Data Centers Are the New Factories...

And Energy Policy Needs to Catch Up

In the 20th century, factories rose up wherever raw materials, rail lines, and rivers made production possible. Today, it’s server racks instead of smokestacks — and energy is the new iron ore. As data centers emerge as the economic engines of the digital age, site selection and energy policy are converging in ways that demand a rethinking of how we plan, permit, and power our future.

It’s not just a clever metaphor. Like factories, modern data centers are capital-intensive, geographically rooted infrastructure. And like factories, they’re increasingly drawn to regions with access to key resources — only now, it’s cheap electricity, robust transmission, and reliable interconnection. From Virginia to Mississippi, states are branding themselves as data center-friendly, offering massive tax incentives for minimal job creation. In Loudoun County, Virginia, data centers account for one-third of all local tax revenue — more than 80 times what they did just a decade ago.

But the math only works if the megawatts show up.

The rise of artificial intelligence and cloud computing has pushed data center energy consumption into the spotlight. What used to be a rounding error in load forecasts is now a primary driver of electricity demand. Industry projections suggest that data centers could consume up to 9% of U.S. electricity by 2030 — a share comparable to the industrial sector’s footprint in earlier decades. A single hyperscale campus can draw as much power as 750,000 homes. For utilities and regulators accustomed to flat or declining load, it’s a whiplash moment.

And with that demand surge comes siting pressure. The availability of power — not just land or fiber — now determines where data centers can go. Developers are increasingly prioritizing sites with existing substations, fast-track interconnection options, or legacy industrial infrastructure. In some cases, companies are repurposing shuttered power plants as data campuses simply because the electrons are already wired in.

When grid constraints hit, they hit hard. In Northern Virginia, the most concentrated data center cluster in the world, Dominion Energy was forced to warn that dozens of new projects might be delayed due to overloaded transmission infrastructure. It wasn’t a generation shortfall — it was a bottleneck at the substations. Even in regions eager to host data centers, permitting delays, community pushback, and interconnection queue backlogs can turn a shovel-ready project into a stranded one.

That’s why proactive planning matters more than ever. Some utilities are beginning to treat data center growth like they once did aluminum smelters or manufacturing parks — incorporating scenario planning into their integrated resource plans, building out capacity ahead of formal load commitments, and partnering directly with developers on timelines. Meanwhile, states like Georgia and Ohio have streamlined permitting and transmission siting to get ahead of the curve.

But we’re still playing catch-up. Many regulatory bodies remain stuck in a reactive posture — assessing infrastructure needs only after project announcements hit the news. Worse, some rely on outdated assumptions that treat data center loads as marginal or speculative. That’s no longer the case. These projects are real, capitalized, and often tied to multi-decade power purchase agreements with clean energy developers. They’re not just the clients — they’re becoming part of the grid.

It’s time for energy policy to catch up to the economic reality. That means embracing digital infrastructure as critical infrastructure. It means building permitting pathways that match the speed of deployment. It means ensuring that grid modernization, renewable integration, and industrial siting are planned in tandem — not in silos.

We used to build transmission to power factories. Today, we build it to power the cloud.

The digital economy is anchored in physical infrastructure. The servers that train AI, deliver streaming content, and store our data don’t float in the ether. They’re grounded in real communities, drawing real power. Treating them like factories — with all the seriousness, planning, and policy alignment that entails — isn’t just metaphor. It’s what the moment requires.

###