The first data centers were initially built back in the 1950s and were very large encompassing entire rooms that had to be built with their environments designed to support the needs of the very large beginnings of technology and later computers needed to process the information. This was how the first few data centers tended to look and were—compared to today’s standard, pretty rudimentary.

Even though AI data centers have been around for 70 years they have only recently begun to rise into mainstream media. The introduction of AI tools has begun: the most well known one being ChatGPT and it is owned by OpenAI.

OpenAI owns only one major data center currently but has announced plans for up to five more being built throughout the country and this raises the question—how will these data centers affect the environment and communities around the areas they are built?

The likely most immediate affect people living around a newly built data center is an increase to people energy bills as the energy it takes to power a full size data center is colossal—around 100 times the needed power for a full scale office building and can increase energy bills by 80% all the way up to 267% comparing to the average from five years ago.

The increased energy bills caused by the AI data centers can cause surrounding businesses to have to shut down or relocate due to increased operating costs and this can be devastating for local economies.

With the increase in electricity bills the AI data centers puts intense strain on the power plants that power the towns and cities they are built near and can cause entire blackouts due to the heavy additional strain they put onto the power grids.

Increasing electricity bills and causing blackouts are not the only issue with AI data centers as they don’t only require lots of power, they also require lots of water to cool everything down.

As the computers that are used produce ridiculous amounts of heat causing the severe need for water this has caused local water bills to rise as well as AI data centers can use up to 5 million gallons of water per day. That’s equivalent of 10 thousand to 50 thousand people.

AI data centers also affect water treatment facilities as the water they don’t burn off is often sent back into the city water system. The water has to be treated again due to anti-corrosives, silicone, and cleaners in the pipes/computer chips and from off data centers, and most water treatment facilities are not built for the amount of waste water data centers produce. The influx in water use can also cause the facilities to be more damaged and can cause untreated sewage to go into the environment with overworked facilities.

AI data centers can also take up ridiculous amounts of space and some of the larger ones spanning 17 football fields. These can exist in nearby suburban neighborhoods and cities and those areas could be used for more residential areas or even use for more economic reasons like businesses. This wouldn’t be nearly as much of an issue if the data centers provided more jobs however they don’t provide much for local communities and can cause issues for said communities like the ones stated earlier.

Since these huge data centers require so much energy to power them and water to cool them building in rural areas can be expensive. Building them in areas with neighborhoods or communities can be much cheaper but with the cost of the communities paying more for water and electricity this could grow into a large issue for businesses that require electricity and water.

Building these data centers doesn’t just affect the people in the city they also affect the surrounding ecosystems. Other than polluting the water, they can also destroy miles of habitats in their construction alone. They also produce a consistent humming which can cause the animals to be stressed or prevent predators from finding prey as the constant noise can impede their hearing and normal hunting tendencies.

These data centers use so much energy and about 56% to 60% of the energy they use is entirely from fossil fuels. This can cause them to affect not only the local environments but the entire world, as the burning of fossil fuels for energy is the leading cause of global warming.

There is a concept that has been spread throughout the internet and it is the “AI bubble.” In recent years many companies have dumped billions of dollars into the betterment of AI. Many hope that this “AI bubble” pops because all of the issues mentioned here and more. The building of these data centers also contribute to issues like graphics processing unit (GPUs) and random access memory (RAM) becoming incredibly expensive. As these data centers purchase thousands to process everything they need to, the “bubble” popping would slow down the building of these data centers exponentially.

Companies will stop funding these AI projects and some may shut down, making it so electric and water bills aren’t so expensive. As they are in the areas around the data centers, this would also likely cause an influx of RAM and GPUs to go onto the personal computer (PC) market. Even with the benefits that would likely come with the shutdown of all these data centers the companies that fund them all ask the same question—after everything are these data centers really worth the sacrifice of the local communities and ecosystems?