The hidden costs of the digital gold rush: When the AI ​​boom meets the reality of rural communities

Thirstier than an entire city? The shocking water consumption of new AI data centers

The current artificial intelligence boom, hailed as the fourth industrial revolution, reveals a remarkable discrepancy between the promises of technology giants and their actual impact on local communities. While companies like Amazon, Microsoft, Meta, and Google plan to invest an estimated $600 billion in AI infrastructure by 2028, a growing bipartisan resistance to the construction of data centers is forming in the United States. This development reveals fundamental economic and social contradictions of a growth strategy based on outsourcing costs to local communities while retaining profits for a few global technology giants.

The scope of this opposition is considerable. According to Data Center Watch, data center projects worth $64 billion have been blocked or delayed in the last two years, $18 billion of which have been completely halted and another $46 billion postponed. These figures do not represent mere statistics, but signal a profound conflict between global capital and local self-determination. At least 142 activist groups in 24 states are organizing against the construction of new data centers, a mobilization that is remarkable because it transcends traditional political boundaries.

The economic deception

The promise of jobs

The rhetoric of technology companies and their political allies consistently emphasizes job creation as the central argument for data centers. However, a closer look at the empirical data reveals a fundamentally different picture. A study commissioned by the lobbying group Data Center Coalition and conducted by PwC claims that the data center industry supported 4.7 million jobs in the US in 2023. However, this figure is highly misleading.

Of these 4.7 million jobs, only 603,900 were actually direct jobs in the data center industry itself. The remaining 4.1 million jobs were classified as indirect or induced jobs, a methodological construct based on the controversial IMPLAN model. This model calculates a multiplier effect of 7.8, meaning that each direct job supposedly creates 7.8 additional jobs in the overall economy. Independent economists such as Nathan Jensen of the University of Texas call these figures unrealistic, pointing out that a multiplier of one to two would be much more plausible.

The reality of job creation is sobering. A typical data center employs between a few dozen and a few hundred people upon completion, depending on its size and business model. Even hyperscale data centers, representing billions of dollars in investments, require only a few dozen full-time employees to operate. A 40-megawatt data center typically employs about 45 people after the construction phase is complete. In contrast, companies and politicians often promise thousands of jobs, a discrepancy systematically reproduced in media reports.

While the construction phase of a data center temporarily creates hundreds to over a thousand jobs in the construction industry, these are temporary and disappear after the project's completion. The often-cited indirect jobs in the service sector, from hospitality to retail, are precarious and poorly paid. They hardly justify the massive tax exemptions and infrastructure investments that municipalities provide for data centers.

Tax incentives and fiscal shifts

The fiscal impact of data centers presents a complex paradox. On the one hand, they generate significant tax revenue for certain municipalities, while on the other, they result in massive state tax losses due to generous incentive programs. At least 41 U.S. states offer tax exemptions for data centers. While the design varies considerably, the basic structure is similar: exemption from sales and use taxes on equipment, construction materials, and often even electricity consumption.

Virginia embodies the fiscal contradictions of this policy in a particularly dramatic way. The cost of the state's data center tax exemption program exploded from $65 million in 2017 to $750 million in 2023, an increase of 1,054 percent in just six years. These losses are borne by all 8.6 million Virginia residents, equivalent to approximately $87 per person, while only certain communities benefit from the revenue.

Loudoun County, Virginia, dubbed the Data Center Capital of the World, illustrates the concentrated benefits. Estimated annual data center tax revenues are $890 million, representing 95 percent of the county's total $940 million operating budget. This revenue comes primarily from taxing the computer equipment within the data centers, not from traditional property taxes. For every dollar of data center tax revenue, the county spends only $0.04 on public services, compared to $0.25 for traditional businesses. This has enabled Loudoun County to maintain the lowest property tax rate in Northern Virginia, about 25 percent lower than neighboring counties.

However, this model creates a precarious fiscal dependency. Projections indicate that tax revenues from computer equipment could rise to $1.37 billion by 2026 and $1.5 billion to $2.5 billion by 2030. These revenues could exceed traditional property tax revenues, creating what county officials themselves describe as a worrying overdependence on a single, volatile revenue source. Computer equipment typically has a lifespan of only a few years and can be relocated relatively easily to other locations if Virginia changes its incentive policies or other regions become more attractive.

