If the US Has to Build Data Centers, Here’s Where They Should Go
Technology companies have invested so much money in building data centers in recent months, it is actively boosting the US economy and the race for AI shows no signs of slowing down. Meta chief Mark Zuckerberg told President Donald Trump last week that the company would spend $600 billion on US infrastructure, including data centers, by 2028, while OpenAI has already committed to spending $1.4 trillion.
An in-depth new analysis examines the environmental footprint of U.S. data centers to get a sense of exactly what the country could face as this construction continues over the coming years – and where the U.S. should build data centers to avoid the most harmful environmental impacts.
The study, published Monday in the journal Nature Communications, uses various data, including demand for AI chips and information on the state’s electricity and water shortage, to project the potential environmental impacts of future data centers through the end of the decade. The study models a number of possible scenarios for how data centers could affect the United States and the planet, and warns that net-zero emissions promises from tech companies likely won’t account for the energy and water needs of the massive facilities they’re building.
Fengqi You, a professor of energy systems engineering at Cornell and one of the analysis’s authors, says the study, which began three years ago, comes “at the perfect time to understand how AI impacts climate systems and water use and consumption.”
The AI industry is “growing much faster than expected,” he adds, especially with the Trump administration’s focus on the industry. “This is all growing so big right now.”
Not all data centers are created equally environmentally: much of their water and carbon footprint depends on where they are located. Some US states may have grids running more on renewable energy, or are making big strides in putting more clean energy on the grid; this significantly reduces carbon emissions from data centers that draw power from these networks. Likewise, states with less water scarcity are better positioned to provide the large quantities of water needed to cool data centers. (Cooling also makes up a significant portion of data center energy consumption.) The best locations for a data center over the next few years in the United States are states that strike a balance between these two inputs: Texas, Montana, Nebraska and South Dakota, according to the analysis, are “optimal candidates for installing AI servers.”
Much of the construction of data centers in the United States has historically focused on places like Virginia, the data center hub of the United States, and Northern California. Proximity to Washington, DC and Silicon Valley was important to data center companies, as was the density of fiber connectivity in these regions and their skilled workforce. Virginia has also offered substantial tax breaks for data centers for years — a technique other states are turning to to attract development. According to Data Center Map, an industry tool that tracks data center development, of more than 4,000 data centers in the United States, more than 650 are in Virginia – the most in the country – and California has more than 320, ranking third.
