Recent polling indicates that a solid majority of the public is uncomfortable with the rapid spread of AI-powered data centers. A Gallup survey released earlier this year found that 71 percent of Americans would rather not see such facilities built near their homes, and the sentiment cuts across party lines—Democrats, Republicans and independents all registered majorities against local construction.
Historically, these structures were treated like ordinary warehouses: developers applied for zoning, received a stamp, and the building began. That low-profile approach changed roughly two years ago, when journalists started probing the environmental and social footprints of the industry. Nicole Thomas, senior vice president for media and community engagement at JSA, notes that the sector now recognizes the need for community engagement and public education to explain why every smartphone app ultimately relies on a data center to store and process information.
Texas puts a pause on new AI-powered facilities
In September, Texas Governor Greg Abbott ordered state regulators to suspend all permits for new data centers that would support artificial intelligence until comprehensive audits could assess their impact on the power grid and water resources. The move reflects growing concern over the massive electricity draw and water consumption associated with large-scale AI workloads, as well as the noise and visual footprint of sprawling campuses. Abbott’s directive is one of the most visible examples of a state-level response to the national backlash.
From Utah to Indiana: communities test the limits of transparency
Kevin O’Leary’s Utah proposal sparks surprise
When entrepreneur Kevin O’Leary, known as “Mr. Wonderful,” announced plans for a 1-gigawatt data center on a 1,300-acre site in Utah, the proposal caught many residents off guard. The project mirrors the scale of facilities needed for advanced artificial intelligence models, and it quickly ran into the same 70 percent opposition figure that appears in national polls. Critics argue that the size alone would demand a level of energy and water that could strain local supplies, while supporters point to potential tax revenue and job creation.
Hyperscaler arrival in Indiana meets criticism
In Indiana, a large hyperscale operator set up shop in a mid-sized city, prompting local activists to demand clearer communication. Thomas, who grew up in Indiana, describes the rollout as “basically non-existent” in terms of public outreach. Residents complained that they were not informed about the facility’s capacity, which is far smaller than the gigawatt projects but still significant enough to raise questions about long-term environmental impact. The episode underscores the industry’s broader challenge: aligning rapid deployment with genuine neighborhood dialogue.
Pennsylvania’s $165 million data-center showdown
Hazle Township, a community of roughly 4,500 people in the Poconos, received letters from developer NorthPoint offering a $10,000 payment to each resident in exchange for permission to construct a 1,300-acre, $165 million data center. Simultaneously, about 30 households in neighboring Sugarloaf Township faced eminent-domain notices for a high-voltage transmission line—dubbed Project Hazelnut—intended to power the same facility. Opposition groups, coalescing under the banner Alliance to Stop the Line, lodged 20 formal protests and filed right-to-know requests that revealed Amazon as the prospective end user, though the company later said it was merely evaluating sites.
The Pennsylvania episode reflects a larger trend: According to the same Gallup poll, 75 percent of Democrats and 63 percent of Republicans oppose local data center construction. A Data Center Watch report released in June recorded that at least 75 projects valued over $130 billion were delayed or canceled in the first quarter of the year, while the number of active opposition groups more than doubled—from 396 to 833—in the same period. Scholars such as Tatishe Nteta argue that the public’s resistance stems from a communication gap: benefits are rarely highlighted, while costs—environmental, economic, and job-related—are front-and-center.
Looking ahead, industry spokespeople like Thomas predict a three-to-five-year window in which sustained outreach could shift perceptions. By demystifying the difference between a modest 5-megawatt facility—unlikely to host large-scale AI models—and a gigawatt-scale campus, the sector hopes to plant “educational seeds” that mature into informed community decisions. Whether those seeds grow into acceptance or continued opposition will likely depend on how transparently developers address the twin challenges of energy consumption and environmental impact while delivering tangible economic benefits.