The fundamental problem lies in the structure of these fiscal arrangements: diffuse costs and concentrated benefits. While a single county generates massive revenue, the entire state bears the cost of the tax exemptions. Loudoun County's 440,000 residents gain approximately $1,506 per person, while the rest of Virginians lose approximately $87 each. This asymmetry creates political dynamics in which local elites benefit from data centers while the broader social costs are externalized.

Critics argue that these tax incentives are ineffective. Data center location decisions are primarily determined by other factors: access to reliable energy, water supplies, fiber optic infrastructure, and proximity to major internet exchanges. Virginia's cool climate and excellent internet infrastructure would attract data centers even without massive tax breaks. Nevertheless, the state is foregoing hundreds of millions of dollars in revenue that could be used for schools, roads, and other public services.

Resource consumption and ecological externalization

Energy as a limiting factor

Data center energy consumption represents one of the greatest economic and environmental challenges of digital transformation. In 2023, US data centers consumed 183 terawatt-hours of electricity, equivalent to 4.4 percent of total US electricity consumption. By 2030, this consumption is expected to rise to 426 terawatt-hours, an increase of 133 percent. This would mean that data centers would consume between 6.7 and 12 percent of total US electricity consumption.

However, these figures obscure the actual dimensions of individual facilities. Traditional data centers typically require 5 to 10 megawatts of power, while modern hyperscale facilities for artificial intelligence consume 100 megawatts or more. The largest planned data centers are expected to require up to 2,000 megawatts, or 2 gigawatts, equivalent to the output of two large nuclear power plants. Data center campuses in the early planning stages on 50,000 acres could consume up to 5 gigawatts.

This exponentially growing demand is hitting an electrical grid that is already under strain. Goldman Sachs estimates that approximately $720 billion in grid infrastructure investments will be required by 2030 to meet data center demand. These costs are ultimately borne by all electricity customers, leading to rising energy prices for homes and businesses.

The regional impacts are particularly dramatic. In Virginia, data centers consumed approximately 26 percent of the state's total electricity consumption in 2023, a concentration that requires massive investments in new generation capacity. In other states, such as North Dakota, Nebraska, Iowa, and Oregon, data centers consume between 11 and 15 percent of electricity consumption.

The question of energy sources exacerbates environmental concerns. Although technology companies have made commitments to 100 percent renewable energy, the reality paints a different picture. The International Energy Agency predicts that, despite a growing share of renewable energy, gas-fired power generation for data centers will more than double from 120 terawatt-hours in 2024 to 293 terawatt-hours in 2035, with the majority of this growth occurring in the United States. Global Energy Monitor identified 38 gigawatts of gas-fired power capacity under development specifically planned for data centers, representing about a quarter of all such projects.

Some companies are even considering extending the lifespan of coal-fired power plants or building new fossil-fuel power plants to meet their data centers' energy needs. This development directly contradicts national and international climate goals. Researchers warn that the power consumption of artificial intelligence runs counter to the massive efficiency gains needed to achieve net-zero emissions.

For rural communities, the establishment of data centers often means rising electricity bills. A study by the Virginia Legislature estimates that average households in the state could pay an additional $37.50 per month in energy costs due to data centers. The reason lies in the structure of electricity pricing: The costs of grid expansion and new generation capacity are passed on to all consumers, while data centers can often negotiate special rate agreements.

Water as a scarce resource

Data center water consumption represents a growing environmental and economic challenge, especially in water-scarce regions of the United States. A single large data center can consume up to 5 million gallons of drinking water per day, enough to supply thousands of homes or farms. Google, one of the industry leaders, consumed 5.6 billion gallons of water worldwide in 2022, and this consumption is expected to continue to rise due to the generative AI revolution.

Data center water use is concentrated in three main areas. First, direct on-site cooling, which evaporates an average of 0.26 to 2.4 gallons per kilowatt-hour of server power. Second, water-intensive power generation in thermal and hydroelectric power plants, which require an average of 2.0 gallons of evaporated water per kilowatt-hour of electricity consumed. Third, water consumption in the supply chain, particularly in semiconductor production, where the manufacture of a single microchip requires 2.1 to 2.6 gallons of water.

The spatial distribution of data centers exacerbates the water problem. Approximately 20 percent of US data centers draw water from moderately to heavily polluted watersheds in the Western United States. The dry air in these regions makes them technically attractive for data centers, as moisture can cause corrosion and electrical problems in sensitive equipment. At the same time, these regions have the highest marginal costs in terms of water consumption.

