Exactly how is AI affecting our lives and the environment?

     "Education is the most powerful weapon which you can use to change the world.” ―Nelson Mandela

     You may have heard of the familiar, yet daunting phrase “AI consumes as much as a 16-oz water bottle for every 5-50 prompts”. But what exactly could this statement mean; and how does it reveal the raw, detrimental impacts of AI usage upon millions of people? Although advancements in technology were originally constructed for the benefit of society, the rise in vast utilization of AI is not only in opposition to its initial objectives, but have also contributed to a chain of several national conflicts. This ‘ripple effect’ includes the practical procedures utilized to continue the recession of accessibility for vital resources, in addition to the rise and circulation of environmental racism.

     To begin with: what even are data centers, anyway?

     Data centers are non-renewable, physical facilities made up of several interconnected computer servers. Their main functions include storing, processing, and distributing large amounts of data. Not only do data centers house every photo, text, and document from a specific organization, they also act as the primary vessel generating AI prompts. Due to the excessive energy needed to operate these functions, data centers require water in order to cool their servers and prevent damage from overheating. In particular, AI uses machine learning for their generative responses which requires large amounts of computational power. This means that significantly more heat is generated and requires more water for cooling to maintain performance, leading to the exhaustion of several water sources.

     At the same time, these water sources that are prone to being consumed by data centers are actually the same vital freshwater that is necessary for human survival. According to the Bureau of Reclamation California-Great Basin, only 3% of Earth’s water is freshwater, and just 0.5% of all water is accessible and safe for human consumption. Alongside this, an article in the Washington Post states that a single data center can each “drink” up to 5 million gallons per day, or about 1.8 billion annually. This is equivalent to the water usage from a town of 10,000 to 50,000 people. Let us reflect on the real-life situations of how citizens that inhabit rural spaces around the world, who already lack accessible clean water, are being directly affected by this cause, especially with data centers being so close to their space of living. Moreover, according to a research study at UCR,  approximately 80% of the water (typically freshwater) withdrawn by data centers evaporates, with the remaining water either discharged to municipal wastewater facilities or is otherwise taken out of immediate human usage, effectively removing it from the local water cycle and disrupting our climate. As growth in the technology industry increases reliance on these cloud computing data facilities, the demand for construction of machine learning applications result in further pressure on local water resources and decreases society’s accessibility to these resources.

     Given a basic understanding behind the practical procedure of data centers, we can recognize its application upon society and its negative effects. Environmental racism, defined by the World Economic Forum, is stated as “a form of systemic racism whereby communities of colour are disproportionately burdened with health hazards through policies and practices that force them to live in proximity to sources of toxic waste”. This discriminatory practice has increased alongside the construction of data centers, and a real-life example on the effects of environmental racism are the xAI and Grok data center facilities in Memphis, Tennessee.

     In the report of the non-profit newsroom More Perfect Union, Alexis Humphrey—a Memphis resident—grew up and currently lives just over a mile from the xAI facilities. She and her family all have chronic lung illnesses including asthma and bronchitis. Her grandfather passed away last year due to a respiratory illness called Chronic Obstructive Pulmonary Disease (COPD) that is mainly caused by smoking. Alexis’s grandfather did not smoke cigarettes, but was visiting the hospital every three weeks or so to receive breathing treatments due to the severity of the air pollution. It is no coincidence that alongside Alexis’s family, the neighbors are also developing chronic lung illnesses that have been unprecedented before the construction of these facilities. Yet despite such tragedies that are evidently occurring to citizens of the area, the Environmental Protective Agency (EPA) has stayed silent in this affair and has taken no action against the project despite its violations of the Clean Air Act. With Elon Musk working alongside Donald Trump in office, he has the control to help limit budgets for the EPA and has even eliminated all environmental justice offices which help heavily polluted communities. By cancelling billions of dollars in funding for addressing pollution, it is clear where the government stands when they are faced with a decision concerning the welfare of people and their desire for power.

     Artificial intelligence has remarkably increased its water footprint in the years past, and it is time to acknowledge its profound environmental and societal impact unitedly. In spite of the fact that the heavy comparison of using AI is tantamount to  wasting a water bottle may seem like another ‘trendy’ and ‘fear-mongering’ idiom, it is important to acknowledge the truth and research that supports such a claim.

In a world where you can’t control the actions of those in power, it becomes a crucial priority for education to forefront an understanding of our  world in order to foster a better future.

Sources:

1. Bureau of Reclamation. “Water Facts - Worldwide Water Supply.” Bureau of Reclamation, 4 Nov. 2020, www.usbr.gov/mp/arwec/water-facts-ww-water-sup.html. ‌

2. Osaka, Shannon. “A New Front in the Water Wars: Your Internet Use.” Washington Post, 4 May 2023, www.washingtonpost.com/climate-environment/2023/04/25/data-centers-drought-water-use/. ‌

3. Li, Pengfei, et al. “Making AI Less “Thirsty.”” Communications of the ACM, vol. 68, no. 7, 27 June 2025, pp. 54–61, https://doi.org/10.1145/3724499. Accessed 9 July 2025. ‌

4. Beech, Peter. “What Is Environmental Racism and How Can We Fight It?” World Economic Forum, 31 July 2020, www.weforum.org/stories/2020/07/what-is-environmental-racism-pollution-covid-systemic/.

5. More Perfect Union. “We Went to the Town Elon Musk Is Poisoning.” YouTube, 30 May 2025, www.youtube.com/watch?v=3VJT2JeDCyw. ‌ ‌