Summer Reading: Data Centers and Artificial Intelligence

There are plenty of ways to pass the time at the beach or by the pool. Whether they involve a Kindle, a quick weather check on the phone, an Uber ride to the boardwalk, or even advanced lifeguard equipment, data centers will inevitably be part of the experience. Despite being criticized by (what seems like) virtually every politician across the country, data centers make modern digital life possible and generate jobs and opportunity in the process. But there is lots of hot air out there swirling around these projects, and the Taxpayers Protection Alliance (TPA) would like to set the record straight. So, lie back in your beach chair, ask your favorite artificial intelligence (AI) application about the local tide charts, and enjoy the first installment of 2026’s “Summer Reading.”

As James Madison Institute policy analyst Turner Loesel notes, “Data centers are foundational infrastructure for the modern economy. In short, they are the computers we use but don’t touch. They power essential services ranging from cloud computing and online commerce to artificial intelligence and secure financial transactions. As the scale and scope of digital services continue to expand, so too does the need for reliable, efficient data centers capable of supporting global connectivity, computation, and storage.”

Besides the minor benefit of making modern life possible, these projects are a boon for the communities they are located in. According to a 2025 report by the American Edge Project, “America’s data center boom is one of the strongest engines of local economic growth. From Meta’s $10 billion Hyperion Advanced Campus in Louisiana to Amazon’s $11 billion AI-optimized facility in Indiana and countless smaller regional projects meeting surging cloud computing demand, data center construction and operation are fueling a surge of new jobs, investment, and tax revenues across the United States. … A typical data center project employs roughly 1,600 to 1,700 local workers during build-out, and adds $243.5 million to the local economy.” Further, the report notes, “Building the fiber and mid-band networks linking data centers to users could generate 3.8 to 4.6 million jobs and contribute $1.4 to $1.7 trillion to U.S. Gross Domestic Product (GDP) by 2030.”

Despite these astounding benefits, naysayers of data centers—and in particular AI data centers—predict a cascade of job losses that will supposedly result from the next iteration of the digital domain. AI pessimists often assume a flawed zero-sum framework: if an AI tool automates a specific task, it inevitably eliminates the corresponding job. Recent economic data demonstrates the opposite outcome. Automating routine tasks lowers operating costs, boosts productivity, increases consumer demand, and elevates human workers to higher-value roles. Data from the Bureau of Labor Statistics reveals robust projected growth across the very fields most exposed to generative AI. For example, employment for data scientists is projected to surge by 33.5 percent and information security analysts by 28.5 percent through 2034, driven by the critical need to structure enterprise data and secure cloud architectures. Software developer jobs are expected to grow up to 15.8 percent, adding over 267,000 net new jobs, while personal financial advisor roles are projected to expand by 10 percent as AI automates back-office analytics and allows advisors to serve larger client bases.

These government projections, which reflect extensive data and incorporate recent employment trends, are backed up by current dynamics in industries that were supposedly due for a contraction because of AI. As a recent report by Build American AI notes, “The clearest decade-long real-world test of artificial intelligence’s interaction with work has produced the opposite of the cataclysmic result predicted in 2016. The radiologist workforce, its training pipeline, and the surrounding labor market in imaging informatics have all grown, and the clinical outcomes of AI-supported pathways are better than radiologist-only pathways in the most rigorous trials published to date.” From 2014 to 2023, the radiologist workforce grew by 17.3 percent; “AI-assisted imaging coincides with more radiologists, more residency slots, and a new class of imaging-informatics and model-governance roles.”

This dynamic reflects the classic Jevons Paradox, which, as R Street Institute Senior Fellow Adam Thierer notes, “is named after a 19th-century English economist who first identified the fact that efficiency improvements for many resources or goods are oftentimes accompanied by an increase, rather than decrease, in total consumption of that resource. This is because lower costs often spur greater demand. This means that AI-enabled automation improvements often encourage all new applications and a boost in overall demand for new systems and services – and the labour that makes them.” Organizations leverage lower costs to build more complex tools, resolve long-neglected technical debts, and deploy new applications, driving higher overall demand for skilled labor. Rather than displacing workers, AI makes occupations such as radiology, software development, and financial planning more accessible and valuable, multiplying the total volume of work that organizations can undertake.

Undeterred by evidence, research, and reasoned argument, detractors of the latest technological advancements continue to yell that the sky is falling…and taking the grid and potable water with it into a dreary abyss. For example, detractors of data centers claim that these projects are guzzling water and electricity and have used these unsupported arguments to push for moratoria across the 50 states. Fortunately, TPA is going state by state, countering this misinformation with statistics on water and energy use. As TPA noted in the case of Maryland, “Prominent Maryland policymakers have claimed — without evidence — that data centers are depleting resources and increasing costs. … [P]olicymakers in Prince George’s County are concerned about the impact of data centers on the ‘regional water supply,’ and local lawmakers will likely cave to a petition campaign that cites environmental fears. Similarly, Baltimore City’s recently passed one-year moratorium on data centers was borne out of water- and electricity-related panic. But, according to TPA’s analysis, data centers accounted for just 0.06 percent of all water consumption in Maryland in 2025. Furthermore, data centers accounted for only 2.5 percent of all Maryland electricity consumption in 2025.”

This may sound surprising, but it really isn’t. AI researcher and policy analyst Andy Masley notes, “in 2030, AI in data centers specifically will be using 0.08% of America’s freshwater. This means it will rise to the level of 5% of America’s current water used on golf courses, or 5% of U.S. steel production, or be about 173 square miles of irrigated corn farms.” Data centers’ share of U.S. annual electricity consumption is also far lower than critics suggest — less than 5% — and states with high load growth are actually seeing inflation-adjusted power price declines.

The truth is far less scary than the dystopian portrait that critics of AI and data centers paint. So, sit back, relax, and enjoy the fruits of these latest technological advances—hopefully on a beach somewhere.