Phoenix, Arizona, illustrates the dimensions of the problem. The region is home to over 58 data centers. If each of these data centers uses 3 million gallons of water per day for cooling, this equates to a daily consumption of more than 170 million gallons of drinking water for data center cooling alone. This massive consumption puts a strain on an already fragile water supply and raises ethical questions about whether the needs of tech giants should take precedence over the basic needs of residents and agriculture.

Water pricing reinforces this inequality. In many cases, technology companies pay lower water rates than local residents. In Mesa, Arizona, Google negotiated a rate of $6.08 per 1,000 gallons of water, while residents paid $10.80 per 1,000 gallons. This arrangement sparked outrage among residents who felt the tech giant was receiving preferential treatment at the expense of the community.

The regulatory structure of water pricing contributes to this problem. Water tariffs are often set by public authorities based on the costs of water treatment, distribution, and infrastructure maintenance, rather than by supply and demand in a competitive market. This creates a situation where technology companies can negotiate favorable water tariffs that do not fully reflect the marginal costs of their water use. This leads to a lack of incentive for these companies to conserve water or invest in more efficient cooling technologies.

The drinking water used to cool data centers is often treated with chemicals to prevent corrosion and bacterial growth, making it unsuitable for human consumption or agricultural use. This means that data centers not only consume large quantities of drinking water but also effectively remove it from the local water cycle.

In Georgia, residents near a data center reported disruptions to their water supply, with some saying they can no longer drink the water. These anecdotal reports suggest potential impacts on water quality beyond consumption.

Noise pollution as an underestimated externality

Noise pollution from data centers represents an often overlooked but significant negative externality that impacts the quality of life and health of neighboring communities. The primary sources of noise are diesel generators for backup power, cooling systems, and high electricity consumption, which generate a low-frequency hum.

Diesel generators are the most common backup power source for data centers. Small data centers under 5,000 square feet typically use two to five generators, while hyperscale data centers may require dozens. To ensure their functionality, these generators must be tested at least monthly. Noise emissions vary depending on the size of the generators: Small generators run at around 85 decibels, while larger generators are closer to 100 decibels. Since data centers typically run multiple generators simultaneously, the decibel level increases accordingly.

Cooling systems generate continuous noise. HVAC fans in data centers produce noise levels between 55 and 85 decibels. With the rise of artificial intelligence and data storage needs, servers consume more energy daily. Temperatures rise more quickly when servers have high workloads, so HVAC systems operate continuously at increased speeds to cool the servers and aisles.

For comparison, according to the American Speech-Language-Hearing Association, safe sound levels are 70 decibels or below. Exposure to noises of 85 decibels and above is harmful to hearing. Some data centers reach noise levels of up to 96 decibels in server areas.

A particularly well-documented case is the Great Oaks community in Virginia. John Biess and his wife, Gloria, called the county police in May 2022 to complain about the screeching, humming, and roaring coming from the data centers they were building 600 feet to the north, behind a forest of oak trees. The first police officer who arrived confirmed that it was quite loud. Other residents said the never-ending noise made it difficult to sleep, gave them headaches, and ruined outdoor activities. Some said it was worse at night, a point later confirmed by the Biess's decibel meter, which recorded noise levels as high as 65 decibels at night. The county noise ordinance limits residential noise to 55 decibels at night, but at the time, it exempted noise from cooling systems.

Carlos Yanes, another Great Oaks resident, ordered $20,000 in new windows and moved his one-year-old child's crib to the basement. Several residents were talking about moving away. After many meetings with Amazon and costly engineering work, the data center operator managed to reduce the noise by 10 decibels.

A growing body of research shows that the type of chronic noise emitted by data centers poses a hidden health threat, increasing the risk of high blood pressure, strokes, and heart attacks. Residents describe living near a data center as like having a lawnmower running in their living room 24 hours a day, 7 days a week.

Noise pollution is especially noticeable in rural areas, where massive, featureless buildings replace spaces that were once forests or farmland. Even 60 decibels, the low end of the typical spectrum, sounds like overlapping conversations or background music. People often describe the noise as a hum, a tinny whine, or a low-frequency drone. Data centers operate 24/7, so the volume doesn't increase after hours, but its volume is more noticeable when things get quiet.

