{"id":10193,"date":"2026-04-06T20:22:01","date_gmt":"2026-04-06T20:22:01","guid":{"rendered":"https:\/\/wildgreenquest.com\/?p=10193"},"modified":"2026-04-06T20:22:01","modified_gmt":"2026-04-06T20:22:01","slug":"how-much-water-does-ai-use-an-expert-analysis-of-the-real-footprint","status":"publish","type":"post","link":"https:\/\/wildgreenquest.com\/?p=10193","title":{"rendered":"How Much Water Does AI Use? An Expert Analysis of the Real Footprint."},"content":{"rendered":"<p><br \/>\n<\/p>\n<div>\n<div class=\"justify-start\">\n<nav class=\"align-left col-span-full mb-base\" data-pom-e2e-test-id=\"breadcrumbs\"\/>\n<p>An expert on environmental policy measured every drop of water he used during months of heavy AI work. The findings reveal we may be worrying about the wrong environmental crisis.<\/p>\n<\/div>\n<div class=\"border-border-light border-t py-base-tight\">\n<div class=\"flex h-4 justify-between\">\n<div class=\"flex gap-x-base-tight\">\n<div class=\"\"><button class=\"inline-flex shrink-0 items-center justify-center rounded-full hover:cursor-pointer bg-bg-surface hover:bg-bg-light focus:bg-bg-dark text-primary border border-solid border-border-light aria-pressed:bg-brand-primary aria-pressed:text-text-surface py-very-tight px-base-tight gap-tight font-semibold font-utility-2 opacity-50\" aria-label=\"Loading audio\" aria-pressed=\"false\" id=\"article-listen-button\" disabled=\"\"><svg xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"18\" height=\"18\" viewbox=\"0 0 18 18\" fill=\"none\" class=\"\"><title>Listen to this article<\/title><path d=\"M12.5265 16.2C12.2326 16.2 12.0122 16.1265 11.7184 16.053C11.2775 15.9061 10.9102 15.6122 10.6898 15.2449C10.4694 14.8775 10.3959 14.4367 10.5428 13.9959L11.5714 10.3959C11.6449 10.1755 11.7184 9.95509 11.8653 9.80815C12.0122 9.66121 12.1592 9.51427 12.3796 9.36733C12.6 9.29386 12.8204 9.2204 13.0408 9.14693C13.2612 9.14693 13.4816 9.14693 13.702 9.2204H13.7755C14.2163 9.36733 14.5837 9.58774 14.951 9.95509C14.951 9.66121 14.951 9.29386 14.951 8.99999C14.951 7.38366 14.2898 5.8408 13.1877 4.73876C12.0122 3.56325 10.5428 2.9755 8.92652 2.9755C7.31019 2.9755 5.76733 3.63672 4.66529 4.73876C3.48978 5.8408 2.82856 7.38366 2.82856 8.99999C2.82856 9.29386 2.82856 9.66121 2.82856 9.95509C3.12244 9.66121 3.56325 9.36733 4.00407 9.2204H4.07754C4.29795 9.14693 4.51835 9.14693 4.73876 9.14693C4.95917 9.14693 5.17958 9.2204 5.39999 9.36733C5.6204 9.4408 5.76733 9.58774 5.91427 9.80815C6.06121 9.95509 6.13468 10.1755 6.20815 10.3959L7.23672 13.9959C7.38366 14.4367 7.31019 14.8775 7.08978 15.2449C6.86938 15.6122 6.50203 15.9061 6.13468 16.053C5.76733 16.2 5.39999 16.2 5.03264 16.2C4.66529 16.1265 4.29795 16.053 4.00407 15.9061C3.63672 15.6122 3.41631 15.3918 3.12244 15.0979C2.90203 14.8775 2.75509 14.5102 2.68162 14.1428L2.38774 13.0408L2.2408 12.6735C2.2408 12.6 2.16733 12.453 2.16733 12.3796C1.87346 11.2775 1.79999 10.1755 1.79999 8.99999C1.79999 7.08978 2.53468 5.25305 3.9306 3.9306C5.25305 2.60815 7.08978 1.79999 8.99999 1.79999C10.9102 1.79999 12.7469 2.53468 14.0694 3.9306C15.3918 5.25305 16.2 7.08978 16.2 8.99999C16.2 10.102 16.053 11.2775 15.8326 12.3796C15.8326 12.453 15.7592 12.6 15.7592 12.6735L15.6857 12.9673V13.0408L15.1714 14.1428C15.0979 14.5102 14.8775 14.8041 14.6571 15.0979C14.4367 15.3918 14.1428 15.6122 13.849 15.7592C13.5551 15.9061 13.1877 16.053 12.8204 16.053C12.7469 16.1265 12.6735 16.2 12.5265 16.2ZM3.71019 13.849C3.78366 14.0694 3.85713 14.2163 4.00407 14.4367C4.15101 14.6571 4.29795 14.7306 4.51835 14.8775C4.73876 14.951 4.8857 15.0245 5.10611 15.0979C5.32652 15.0979 5.54693 15.0979 5.76733 15.0245C5.98774 14.951 6.06121 14.8775 6.13468 14.7306C6.20815 14.5837 6.20815 14.4367 6.20815 14.2898L5.17958 10.6898C5.17958 10.6163 5.10611 10.5428 5.10611 10.4694C5.03264 10.4694 4.95917 10.3959 4.8857 10.3224C4.81223 10.3224 4.73876 10.249 4.66529 10.249C4.59182 10.249 4.51836 10.249 4.44489 10.249H4.37142C3.9306 10.3959 3.63672 10.6898 3.41631 11.0571C3.26938 11.351 3.19591 11.7184 3.26938 12.0857C3.34285 12.3061 3.34284 12.5265 3.41631 12.6735L3.71019 13.849ZM13.2612 10.3224C13.1877 10.3224 13.1877 10.3224 13.2612 10.3224C13.1143 10.3224 13.0408 10.3224 12.9673 10.3959C12.8939 10.4694 12.8204 10.4694 12.8204 10.4694C12.7469 10.5428 12.7469 10.6163 12.7469 10.6898L11.7184 14.2898C11.6449 14.4367 11.7184 14.5837 11.7918 14.7306C11.8653 14.8775 12.0122 14.951 12.0857 15.0245C12.3796 15.0979 12.5265 15.0979 12.7469 15.0979C12.9673 15.0979 13.1877 15.0245 13.3347 14.8775C13.5551 14.8041 13.702 14.6571 13.849 14.4367C13.9959 14.2898 14.0694 14.0694 14.1428 13.849L14.5102 12.6735C14.5837 12.5265 14.5837 12.3061 14.6571 12.1592C14.7306 11.7918 14.6571 11.4245 14.5102 11.1306C14.2898 10.7633 13.9959 10.4694 13.5551 