Logo
FrontierNews.ai

Tech Giants Say AI Data Centers Are Getting Less Thirsty, But There's a Catch

Tech companies are deploying new cooling technologies to slash water consumption at AI data centers, addressing mounting public backlash over environmental impact. Nvidia reported in June that its latest system, called DSX, can nearly eliminate water consumption at some facilities, while Microsoft and Amazon Web Services (AWS) say their closed-loop cooling systems involve no net water loss. However, independent researchers caution that cutting water use often means burning more electricity, creating a hidden environmental trade-off.

Why Are Data Centers Consuming So Much Water?

Data centers are massive warehouses filled with computer servers that power the internet and increasingly run artificial intelligence systems. These servers generate intense heat, and keeping them cool requires enormous amounts of water. According to consultancy Rystad Energy, data centers consumed 222 billion liters, or about 59 billion gallons, of water worldwide in 2025 for cooling purposes. Without adaptive measures, that figure could nearly triple to 644 billion liters by 2030.

The scale of this problem has sparked significant public anger across the United States. Communities are pushing back against new data center projects, concerned about water depletion and power grid strain. This pressure is forcing tech giants to invest in solutions that can demonstrate environmental responsibility.

How Are Companies Reducing Water Use in Data Centers?

The leading approach involves closed-loop cooling systems, where liquid flows directly through the servers, as close as possible to the chips themselves. These chips can reach temperatures above 80 degrees Celsius, or 176 degrees Fahrenheit. Nvidia's DSX system takes a different approach by allowing the cooling liquid to enter servers at a warmer temperature of 45 degrees Celsius, compared to the roughly 32 degrees Celsius used by most other closed-loop systems in 2024.

By starting with warmer water, Nvidia reduces the need for year-round air cooling. According to Josh Parker, Nvidia's head of sustainability, simple fans circulating air are often sufficient, though extreme climates or heat waves may still require additional cooling methods. Microsoft, AWS, and Meta have all adopted similar closed-loop systems and report improved water efficiency.

Steps to Understand the Water-Energy Trade-Off in Data Center Cooling

  • The Core Problem: There exists a direct trade-off between water consumption and energy consumption when cooling data centers, according to Andy Masley, an independent researcher who covers AI and data centers. Reducing water use typically requires more electricity to cool the liquid in sealed pipes through air circulation.
  • Hidden Water Costs: The water a data center uses directly is only part of the story. Water is also consumed to generate the electricity powering the facility and to manufacture its chips and servers. In the United States, this hidden water use can be twice the amount a data center consumes on its own.
  • Economic Incentives: Because water is generally much cheaper than electricity, companies have less incentive to cut water use on cost grounds alone. Instead, the primary motivation is public relations amid growing backlash to data centers across the United States, according to Shaolei Ren, an engineering professor at the University of California, Riverside.

The efficiency improvements reported by major cloud providers are substantial. Microsoft improved its water use efficiency by 25 percent between 2022 and 2025, while AWS achieved a 37 percent improvement over the same period, according to their most recent sustainability reports. However, both companies used more water overall during this timeframe as they expanded their data center footprints.

What's Preventing Faster Adoption of Water-Saving Technology?

One significant barrier is cost. Upgrading older data centers to newer, less thirsty technology requires substantial capital investment. However, this obstacle may matter less than it initially appears. Older data centers are smaller and less powerful than the enormous new facilities being built today, so they require less cooling in the first place, according to Minh K. Le, who leads data center and hydrogen research at Rystad.

Another challenge is inconsistency in how companies report environmental data. There is no industry-wide standard for how organizations disclose their environmental, social, and governance, or ESG, efforts. Elon Musk's SpaceX, which became a major player in data centers after acquiring his artificial intelligence company xAI, has never published an ESG report. In June, ratings agency MSCI gave SpaceX its lowest ESG score.

"There's a pretty direct trade-off between how much water is used and how much energy is used for temperature control," explained Andy Masley, an independent researcher who covers AI and data centers.

Andy Masley, Independent Researcher, AI and Data Centers

The path forward remains uncertain. While tech giants are making genuine progress on water efficiency, the underlying tension between water and energy consumption suggests that solving one environmental problem may simply shift the burden to another. As data center demand continues to surge alongside AI adoption, the industry will need to find ways to address both challenges simultaneously, not just trade one for the other.