Our US expertise in business development, sales and marketing - Image: Xpert.Digital

Industry focus: B2B, digitalization (from AI to XR), mechanical engineering, logistics, renewable energies and industry

More about it here:

• Xpert Business Hub

A topic hub with insights and expertise:

• Knowledge platform on the global and regional economy, innovation and industry-specific trends
• Collection of analyses, impulses and background information from our focus areas
• A place for expertise and information on current developments in business and technology
• Topic hub for companies that want to learn about markets, digitalization and industry innovations

The Anatomy of Resistance

Cross-party opposition

One of the most remarkable aspects of the data center opposition is its bipartisan nature. The opposition to data centers doesn't follow the typical ideological fault lines of American politics. There have been blocked projects in both Republican and Democratic states, and there are both Republican and Democratic officials opposing new projects.

A review of public statements by elected officials in counties with large data center projects found that 55 percent of politicians who had taken public positions opposing data center projects were Republicans and 45 percent were Democrats. This bipartisan opposition is notable because large data center developments tend to be located in Republican-leaning states, with Virginia and Oregon being notable exceptions. Even in Texas, known for being particularly pro-business, there is bipartisan support for additional regulations on data center development in the state Senate.

The motivations of the opposition vary along political lines. Republicans tend to focus on tax incentives and strain on the energy grid, while Democrats are more concerned about environmental impacts and resource consumption. However, both parties agree that they do not want data centers in their communities.

A recent nationwide survey by Heatmap found that only 44 percent of respondents would welcome a data center near where they live. Surprisingly, data centers were less popular than almost every other type of energy project. The American public, according to Heatmap's survey, is more skeptical of data centers, which, once built, are essentially warehouses, than of gas-fired power plants, which emit nitrogen oxide and sulfur dioxide in addition to greenhouse gases. They oppose data centers more than wind farms with their towering turbines and mechanical humming noises, more than battery storage facilities that can erupt in super-hot fires, or even nuclear power plants, long the go-to reference for scary energy facilities.

Successful blockades and moratoriums

Several communities have successfully blocked data center projects or imposed moratoriums, serving as models for other regions. In August 2025, St. Charles, Missouri, became the first city in the nation to impose a citywide, one-year moratorium on data center construction. The moratorium was prompted by widespread public concern about a proposed 440-acre data center project called Project Cumulus, located in an environmentally sensitive area near water well sites and in a floodplain. Residents expressed strong opposition, citing a lack of transparency due to non-disclosure agreements, potential threats to water supplies, strain on electrical infrastructure, and environmental risks. The City Council passed the moratorium unanimously.

This success encouraged other communities. St. Louis is also considering a moratorium on new data center projects after the Planning Commission recommended a pause while rules are drafted. The city's planning director, Don Roe, recommended a temporary pause in a memo. The city's zoning plan, he wrote, is not designed for data centers, facilities that look like warehouses but consume massive amounts of electricity and water.

Tarboro, North Carolina, is another example. After more than five hours of deliberation, the city council voted 6 to 1 against granting a special use permit for a proposed $6.2 billion hyperscale data center on a 50-acre site already zoned for heavy industrial use.

Saline Township, Michigan, voted against rezoning 575 acres for a data center. The landowners and developer Related Digital filed a lawsuit against the township a few days later. In Augusta Township, Michigan, a petition successfully pushed forward a vote on rezoning for a $1 billion data center project.

Prince George's County, Maryland, has placed a pause on all data center development in the region while they continue to investigate potential community impacts. A proposal to convert an abandoned shopping center into a data center sparked intense opposition at local meetings and a 20,000-signature petition to stop the project.

In Ohio, the state imposed a nine-month moratorium on data centers to assess community impacts. This statewide action signals that opposition is rising beyond local communities to the state level.

Internationally, other countries have also imposed moratoriums. The Dutch government imposed a nine-month moratorium on new hyperscale data center permits in February, with exceptions in parts of Groningen and North Holland. Singapore has also implemented a moratorium. As a city-state with a population of just under six million people crammed onto an island half the size of London, local authorities have grown concerned in recent years that data center development has exceeded the country's capacity to supply these facilities with clean electrical energy.

Organizational strategies and community mobilization

Resistance to data centers has developed a sophisticated organizational infrastructure. Virginia has become the focal point of community opposition to data centers in the United States, with 42 activist groups working to slow, stop, or further regulate data center development. Opposition in Virginia is becoming increasingly professional and organized. In 2023, the Data Center Reform Coalition was formed to coordinate efforts among environmental, conservation, and homeowner associations opposing data center projects. The Data Center Reform Coalition is a growing organization, adding new members to its platform as opposition to data centers continues to grow in Virginia.