10.3224H13.4816C13.3347 10.3224 13.2612 10.3224 13.2612 10.3224Z\" fill=\"currentColor\"\/><\/svg><span class=\"hidden sm:inline\">Listen<\/span><\/button><\/div>\n<\/div>\n<\/div>\n<\/div>\n<div><\/p>\n<p class=\"fp-leadCaption py-tight text-left font-utility text-utility3-size leading-utility3-line-height text-secondary\"> (Photo: Craig Hastings \/ Getty)<\/p>\n<\/div>\n<\/div>\n<div>\n<div class=\"article-body\">\n<p>Published April 6, 2026 02:14PM<\/p>\n<\/div>\n<p>People are worried about Artificial Intelligence (AI) and its use of water. If you care about the Colorado River, if you have watched Lake Powell\u2019s water level drop and Lake Mead shrink, and have felt the dread of living in a place that is running out of its most essential resource, then hearing that a new technology is guzzling water hits a nerve. It should. The instinct to protect what is disappearing is a good one. But it turns out that AI isn\u2019t as dire a threat to our water as people may think.<\/p>\n<p>I work on the Colorado River water for a living as a filmmaker and storyteller. I have a PhD in engineering and public policy. I am Din\u00e9. The threats to the river are not abstract to me; they are very real. So earlier this year, I decided to quantify something that has been missing in the conversation about AI and water: I measured my own AI water use.<\/p>\n<p>For 11 weeks, I tracked all of my AI use. One hundred sessions. I counted the tokens processed and applied publicly available numbers on per-token energy and water intensity from Epoch AI and operator-reported data from Microsoft and Google. Anyone can run this math.<\/p>\n<p>In those 11 weeks, I built an iOS app from scratch and wrote policy briefs on extreme heat for nonprofits I work with. I produced documentary pitch decks and drafted a 15,000-word climate fiction piece about the Colorado River collapse. I used AI every single day, often for hours at a time.<\/p>\n<p>Total lifecycle water footprint of all that work: about five gallons. That accounts for everything: the water used to cool the data centers, the water consumed at power plants to generate the electricity, and the water embedded in manufacturing the hardware.<\/p>\n<p>When an <i>Outside <\/i>editor reached out to ask me to write this story, I was on a trip to Marble Canyon, Arizona, to train raft guide companies on what is happening with the river. I drove my diesel Sprinter van from Tucson to the site, which tallied 383 miles at 20 miles per gallon of gasoline. When I ran the numbers later, the lifecycle water footprint of my fuel was around 110 gallons. One drive to the work I do on the Colorado River used more than 20 times the water of everything I did with AI in 11 weeks. That comparison stopped me cold\u2014and I study this for a living.<\/p>\n<p>You may have read stories about how data centers use lots of water, and how these massive warehouses of computer servers are being built across the country to help with the expanding use of AI. Here is the part that I think gets lost in the discourse. All U.S. data centers combined\u2014not just AI, <i>all of them<\/i>\u2014account for roughly 0.3 percent of total national water withdrawals. Agriculture consumes approximately 80 percent of Colorado River water. In my home state of Arizona, agriculture is 86 percent of the state\u2019s water use. These are not competing concerns on the same scale. They are separated by orders of magnitude.<\/p>\n<p>I know what the next question is, because I get it every time: Sure, but AI is growing exponentially. Will it not eventually become the problem? It is a fair question, and it deserves a real answer.<\/p>\n<p>The evidence says no. Inference efficiency, meaning how much energy it takes to actually answer a single query, is improving dramatically. A 2025 Microsoft Research paper found that combined advances in hardware, software, and model architecture can deliver 8 to 20 times reductions in energy per query. The cost of running AI systems comparable to GPT-3.5 dropped more than 280 times between late 2022 and late 2024. Hardware efficiency gains are running at about 40 percent per year. And AI companies have an enormous financial incentive to keep pushing efficiency.<\/p>\n<p>Electricity is one of the biggest line items on their balance sheets. They are not going to burn more power than they have to. Even the International <a rel=\"nofollow\" target=\"_blank\" class=\"text-brand-primary underline hover:text-brand-primary\/85 break-words overflow-wrap-anywhere underline-offset-[3px]\" data-afl-p=\"0\" href=\"https:\/\/build-up.ec.europa.eu\/en\/resources-and-tools\/publications\/international-energy-agency-iea-new-report-energy-and-ai\">Energy Agency\u2019s 2025 Energy and AI report<\/a> projects data centers will account for roughly three percent of global electricity by 2030. That is worth monitoring. It is not the crisis.