The Economic Cost of AI: A Four-Part Series
AI isn’t just burning through chips and cash; it’s burning through our power supply.
This is a column about technology. See my full ethics disclosure here.
By 2030, Goldman Sachs projects that U.S. data centers could consume as much electricity as 75 million homes. Every chatbot reply, image generation, and background service draws energy from the same wires that power our lights and home air-conditioning. These billion-dollar builds aren’t creating thousands of jobs. They’re leaving communities with higher bills, drained aquifers, and brittle infrastructure.
That’s the uncomfortable truth behind the AI boom: what looks like innovation at the surface is quietly reshaping the social contract of electricity.
We’ve seen this story before. Railroads, oil, and steel - each industrial boom left a permanent mark on rural America. But here’s the difference: a steel mill might employ thousands, while a hyperscale data center can demand the same electricity as an entire city while employing fewer people than a grocery store.
The imbalance is staggering. In Georgia, families are already paying $43 more a month as data centers consume up to 80% of new supply. ERCOT in Texas warns demand could double by 2030.
This series examines how AI’s insatiable appetite for power risks transforming rural towns into the digital equivalent of company towns—captive suppliers of cheap electricity and water, with little to show in return.
Over four parts, we’ll look at:
The power gird is the real bottleneck (why access to power now defines the AI race).
Who pays for innovation (and why the bill keeps landing in residents’ mailboxes).
Data centers as modern company towns (and whether communities can demand equity).
The future of the grid itself (will AI bankrupt it, or force a reinvention?).
AI’s biggest problem isn’t hype; it’s the electricity demand. This demand could triple U.S. electricity use, forcing hard choices about how we power both our communities and our technologies. The media hype cycle celebrates trillion-dollar valuations. The reality is millions of people subsidizing technology that may never provide them positive value.
This series asks a blunt question: Will AI bankrupt rural America or reinvent it?
Subscribe now to follow along. Because the future of AI isn’t being decided in boardrooms—it’s being decided at your local utility commission.
Read the first post here.