The AI Boom Is Turning Unlikely Places Into Infrastructure Hotspots

For decades, America’s data centers clustered around established technology hubs. Silicon Valley, for example, is home to many of the companies that built the modern internet, while Northern Virginia's proximity to Washington, D.C., and its role as a major digital communications nexus have helped make it one of the industry's largest markets.

Power and land shortages in those hubs had already begun pushing some development elsewhere. Then the rapid adoption of consumer-facing AI tools beginning in late 2022 unleashed a wave of investment in computing capacity that established markets struggle to absorb¹.

To accommodate this growth, developers are looking outside the industry's traditional strongholds, including in rural communities and smaller cities. Arizona, Texas, Ohio, North Carolina and Wyoming are among the places attracting major AI infrastructure investment, says William Kellogg, vice president of strategic markets at Spectrum Business®. The draw is primarily access to electricity and land. But securing those resources is only part of making a new AI campus viable. “Delays in any resource that is not coordinated at the genesis might be a delay in turning up the facility,” Kellogg says. “Network builds are much greater than just the campus being planned.” A promising site can be far from the high-capacity networks needed to connect it with customers and other computing facilities.

The Expanding Geography Of AI Infrastructure

For large-scale projects, developers typically need hundreds of acres of land, a climate or water supply that keeps servers cool and county and state governments willing to support the build. For operations devoted to training AI models, locations far from major population centers can make sense. Unlike applications that serve users directly, training doesn't require near-instant responses, giving developers more geographical flexibility.