<\/p>\n<p>It is also worth understanding where AI\u2019s water footprint actually comes from. Most of it is not water running through a data center\u2019s cooling towers. Most of the water used for AI is from generating the electricity that powers the servers. That is scope-2 water, the water consumed at power plants through evaporative cooling and steam generation. We do not hold this against any other electricity consumer. Nobody is calculating the water footprint of your refrigerator or your electric car or the subway.<\/p>\n<p>But when AI uses that same grid electricity, suddenly it is labeled a water crisis. The water is real. The inconsistency in how we talk about it is also real.<\/p>\n<p>The current crisis facing the Colorado River is the same one it\u2019s faced for 100 years. The Colorado River was divided up in 1922 based on flow measurements that, as it turned out, were calculated from some of the wettest years on record. We over-allocated water use from a river we overestimated, and then the climate started warming.<\/p>\n<p>Streamflow has dropped roughly 20 percent since 2000. The math was never going to work. It is not working now. And the 2026 Compact renegotiation, the most consequential water policy event in a century, is happening right now with a fraction of the public attention that a viral video about ChatGPT\u2019s water use receives. Alfalfa irrigation in California\u2019s Imperial Valley alone consumes over 800 billion gallons a year. That is where the water is going. That is what needs your attention.<\/p>\n<p>I am not asking anyone to stop caring about AI\u2019s water use; I am asking you to instead focus on the scale of water use by all of the industries that rely on the Colorado River. At the local level, a single data center can stress a small community\u2019s water supply, and that is worth watching. But when the broader discourse frames AI as the driver of the Western water crisis, it pulls focus from the systems that actually drain the river. The river needs you. It just needs you pointed at the 80 percent, not the 0.3.<\/p>\n<h2><b><i>How I Calculated My Research<\/i><\/b><\/h2>\n<p><i>Water withdrawal and use data come from the U.S. Geological Survey\u2019s 2021 national water use estimates and the Bureau of Reclamation\u2019s Colorado River accounting. Per-token energy intensity is drawn from Epoch AI and Lin (2025), with water use derived from operator-reported Water Usage Effectiveness data published by Microsoft and Google. The 0.3 percent data center figure is consistent with Lawrence Berkeley National Laboratory\u2019s 2024 U.S. data center energy report. Inference efficiency projections reference a 2025 Microsoft Research paper on AI inference energy pathways (Oviedo et al.). The IEA\u2019s base case projection for global data center electricity demand comes from its 2025 Energy and AI special report. The diesel lifecycle water footprint is calculated using Argonne National Laboratory\u2019s GREET model. Colorado River over-allocation history and streamflow decline data are drawn from Bureau of Reclamation records and peer-reviewed hydrology research, including Udall and Overpeck (2017) on Colorado River flow loss and Milly and Dunne (2020) on evapotranspiration trends.<\/i><\/p>\n<p><!-- --><\/div>\n<p><br \/>\n<br \/><a href=\"https:\/\/www.outsideonline.com\/outdoor-adventure\/environment\/ai-water-use-colorado-river-footprint\/\">Source link <\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>An expert on environmental policy measured every drop of water he used during months of heavy AI work. The findings reveal we may be worrying about the wrong environmental crisis. Listen to this articleListen (Photo: Craig Hastings \/ Getty) Published April 6, 2026 02:14PM People are worried about Artificial Intelligence (AI) and its use of<\/p>\n","protected":false},"author":1,"featured_media":10194,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[35],"tags":[],"class_list":{"0":"post-10193","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-wild-living"},"_links":{"self":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts\/10193","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=10193"}],"version-history":[{"count":0,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts\/10193\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/media\/10194"}],"wp:attachment":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=10193"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=10193"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=10193"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}