AI Infrastructure Spending Boom Threatens to Overheat US Economy, Complicating Fed's Inflation Fight

The United States faces a paradox: its economy is too strong, not too weak, and the artificial intelligence buildout is compounding inflationary pressure. AI infrastructure spending is projected to reach roughly $1 trillion in 2026 — exceeding annual US military expenditure — and $10.3 trillion through 2032, according to Columbia University economist Stijn Van Nieuwerburgh. That equals 3.6% of GDP annually, surpassing prior US investment booms in canals, railroads and highways. Demand for memory chips, building materials, electricity and skilled labor is outpacing supply, driving up costs economy-wide. Chicago Fed President Austan Goolsbee warned on September 21 that data center construction may be "spilling out of its own lane," signaling demand overheating that could force further rate hikes. The spending lands on an already hot economy: unemployment at 4.1%, August retail sales up 1.2%, and an AI-driven stock rally fueling consumption among top earners. Even if the Iran war ended and tariffs were lifted, inflation — above the Fed's 2% target for over five years — would likely persist.