Experts on data center opposition emphasize the importance of grassroots mobilization. Steven Gonzalez Monserrate, a data center expert, explains that grassroots mobilization has recently had a much greater impact than many in the data center industry expected. In the case of Chandler, Arizona, he worked with a group of individuals who experienced noise pollution as a result of living near data centers. After many years of meetings, protests, and community organizing, they successfully passed the first municipal noise ordinance written specifically for data centers in the United States.

Recommendations for communities affected by planned data centers include: Early organization, as data centers are very secretive and try to make deals behind the scenes, so at the time of announcement, it may seem like nothing can be done. It is therefore important to make noise and raise awareness as quickly as possible. Talk to local politicians to hold them accountable, as they have often been kept out of the loop by the central government and may have been misinformed. Contact local media, which can often be the best advocates. International outreach, as this creates a network around the world facing the exact same issues and can be a great support. Seek out experts in water and electricity infrastructure, zoning laws, and data center design.

In November 2025, a one-day summit was held in Georgia to unite community members, students, and advocates. The summit included workshops, panels, and a networking fair to build successful organizing skills to combat the proliferation of data centers and crypto mining in Georgia communities. The agenda included topics such as successful community opposition, successful organizing strategies, the legal landscape, successful grassroots campaigns, environmental permitting, technical communications, and communication with elected officials.

Structural power asymmetries

Transparency deficit and democratic erosion

One of the most fundamental criticisms of data center development concerns the systematic lack of transparency that undermines democratic decision-making. Data center companies often operate behind non-disclosure agreements and shell companies, making it difficult for communities to make informed decisions about projects that will have profound impacts on their environment, infrastructure, and quality of life.

The Cumulus project in St. Charles illustrates this problem. The developers of CRG Cumulus invoked non-disclosure agreements, preventing full disclosure of project details. Residents sharply criticized the lack of transparency, ultimately leading to a citywide moratorium.

In St. Louis, Lauren Filla, treasurer of the Eco-Socialist Green Party of Eastern Missouri, expressed this frustration: "This is exactly what we predicted, that there would be a kind of dilution of residents' opinions and voices into these areas and away from city leadership. We don't want them to pass the buck. City leadership needs to take responsibility for protecting St. Louis from these monstrosities."

Federal policies have exacerbated this problem. The Trump administration's Big Beautiful Bill included provisions designed to preempt state and local standards by making federal assistance contingent on jurisdictions' willingness to adopt lighter regulatory standards. These provisions effectively closed the door to community participation, eliminating public notice requirements, shortening or circumventing comment periods, and limiting legal remedies traditionally available to residents to fight high-paying projects.

These shifts create a regulatory environment in which tech giants can operate with near impunity, confident that even blatant violations of environmental laws will go unchecked. Environmental laws remain on paper, but political priorities have eroded their power. The United States is recalibrating its governance model to prioritize hyperscale AI development over democratic accountability, exposing already vulnerable communities to the unchecked social, environmental, and infrastructural costs of Big Tech's expansion.

Environmental justice and spatial inequality

The spatial distribution of data centers follows patterns of structural inequality. A national study found that although data centers are not disproportionately located in census tracts with high Environmental Justice Indexes overall, there is a very strong correlation between their location and social vulnerability, with poverty and lower educational attainment being important factors. Almost half of all facilities are located in census tracts whose social vulnerability indicators are above the national median. Regionally, the picture is even clearer: States such as California, Texas, and Illinois host clusters of data centers in areas with high or very high environmental justice burdens.

In California alone, nearly a third of data centers are located in the state's most polluted neighborhoods. This location is not accidental. Low-income neighborhoods and communities of color, areas already burdened by environmental and economic injustice, have weakened political power and a diminished ability to resist powerful corporate interests.

The lack of transparency created by the policies discussed in Part 1 of the series has been used as a valuable tool by Big Tech to stifle meaningful community participation and turn resistance into an uphill battle that is nearly impossible to win.

Regulatory systems frequently fail, and government agencies align themselves with industry. Historically, communities have been forced to fight for their rights to clean air, water, and meaningful participation in decisions that affect their lives. These struggles have been long, grueling, and often uphill battles against corporate giants and the agencies charged with protecting the public. They have also produced some of the most powerful models of grassroots resistance—models that can and should inform today's fight against the unchecked expansion of Big Tech.

