The development of artificial intelligence (AI) has caused major changes in nearly every aspect of the world. As AI gains stronger influence and becomes widespread, the energy needed to build, train and run these systems grows fast, as well as their negative environmental impact.
Training an advanced AI model requires an enormous amount of computing power. Machine learning algorithms, especially ones like ChatGPT or image generators, require huge datasets and complex calculations that can take weeks or even months to complete. This process usually happens on massive server farms powered by electricity, multiple of which still come from burning fossil fuels. Studies show that training a single AI model can emit as much carbon dioxide as five cars would over their entire lifetime. The water that these AI models use generates a substantial amount of water vapor, which is classified as a greenhouse gas, contributing to the rising global temperatures.
“Although I really enjoy ChatGPT, I definitely think that there are more cons than pros. It makes people lazy and takes over people’s jobs, which has been a problem since machinery was first introduced into the workplace. Also, its contribution to climate change is another turn-off because there is no point in having anything if it is just going to kill us eventually,” freshman Monica Lamb said.
Once trained, AI models do not prove energy-free. Every time a person asks an AI assistant a question, streams AI-generated content or relies on smart services, servers activate to process the request. When millions — or even billions — of people use AI tools daily, the energy use adds up quickly. Data centers — the giant warehouses full of servers that power AI — already account for nearly 1-2% of global electricity use, and that number projects extreme growth with the increased usage of AI.
Building AI does not just consume electricity — it also requires hardware such as graphics processing units (GPUs) and specialized chips. Manufacturing these components involves mining for rare earth metals, using water-intensive processes and producing comprehensive amounts of waste. The environmental impact of creating and disposing of AI hardware serves as a major contributor to pollution and resource depletion.
“AI is a really controversial topic because it impacts so many different sectors of the world and the economy. I think that while I enjoy using it, I know that I shouldn’t because of the amount of water used to cool the overheating systems. I wish there was a more sustainable way to use these systems,” magnet junior Kassidy Brown said.
As AI advances, a substantial number of industries adopt it — from healthcare to entertainment to finance. This growing demand drives the expansion of data centers, usually located where energy costs less, not where it comes cleanest. Unless data centers switch to renewable energy sources, the AI boom could significantly increase global carbon emissions over the next decade.
While AI acts as a convenient resource for students, workers and other individuals, the environment suffers at the expense of efficiency. Considering cleaner energy sources as fuel to maintain the widespread data centers used for AI would significantly help reduce the stress on the Earth. AI offers remarkable potential to solve global challenges, but remembering its environmental impact remains crucial.
