{"id":187843,"date":"2025-11-30T10:38:06","date_gmt":"2025-11-30T10:38:06","guid":{"rendered":"https:\/\/teknomers.com\/en\/data-centers-use-more-water-than-we-thought-but-a-book-is-to-blame\/"},"modified":"2025-11-30T10:38:07","modified_gmt":"2025-11-30T10:38:07","slug":"data-centers-use-more-water-than-we-thought-but-a-book-is-to-blame","status":"publish","type":"post","link":"https:\/\/teknomers.com\/en\/data-centers-use-more-water-than-we-thought-but-a-book-is-to-blame\/","title":{"rendered":"Data Centers Use More Water Than We Thought, but a Book Is to Blame"},"content":{"rendered":"\n
\n

The Misconception about Data Center Water Consumption<\/h2>\n

Critics of the AI boom often highlight its environmental impact, notably the water consumption required to cool servers. This issue has gained attention, yet recent evidence suggests that the problem may not be as severe as initially portrayed, stemming from a significant miscalculation in a popular book.<\/p>\n

The Book That Started It All<\/h3>\n

The book in question is ‘Empire of AI’ by Karen Hao. In her analysis, Hao interviews numerous former employees and insiders of OpenAI, shedding light on what she describes as the excessive water consumption of AI systems. She controversially claimed that a data center could consume 1,000 times more water than a city of 88,000 residents.<\/p>\n

\n
\n
\n
<\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n

Criticism and Clarification<\/h3>\n

According to Andy Masley, who addressed this issue in his newsletter, ‘The Weird Turn Pro’, the real numbers indicate that the data center would consume only around 22% of what the city uses, equivalent to 3% of the entire municipal system. Masley also critiqued Hao\u2019s work for confusing water extraction\u2014temporary withdrawals that are returned to the network\u2014with actual water consumption.<\/p>\n

Understanding the Calculation Error<\/h3>\n

Hao herself responded to Masley’s findings, noting a communication mishap with the Municipal Drinking Water and Sewage Service of Chile (SMAPA). She had requested water consumption data for certain towns but did not receive the information in the expected format. It appears that the figures she cited were likely in cubic meters instead of liters, leading to a massive discrepancy in her conclusions. She is currently seeking clarification from SMAPA on the actual figures.<\/p>\n

Recent Estimates on AI Water Consumption<\/h3>\n

The question of how much water AI truly consumes has sparked significant interest. A 2024 study by the Washington Post suggested that generating a 100-word text with ChatGPT may use around 519 milliliters of water. This calculation takes into account the total annual consumption of data centers and the cooling methods employed.<\/p>\n

Company Transparency Issues<\/h3>\n

AI companies tend to maintain a level of opacity when it comes to disclosing their water and energy consumption. While major tech firms present total annual consumption data in their sustainability reports, the precise details for individual searches remain unclear. Google stands out as the only company to publish detailed water and energy usage for its AI services, claiming that each Gemini query consumes a mere 0.26 milliliters, roughly equivalent to five drops.<\/p>\n

\n
\n
\n
\n \"Investing\n <\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n

The Ongoing Water Controversy<\/h3>\n

While these revelations may appear reassuring, they do not negate the water concerns related to AI. For instance, the Cerrillos data center, which Hao referenced, was halted by the Chilean justice system due to its anticipated negative climate impact amidst an existing drought. This illustrates that data centers indeed require significant water resources, prompting innovative cooling methods such as submerging data centers in ocean water.<\/p>\n

Beyond Water: The Bigger Challenge<\/h3>\n

Water is just one facet of the challenges faced by data centers; energy consumption poses an even larger dilemma. By 2024, data centers in the U.S. accounted for approximately 4% of total electricity consumption, with some local electricity bills spiking by as much as 267% in recent years. Major tech firms are voicing concerns over energy shortages, contemplating solutions that range from nuclear power plants to potentially relocating data centers into space.<\/p>\n

The complex interplay of water and energy consumption highlights the urgent need for transparent reporting and innovative solutions in the evolving landscape of AI technology.<\/p>\n<\/div>\n


\n
General News – 2<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

The Misconception about Data Center Water Consumption Critics of the AI boom often highlight its environmental impact, notably the water consumption required to cool servers. This issue has gained attention, yet recent evidence suggests that the problem may not be as severe as initially portrayed, stemming from a significant miscalculation in a popular book. The […]<\/p>\n","protected":false},"author":1,"featured_media":187844,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[36399],"tags":[542,2107,325,4898,1813,896],"class_list":["post-187843","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-technology","tag-blame","tag-book","tag-centers","tag-data","tag-thought","tag-water"],"_links":{"self":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/187843","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/comments?post=187843"}],"version-history":[{"count":0,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/posts\/187843\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/media\/187844"}],"wp:attachment":[{"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/media?parent=187843"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/categories?post=187843"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/teknomers.com\/en\/wp-json\/wp\/v2\/tags?post=187843"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}