Why AI data centers became a bipartisan local revolt

The public argument over AI infrastructure has moved from abstract promises to a local question: what does a data center change for the people who live beside it?

Nathaniel Whittemore's episode of The AI Daily Brief argues that opposition is rising because residents connect data centers with costs they can see or fear immediately: pressure on the power grid, water consumption, lower property values, noise, and higher taxes. The episode also describes a second reality: data centers can bring tax revenue, construction work, and public infrastructure. The political problem is the gap between those two accounts. Companies and officials make broad claims about economic benefit, while residents want measurable benefits in their own communities. 1

Whittemore cites several recent polling results. Gallup found that 71% of respondents opposed data centers in their area, including 48% who strongly opposed them. HeatMap's measure of opposition near respondents' homes rose from 51% in February to 75%, while strong opposition rose from 24% to 61% over 12 months. 2

The Puck figures cited in the episode show what respondents think is driving that opposition. Seventy-eight percent said local power-grid strain was definitely or probably true of data centers. Seventy-five percent said the facilities use too much local water. Fifty-nine percent associated them with lower property values, and another 59% associated them with constant disruptive noise. Fifty-four percent said the facilities would create too few local jobs. The episode cites 48% for higher local taxes. 2

These figures measure public belief and opposition. They do not establish that every feared effect occurs at the same scale in every town. They do establish why a national promise about AI can become a local campaign against a particular project: residents are evaluating a facility through the bills, roads, utilities, and property around them.

The episode turns to Quincy, Washington, as an example of what a local benefit can look like when officials can attach it to projects and budgets. Quincy has about 30 data centers. Whittemore cites roughly 900 direct engineering and construction jobs, plus an estimate from a Washington State analysis that each direct job supports four to six additional jobs. Data centers contribute 57% of Quincy's property taxes. 2

The cited public projects make the revenue easier to understand. Quincy used data-center-related tax income for a $15 million pool and a $120 million high school, according to the episode. Whittemore also cites the city's poverty rate falling from 29.4% in 2012 to 6.2% in 2024. Those outcomes do not prove that data centers caused every improvement, but they give residents something concrete to inspect: a tax base connected to visible public spending. 2

The episode's argument is strongest when it treats the conflict as a bargaining problem. A town may accept a large facility when the town can see how the facility pays for grid upgrades, water systems, schools, emergency services, or lower household costs. A town may reject the project when the public receives only a forecast of future jobs while carrying immediate infrastructure and environmental risks.

Whittemore mentions a possible benefit for some Dominion Energy customers in Virginia: data-center demand could contribute to lower electricity rates for those customers under a particular arrangement. The example matters because it turns an industry-level benefit into a household-level question. Who receives the savings? When do the savings arrive? What costs remain with other customers? A claim about lower rates becomes politically useful only when those questions have public answers. 2

This also explains why the backlash cuts across party lines. The concerns are rooted in land use, utility capacity, water, taxes, and property values. Those issues belong to local government and household budgets before they belong to a debate about whether AI is beneficial in general. A resident can support AI research and still oppose a facility whose local terms look unfavorable.

The episode points toward a more exact standard for data-center approvals. Officials should publish the project's expected power and water demand, the infrastructure it will fund, the jobs that will remain after construction, the tax revenue the town will receive, and the conditions that trigger additional payments when usage grows. Residents also need a way to compare promised benefits with results after the facility opens.

That standard does not guarantee support. It changes the argument from trust to accounting. Companies can make a case with numbers that residents can check, and residents can challenge assumptions without arguing about the entire future of AI. The approval question becomes local and specific: what costs will this project impose, what benefits will it deliver, and who receives each one?

The episode's title captures the political shift, but the underlying issue is narrower than a general rejection of AI. Opposition rises when communities see a private project drawing on public systems without a clear local return. The projects that survive will be the ones that make the return visible in power, water, taxes, jobs, and public services.

Listen to the full conversation through the Apple Podcasts episode page.