Data centers powering artificial intelligence consume significant amounts of water, highlighting the need for greater transparency regarding water use in both existing and planned facilities.
By Alberto Montanari, EOS
Many are impressed by the power of artificial intelligence (AI) and its astonishing capability to make a synthesis of information. But, what we don’t realize is the price that we have to pay in terms of environmental impact for the development of AI.
Privette et al. [2026] assess the multi-faceted water footprint of data centers that make the necessary calculations to empower AI. Their commentary claims more transparency is necessary for communicating the water usage of data centers, and for supporting effective regulation and community planning. Ensuring the sustainability of data centers in the face of growing needs requires a cooperative effort by public administrations, industrial sector and communities of users. The authors emphasize that rigorous monitoring of environmental impact indicators is necessary to ensure that AI effectively supports the progress of our knowledge.
RESEARCH PAPER: Data Centers Water Footprint: The Need for More Transparency
By Ana Pinheiro Privette, Ana Barros, Ximing Cai
Abstract: As artificial intelligence (AI) grows, so does the number of data centers—and they use a lot of water, especially for cooling. While their total water use in aggregate may seem small, in some communities it puts serious pressure on local water systems. The problem is, we do not have data about how much water these centers actually use. This makes it hard for towns and planners to prepare. To grow responsibly and better integrate with the local communities where they operate, tech companies need to be more open about water demand for existing and planned data centers and work closely with local communities to sustainably use shared resources.
