A severe heatwave affecting much of the United States is intensifying concerns about the escalating energy and water demands of artificial intelligence (AI) data centers. These facilities, essential to supporting the rapid growth of AI technologies, consume significant amounts of electricity and water, raising questions about the sustainability of their expansion amid climate-driven infrastructure stresses.
The heatwave, which has brought record temperatures across the Eastern Seaboard, is putting considerable strain on the national power grid and water resources. Temperatures in cities such as Washington, DC, and New York are expected to exceed 100 degrees Fahrenheit (38 Celsius), prompting grid operators like PJM Interconnection, overseeing power supply in 13 states and the District of Columbia, to request data centers switch to backup power within 15 minutes during emergency alerts. This measure aims to free up electricity for residential and commercial usage during peak demand periods.
Data centers currently account for roughly 4 percent of U.S. electricity consumption, a figure projected to climb to 9 percent by 2030 according to the Department of Energy. The majority of these centers are concentrated in the South and Midwest, with 38 percent of Americans living within five miles of one of the approximately 3,000 operational data centers nationwide. Cooling systems alone can consume up to 40 percent of a data center’s power under normal conditions, a demand that increases as ambient temperatures rise.
The rapid pace of data center construction has drawn bipartisan political attention. Texas Governor Greg Abbott has proposed banning new data center construction in rural areas and advocated for facilities to generate their own power and recycle water. Meanwhile, prominent Democratic figures including Senator Bernie Sanders and Representative Alexandria Ocasio-Cortez have called for a moratorium on new data centers. Public opposition reflects these concerns; a recent Gallup poll indicated that 70 percent of Americans oppose local data center development, citing resource consumption as a primary issue.
Beyond energy, water consumption is a critical concern. Many data centers rely on evaporative cooling systems that use potable water, much of which evaporates rather than being recycled. Although their total water use—approximately 627 million gallons per day—is lower than that of agriculture, industry, or residential consumption, demand is rising, especially in water-stressed regions. Since 2022, about two-thirds of new or planned data centers have been located in areas facing water scarcity.
A single large AI data center can consume up to 5 million gallons of water daily, comparable to the needs of a city of 50,000 residents. This dynamic creates a persistent challenge during heatwaves and droughts, when both energy and water supplies are most constrained. In Texas, officials in Corpus Christi have voiced concern over potential water shortages, exacerbated by new data center developments and prolonged drought conditions.
Experts caution that the existing model of data center siting and cooling was designed for average historical conditions, which are becoming obsolete due to climate change. The increasing frequency of extreme weather events underscores the urgent need for more sustainable infrastructure planning, regulatory oversight, and technological innovations to mitigate the environmental footprint of the growing AI sector.