The AI buildout is the largest raw mobilization of resources in human history. These are the questions I'm trying to answer about it — slowly, with receipts, as an anthropological problem rather than a news beat.
Four questions. Nothing published under them yet — this is where the long-form work will land as it's built.
The hyperscalers and frontier labs are spending on the order of three-quarters of a trillion dollars a year. What does that actually buy — chips, land, power, cooling, networking, labor — and how much of it is the industry paying itself?
Behind the buildout sits demand. Who actually pays for AI — consumers, enterprises, governments — how much is real recognized revenue versus a claimed run-rate, and what are the committed capacity contracts genuinely worth?
Every datacenter on earth owned by, operated by, or dedicated to the top players — and how much power each one draws. A census of the physical stock the money builds.
The governance and management layer — how the firms behind the buildout are actually structured, controlled, and run internally.
