A.I. is just another environmental concern stacked on top of the already endless list
Like a ticking wristwatch, a computer runs with the synchronization of all the necessary components. The computer pulls electricity from the outlet, which is from a generator or power grid, and as the volume of use increases, the computer pulls water from the cooling system to prevent overheating.
Copious amounts of electricity and a slight amount of water are used to run a modern personal computer. More than many of us ever realize.
But then came what we now know as the data center.
Instead of someone’s medium-sized, room-fit computer, it became the size of the average American house, and there are now thousands of them. The amount of space that something like this takes up is so vast that it takes a formation similar to a neighborhood.
The amount of energy required to run these comically large computers every waking hour or every single day is unfathomable, and the resources needed to generate this all are more than one human might be able to use in a year.
The rapid expansion of data centers are nothing more than another problematic environmental issue placed on top of an already daunting and extensive list. Similarities from large data centers and other environmentally concerning infrastructures like coal power plants, oil rigs, or even fast fashion companies, allow for understanding the impacts this will have on the world.
This time, however, the expansion and function of data centers is quickly impacting many American communities on a large scale.
These centers are often located at awkwardly placed plots of land, seemingly having the largest amount of negative impact possible.
Firstly, the river, reservoir, pond, or whichever water resource the centers pull from is simultaneously the community’s, town’s, and people’s water source. The copious amounts of water constantly being used by the data centers are ultimately tainting these vital resources for the people who live around them, who desperately need them.
The tap water becomes brown, dirty, unfiltered, and contaminated; causing it to become unusable and undrinkable. This is because it is pulling so many gallons of water every hour of every day and evaporating the water, without replenishing the source fast enough, that it begins to drain the water completely. Then, it causes a shortage of water, as foretold by the sediment being pulled into people’s water.
There have been multiple reports and videos of this impact happening in many different communities across America on the harsh effects of mass data centers. Water is a precious and necessary resource, which makes the degradation of these sources all the more evil.
Communities that have been there for decades are being torn apart, and people who have lived in these areas their entire lives are now feeling forced to leave. They get no say and no reprimand for what is being done to them.
Approximately five million gallons of water are consumed every day by the largest data centers. Washington, D.C. has a population of around 700,000 people, and, on average, each American uses 82 gallons of water every day.
Using math, assuming that everyone in Washington, D.C. uses some sort of water resource every day, 57 million gallons of water are used every day by Washington, D.C.
That means that a mere five large data centers end up using an entire city’s water use every day.
The difference is these data centers don’t do a public service or provide a necessity to a mass population, yet they still use a magnificent amount of precious resources such as water. This information came from the United States Environmental Protection Agency and the USA Facts organization.
Secondly, the sheer, unrelenting demand for energy coming from the data centers is simply not enough for these places to ideally supply.
And the power grids or energy supplies of the communities are being pushed as far as possible, not for the people who live and make a living there, but for the businessmen who aim to profit millions of dollars, and the tech executives who aim to push the A.I. landscape further and further, without a means to an end.
The people pay for this through higher electricity bills since they must pay more so that the electricity providers can still profit from the accelerated use of gas/coal power plants.
According to an article written by Miguel Yanez-Barnuevo from the Environmental and Energy Study Institute (EESI), an official organization dedicated to the study/documentation of environment and energy, a push for gas and coal power plants to be built and used are direct result of the hasty expansion of data centers.
This directly ties into spiking bills in electricity for neighboring households. Not only do the bills increase because of existing data centers, but also because of proposed, unbuilt data centers, which call for energy infrastructure investment, causing a simultaneous boom.
The EESI’s data on affordable utility states that, with the increased need for interconnection requests, the price increase has burdened the average person.
There is a proposed $29 billion rate increase across the U.S. in 2025. It is a 27% increase in price per kilowatt since 2019, and in the state of Virginia, the price of electricity has increased a whopping 267% over the span of half a decade.
This further disadvantages lower-income households and communities that, on average, spend 20% of their income on energy bills, and higher-income people spend only 3% of their income on energy. (EESI).
According to the EESI, already, one in four households reported having a difficult time paying their bills; with the boom of data centers, this will certainly make things significantly harder.
Lawmakers have been paving the way for A.I. and tech giants to place their data centers, not only in the remoteness of America’s beautiful natural landscapes, but also in the long-living towns, cities, and communities that are home-town’s of so many people.
Big tech companies like Microsoft, Google, Amazon, Meta, and the people running Open A.I. are lobbying increasingly harder and harder to push an agenda upon U.S. lawmakers for more land opportunities for data centers.
According to Human-Centered Artificial Intelligence (HAI) at Stanford University, global private investments in A.I. have increased by 26%. In 2024 alone, investments in A.I. reached $252.3 billion.
According to an article from the World Resources Institute, there are around 4,000 data centers in the United States. The top ten states with the most centers are (from lowest to highest) Florida, Oregon, New York, Arizona, Ohio, Georgia, Illinois, California, Texas, and Virginia. Virginia, taking the cake with an overwhelming 566 data centers, is the country’s number one. Virginia’s land is around 43,000 square miles in total, so that is about one data center every 75 miles.
A.I. has been a thing for longer than most realize, but the public access of it, as well as the insane advancement of it, is a relatively new thing. A safe starting point of the rapid expansion of artificial intelligence at our current level dates back to 2020.
We have seen this pattern in multiple other advancements in technology/infrastructure in the past. Coal, gold, oil, cars, televisions, computers, and arguably, the most large-scale, the telephone and the smartphone.
A large-scale, speedy boom is a natural course for A.I. to take, but what sets it apart from the rest is how enormously negative the effects are in comparison to the likes of the smartphone or even cars.
In retrospect, many advancements made by humans were developed too quickly and too soon. So much so, we have had numerous negative outcomes because of how unready we were, and this is an unfortunate normalcy that we have taken under.
Things like the smartphone became large-scale, so fast, that, no matter how many negatives we find with it, it is so ingrained in modern society, we would find it exceedingly difficult to live without it.
The most palpable fear with A.I. is that we might get it so fast and, so soon, we will reap the negative effects and have no return. Artificial intelligence is something that humanity simply can not risk being hasty with. To ensure a safe future, we must take our time, especially with a tool that can erase so much human progress.