Macroeconomic implications and bubble risks

AI infrastructure as an economic risk

The massive investments in AI infrastructure are increasingly raising questions about economic sustainability and the risk of bubbles forming. The world's leading AI infrastructure developers, known as hyperscalers, are investing unprecedented sums. The three largest hyperscalers are expanding their largest US data centers from currently less than 500 megawatts to a planned 2,000 megawatts, a doubling to quadrupling of the capacity of completed projects.

The four largest energy consumers in this group, Amazon, Meta, Microsoft, and Google, could spend an estimated $320 billion on capital expenditures in 2025, primarily on AI infrastructure. This is more than Finland's GDP and just below the total revenue generated by ExxonMobil in 2024. The Stargate Initiative, a collaboration between OpenAI and the US government, aims to invest $500 billion in a network of next-generation AI data centers.

This spending is driving GDP growth and creating market optimism. However, some analysts warn that this wave of spending could mask deeper economic weaknesses. A September 2025 report from Deutsche Bank suggested that without AI-related investments, the US economy might already be in recession. Greg Knapp, Managing Partner at Irons Macroeconomics, explained that all this investment is driving GDP, but the S&P 500 is currently quite unbalanced, creating a risk of an investment collapse, especially given that government spending has reached unprecedented levels.

Many observers draw parallels to the dot-com bubble of the late 1990s. In contrast to that era, when companies struggled to generate revenue, many of today's AI powerhouses generate significant revenue. However, some experts worry that this may not be enough to sustain high spending levels. Some companies are turning to the bond market to finance their infrastructure growth by issuing debt that they plan to repay later. Companies like Oracle, Meta, and CoreWeave have collectively secured billions through debt or private credit to support new data center projects.

A Stanford survey found that business adoption of AI increased to 78 percent by 2024 from 55 percent the previous year. However, companies remain hesitant, citing concerns about costs, technical complexity, and unclear returns. An MIT study from August found that, despite massive investments, 95 percent of US companies that had launched generative AI pilot programs had yet to see tangible business benefits.

The fundamental problem lies in the mismatch between investment and return. Tech giants are investing hundreds of billions in infrastructure based on assumptions about future demand and revenue generation that may not materialize. If these expectations aren't met, the outage could reshape the economy, from stock market crashes to communities being left with massive, empty data centers.

Energy inflation and macroeconomic costs

Growing energy demand from data centers is contributing to inflationary pressures that extend beyond the technology sector. Bank of America estimates that while hyperscalers contribute significantly to increased electricity demand, they do not represent the entire scenario. In reality, the majority of the expected increase in US electricity consumption through 2030 will result from electric vehicles, industrial reshoring, and the electrification of buildings.

This combined demand is hitting an electricity grid that has been underinvested in for decades. The resulting capacity constraints are leading to rising electricity prices for all consumers. In regions with high concentrations of data centers, the impacts are particularly pronounced. An analysis by the Washtenaw County Board of Commissioners in Michigan quoted Michelle Martinez, director of the University of Michigan's Tishman Center for Social Justice and the Environment, who said that data centers could make it impossible for the county to achieve its 2035 net-zero goal and potentially increase wholesale electricity costs by 20 percent, raising prices for ratepayers in the area.

The overall economic costs go beyond energy prices. The required investments in grid infrastructure, estimated at $720 billion globally by 2030, will ultimately be borne by all taxpayers and electricity consumers. This reallocation of resources from other productive investments to support data centers represents opportunity costs rarely considered in technology companies' cost-benefit analyses.

Future scenarios and turning points

The limits to growth

The current development trajectory of the data center industry is encountering several physical and political limitations. Physical constraints include energy, water, cooling capacity, and grid infrastructure. Some jurisdictions or utility companies have imposed freezes or moratoria on power supplies to data centers because they cannot guarantee or meet data center demand. This has prompted data center providers to seek alternative cities or regions and alternative energy sources.

Political boundaries are manifesting themselves in growing local opposition. As Data Center Watch notes, opposition to data center construction is spreading as data center development accelerates elsewhere in the country and will likely follow the same pattern as in Virginia. Grassroots democracy and organized opposition are becoming an increasingly effective obstacle to data center expansion.

