--- title: How many cars’ worth of CO2 will US data centres really... description: A Cornell study published in Nature Sustainability projects US AI data centres could emit 24 to 44 million tonnes of CO2-equivalent annually by 2030,... canonical: https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How many cars’ worth of CO2 will US data centres really emit? | daily.dev og:description: A Cornell study published in Nature Sustainability projects US AI data centres could emit 24 to 44 million tonnes of CO2-equivalent annually by 2030,... og:url: https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg og:image: https://api.daily.dev/og/posts/Nbn4RGJCG.png og:image:alt: How many cars’ worth of CO2 will US data centres really emit? og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # How many cars’ worth of CO2 will US data centres really emit? **[The Next Web](https://daily.dev/sources/tnw)** · 4 min read · 1 upvotes · 0 comments ## Summary A Cornell study published in Nature Sustainability projects US AI data centres could emit 24 to 44 million tonnes of CO2-equivalent annually by 2030, equivalent to 5 to 10 million cars, not the 24 million cars figure that spread widely online due to a misreading of the tonnes figure. The research also projects a water footprint of 731 to 1,125 million cubic metres yearly. The range depends heavily on the electricity grid mix: cheap renewables could cut emissions over 15%, while best-practice siting and procurement could reduce emissions by up to 73% and water use by up to 86%. Location matters too, with the Midwest and states like Texas, Montana, Nebraska and South Dakota favored over water-stressed Northern Virginia. The figures are based on planned rather than built capacity, and hyperscalers are pursuing both clean power purchase agreements and gas-fired hedges. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How much CO2 will US AI data centers actually emit by 2030 according to the Cornell study? US AI servers are projected to emit between 24 and 44 million tonnes of CO2-equivalent a year by 2030, equivalent to the emissions of 5 to 10 million cars, not the 24 million cars figure that circulated widely online. The range reflects a spread between a restrained build-out and a more aggressive one, modeled using a hybrid statistical and thermodynamic server efficiency model paired with the US ReEDS grid model. _Anyone weighing AI's real environmental cost can follow accurate coverage of data center emissions research on daily.dev._ ### Why is the '24 million cars' statistic about AI data center emissions wrong? The figure misreads the Cornell study's projection of 24 to 44 million tonnes of CO2-equivalent annually as if it meant 24 million cars, when the researchers themselves translated the emissions into 5 to 10 million cars' worth. The error appears to stem from confusing a tonnage figure with a vehicle count, inflating the popularized number by roughly 2.5 times. _Developers tracking AI's infrastructure footprint can verify claims like this by following primary research coverage on daily.dev._ ### How much could data center emissions be reduced through best practices like site selection? Best-practice interventions in siting and procurement could cut data center emissions by up to 73% and water use by up to 86%, according to the Cornell study's modeling. Locations such as the Midwest, Texas, Montana, Nebraska, and South Dakota are identified as better choices than water-stressed regions like Northern Virginia, and cheap-renewables grid scenarios alone could lower emissions by more than 15%. _Engineers weighing sustainable infrastructure choices can track this kind of siting research through daily.dev._ ## Similar posts on daily.dev - [Datacentres drive up big tech’s carbon emissions to a third of those of France](https://daily.dev/posts/datacentres-drive-up-big-tech-s-carbon-emissions-to-a-third-of-those-of-france-bhif4axs1) · Hacker News · 0 upvotes · 0 comments - [Microsoft emissions jumped 25% as AI data centres grow](https://daily.dev/posts/microsoft-emissions-jumped-25-as-ai-data-centres-grow-kffvskftv) · The Next Web · 0 upvotes · 0 comments - [Stop worrying about your AI footprint. Look at the big picture instead.](https://daily.dev/posts/stop-worrying-about-your-ai-footprint-look-at-the-big-picture-instead--jc0bbqdqy) · MIT Technology Review · 0 upvotes · 0 comments - [US gas power projects tripled in 2025 on AI demand](https://daily.dev/posts/us-gas-power-projects-tripled-in-2025-on-ai-demand-woyyvbubn) · The Register · 0 upvotes · 0 comments - [Quantum Computing Could Cut Industrial Emissions Without an AI-Sized Footprint, BCG Finds](https://daily.dev/posts/quantum-computing-could-cut-industrial-emissions-without-an-ai-sized-footprint-bcg-finds-ioammjsrz) · The Quantum Insider · 0 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#sustainability](https://daily.dev/tags/sustainability) [View this post on daily.dev](https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"How many cars’ worth of CO2 will US data centres really emit?","url":"https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg"},"datePublished":"2026-08-18T10:25:09.017Z","dateModified":"2026-09-14T06:05:45.254Z","description":"A Cornell study published in Nature Sustainability projects US AI data centres could emit 24 to 44 million tonnes of CO2-equivalent annually by 2030,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/5ad645493f56ef70ff83262be07f73eb?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/5ad645493f56ef70ff83262be07f73eb?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The Next Web","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"The Next Web","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tnw","url":"https://daily.dev/sources/tnw"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,sustainability","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The Next Web","item":"https://daily.dev/sources/tnw"},{"@type":"ListItem","position":3,"name":"How many cars’ worth of CO2 will US data centres really emit?"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-many-cars-worth-of-co2-will-us-data-centres-really-emit--nbn4rgjcg#faq","mainEntity":[{"@type":"Question","name":"How much CO2 will US AI data centers actually emit by 2030 according to the Cornell study?","acceptedAnswer":{"@type":"Answer","text":"US AI servers are projected to emit between 24 and 44 million tonnes of CO2-equivalent a year by 2030, equivalent to the emissions of 5 to 10 million cars, not the 24 million cars figure that circulated widely online. The range reflects a spread between a restrained build-out and a more aggressive one, modeled using a hybrid statistical and thermodynamic server efficiency model paired with the US ReEDS grid model. Anyone weighing AI's real environmental cost can follow accurate coverage of data center emissions research on daily.dev."}},{"@type":"Question","name":"Why is the '24 million cars' statistic about AI data center emissions wrong?","acceptedAnswer":{"@type":"Answer","text":"The figure misreads the Cornell study's projection of 24 to 44 million tonnes of CO2-equivalent annually as if it meant 24 million cars, when the researchers themselves translated the emissions into 5 to 10 million cars' worth. The error appears to stem from confusing a tonnage figure with a vehicle count, inflating the popularized number by roughly 2.5 times. Developers tracking AI's infrastructure footprint can verify claims like this by following primary research coverage on daily.dev."}},{"@type":"Question","name":"How much could data center emissions be reduced through best practices like site selection?","acceptedAnswer":{"@type":"Answer","text":"Best-practice interventions in siting and procurement could cut data center emissions by up to 73% and water use by up to 86%, according to the Cornell study's modeling. Locations such as the Midwest, Texas, Montana, Nebraska, and South Dakota are identified as better choices than water-stressed regions like Northern Virginia, and cheap-renewables grid scenarios alone could lower emissions by more than 15%. Engineers weighing sustainable infrastructure choices can track this kind of siting research through daily.dev."}}]} ```