.C. event, Elon Musk said power generation could become one of the biggest constraints on America's ability to compete in artificial intelligence, and NVIDIA's Jensen Huang, on the same panel, called energy the foundation of the AI computing stack (Yahoo Finance / GuruFocus). Musk put a number on the gap: China's electricity production is roughly three times that of the United States.
One proposed answer is space-based solar power, or SBSP: satellites in orbit collect sunlight, convert it, and beam the energy to a ground receiving station called a rectenna, which feeds it into the grid. It is one option alongside new nuclear, geothermal, storage, and transmission buildout. SBSP's economics ride on a different curve: per-kilowatt cost of orbital hardware is set by launch cost times launch cadence.
That curve just moved. On Sept. 28, SpaceX flew Starship on its first orbital flight to deliver meaningful payload, deploying 26 Starlink V3 satellites (Spaceflight Now). Musk has said he expects Starship to reach a weekly or twice-weekly launch cadence next year. DOE and DOD have begun new and large SBSP investments (Stocktitan).
Musk's cadence target is forward-looking, and SpaceX has historically slipped similar goals. SBSP also depends on rectenna siting, wireless transmission efficiency, and spectrum allocation. The Sept. 28 flight made the launch-cost gate visible. The other gates are still open.