Some developers are considering radical alternatives. Space-based data centers could be a viable solution in the next decade. Orbital data centers are expected to dramatically improve efficiency by leveraging the cold vacuum of space for passive cooling and harnessing solar power with up to 40 percent greater efficiency than Earth-based systems. With operating costs of just 0.1 cents per kilowatt-hour compared to 5 cents on Earth and emissions up to 10 times lower, they offer a compelling alternative for sustainable high-performance computing.

Regulatory turnaround

The regulatory landscape is beginning to shift. Numerous states are rethinking their generous tax incentive programs. Georgia passed a bipartisan measure that would have suspended the state's data center sales tax exemption for two years, long enough to study the costs to the state's strained power grid and water systems. Governor Brian Kemp vetoed the legislation, citing the need to support existing investments. Environmental and consumer advocates called the veto a gift to an industry already benefiting from generous federal support.

At the local level, municipalities are developing more demanding regulatory approaches. St. Louis passed an executive order that sets standards for data center development without imposing a complete moratorium. The ordinance requires data centers to use renewable energy, implement enhanced noise mitigation measures, and conduct comprehensive environmental impact assessments.

The Washtenaw County Board of Commissioners passed a data center resolution to support local data center decisions, offering county assistance in collecting data on expected water and energy consumption, noise, and other environmental impacts. Commissioners would also assist municipalities in developing public awareness plans in the form of information on expected impacts.

Alternative development models

Critics of current data center development argue for alternative models that emphasize community benefits. These include stricter local hiring requirements, binding environmental protection agreements, community ownership of data centers, and differentiated regulation based on size and environmental impact.

Some experts are calling for a fundamental overhaul of how data centers are integrated into local economies. Instead of treating them as mere tax revenue generators, municipalities could require data centers to make measurable contributions to local infrastructure, education, and environmental protection. This could include investments in renewable energy generation, water treatment plants, and local training programs.

The debate over data centers touches on fundamental questions about economic development, environmental justice, and democratic decision-making. As tech giants continue to invest massive sums in AI infrastructure, resistance is growing from communities who bear the real costs of this development. The coming years will show whether this resistance is strong enough to force a more sustainable and equitable model of technological development, or whether the power of global capital will continue to overwhelm local concerns.

Economic analysis reveals that the current data center boom is based on an unsustainable externalization of costs. The promises of jobs are proving exaggerated, the tax incentives fiscally inefficient, and the environmental impacts significant. The cross-party opposition signals that these insights have penetrated broad sections of the population. The question is no longer whether the current model needs to be reformed, but how quickly and comprehensively this reform will take place.

A new dimension of digital transformation with 'Managed AI' (Artificial Intelligence) – Platform & B2B Solution | Xpert Consulting - Image: Xpert.Digital

Here you will learn how your company can implement customized AI solutions quickly, securely, and without high entry barriers.

A Managed AI Platform is your all-round, worry-free package for artificial intelligence. Instead of dealing with complex technology, expensive infrastructure, and lengthy development processes, you receive a turnkey solution tailored to your needs from a specialized partner – often within a few days.

The key benefits at a glance:

⚡ Fast implementation: From idea to operational application in days, not months. We deliver practical solutions that create immediate value.

🔒 Maximum data security: Your sensitive data remains with you. We guarantee secure and compliant processing without sharing data with third parties.

💸 No financial risk: You only pay for results. High upfront investments in hardware, software, or personnel are completely eliminated.

🎯 Focus on your core business: Concentrate on what you do best. We handle the entire technical implementation, operation, and maintenance of your AI solution.

📈 Future-proof & Scalable: Your AI grows with you. We ensure ongoing optimization and scalability, and flexibly adapt the models to new requirements.

More about it here:

• The Managed AI Solution - Industrial AI Services: The key to competitiveness in the services, industrial and mechanical engineering sectors

☑️ Our business language is English or German

☑️ NEW: Correspondence in your national language!

Konrad Wolfenstein

I would be happy to serve you and my team as a personal advisor.

You can contact me by filling out the contact form or simply call me on +49 89 89 674 804 (Munich) . My email address is: wolfenstein ∂ xpert.digital

I'm looking forward to our joint project.

☑️ SME support in strategy, consulting, planning and implementation

☑️ Creation or realignment of the digital strategy and digitalization

☑️ Expansion and optimization of international sales processes

☑️ Global & Digital B2B trading platforms

☑️ Pioneer Business Development / Marketing / PR / Trade Fairs