Two weeks ago, a company filed paperwork with the FCC requesting permission to launch up to one million satellites into low Earth orbit. Not communication satellites. Not GPS. Data center satellites. Solar-powered, laser-linked orbital compute nodes, designed to run AI inference workloads from space. The filing included the phrase "first step towards becoming a Kardashev II-level civilization." That's a reference to harnessing the full power of a star. For context, we're currently a Kardashev 0.7. The filing didn't include a deployment timeline, a cost estimate, or detailed technical specifications for the satellites themselves. I've been thinking about this filing, and what fascinates me isn't whether it's feasible. It's what it reveals about how deeply the centralization assumption has embedded itself in how we think about computing infrastructure. The Setup The logic goes like this: AI workloads require enormous compute. Compute requires enormous power.…