top of page
Search

AI Water Consumption & What You Can Do

  • Writer: Asiya Siddiqui
    Asiya Siddiqui
  • Jan 14
  • 3 min read

By: Asiya Siddiqui

January 14, 2025


Who doesn’t love AI? In today’s world, artificial intelligence has become an indispensable tool, assisting in everything from free math tutoring and comprehensive document summaries to even generating grocery lists. Over the last quarter century, we have grown increasingly reliant on machines for support. As a result, AI has seamlessly woven itself into nearly every aspect of our lives—so much so that imagining a world without it feels impossible.

AI data center.


But where does AI receive its intelligence? The answer lies in data centers. A data center is a physical facility that stores computing infrastructure (e.g., servers, network equipment) necessary for IT systems to operate. These installations power the AI you use every day, such as Open AI’s Chat GBT or Google’s Gemini. Unfortunately, this industry, valued at over 600 billion dollars in 2024, is costing us far more than money. It is robbing us of one of our most crucial resources: water.


When researching for this article, I forgot to add “-AI” to my Google search, inadvertently activating Gemini, Google’s AI-generated response system. Ironically, Gemini itself revealed the water-intensive reality behind its operations:

  1. Cooling Systems: AI models require immense computing power, translating to high energy needs in data centers. Data centers rely heavily on cooling systems to prevent overheating, resulting in heavy water consumption. 

  2. Electricity Generation: Much of the electricity powering data centers comes from thermoelectric or hydroelectric plants, both of which demand large amounts of water. 

  3. AI Supply Chains: Producing components like microchips also contributes to water depletion. For example, manufacturing a single microchip requires nearly 3 gallons of water to cool machinery and ensure contaminant-free wafer sheets.


Only 3% of the Earth’s water is fresh, and a mere 0.5% of that is accessible—the rest is trapped in glaciers and polar ice caps. Meanwhile, global water demand continues to grow. In 2021, the UN Water Program reported that over 720 million people lived in countries experiencing high water stress (5). 


Yet, tech companies exacerbate this crisis by strategically locating data centers in lower-income nations to take advantage of cheaper energy and real estate. For instance, Google’s hyper-scale data centers averaged 550,000 gallons (2.1 million liters) of water consumption per day over the past year (4). How long do we have before the rest of our water supply goes dry?


What You Can Do

There are meaningful steps you can take to make a difference:

  1. Support Sustainable AI Practices: Encourage companies and developers to adopt energy-efficient AI practices by supporting those that prioritize green technology. You can push policymakers to implement regulations and incentives that promote environmentally friendly AI practices.

  2. Reduce Your AI Usage: As I previously mentioned, you can add “-AI” to the end of your Google search to avoid Gemini or turn off AI-generated responses altogether. Furthermore, avoid frivolous uses (such as generating grocery lists or drafting emails) that drive unnecessary demand for processing power. 

  3. Raise Awareness: Mention these statistics at dinner and share factual news articles on social media. Collective change in how we interact with AI can lead to a more sustainable future.


Although AI has proven useful in reducing water waste through smart water management, its relentless water consumption threatens our global water security. As a society deeply dependent on technology, we must implement environmental safeguards both on a personal and governmental level.



Support our cause!


References:

  1. Artificial intelligence (ai) market size to reach usd 3,680.47 bn by 2034. (n.d.). Precedence Research. Retrieved January 14, 2025, from https://www.precedenceresearch.com/artificial-intelligence-market#:~:text=The%20global%20artificial%20intelligence%20

  2. Collier, A. (2024, August 22). Artificial intelligence is using a ton of water. here's how to be more resourceful. Veolia. Retrieved January 14, 2025, from https://www.watertechnologies.com/blog/artificial-intelligence-using-ton-water-heres-how-be-more-resourceful

  3. Danelski, D. (2023, April 28). Ai programs consume large volumes of scarce water. UC Riverside News. Retrieved January 14, 2025, from https://news.ucr.edu/articles/2023/04/28/ai-programs-consume-large-volumes-scarce-water#:~:text=Google's%20data%20centers%20in%20the%20U.S.%20alone,by%20the%20journal%20arXiv%20as%20a%20preprint

  4. Pinheiro Privette, A. (2024, October 11). Ai's challenging waters. The Grainger College of Engineering Civil & Environmental Engineering. Retrieved January 14, 2025, from https://cee.illinois.edu/news/AIs-Challenging-Waters#:~:text=Data%20centers%20are%20growing%20like%20weeds%2C%20and%20they%20are%20thirsty%26text=Water%20consumption%20in%20data%20centers,day%20over%20the%20past%20year

  5. Un World Water Development Report 2024. (2024, March 19). UN-Water. Retrieved January 14, 2025, from https://www.unwater.org/publications/un-world-water-development-report-2024

  6. Water facts - worldwide water supply. (2020, November 4). Bureau of Reclamation. Retrieved January 14, 2025, from https://www.usbr.gov/mp/arwec/water-facts-ww-water-sup.html#:~:text=3%25%20of%20the%20earth's%20water,water%20is%20available%20fresh%20water

  7. Zhang, M. (2024, January 17). Data center water usage: A comprehensive guide. Dgtl Infra. Retrieved January 14, 2025, from https://dgtlinfra.com/data-center-water-usage/

 
 
 

Comments

bottom of page