Artificial Intelligence is seen as the savior when it comes to solving complex math problems or helping you finish your homework quickly. AI can definitely make a positive world impact by giving researchers the ability to quickly parse big data sets and create fancy graphs and charts of their work. However, AI isn’t just computer hardware, and it comes at a steep price.
As AI keeps advancing, it will require more data centers to properly function. These data centers run powerful computers that operate non-stop, generating enormous amounts of heat from the calculations they perform every single day. To combat this, companies often rely on liquid cooling—water is the best way to transfer heat from one place to another. The downside is that fresh water is used, and it isn’t just like a few gallons a day: each data center can consume millions of gallons of water in a single year. That is millions of gallons of water that the locals will not be able to utilize to live and survive.
That said, it is important to note that not every AI has the same environmental impact. The majority of water usage comes from large-scale generative AI models that have to run constantly. Additionally, the majority of data center cooling systems use closed-loop cooling systems, which recycle the water rather than permanently consuming it.
What are the actual downsides of AI?
The most visible downside of AI is the rise in the cost of computer parts such as graphics cards and memory. These issues aren’t just limited to gamers or PC enthusiasts. Experts warn that the surge in memory demand for AI models is spilling over into consumer electronics, which means that it could drive up the prices of phones and laptops.
What use does AI have other than homework help?
In the scientific sector, AI is currently assisting scientists and researchers turn heaps of data into visual formats such as graphs and charts. But beyond fancy data visualization, AI is helping researchers notice patterns that we humans fail to see. For example, when presented with an x-ray of a patient, AI has made faster and more accurate diagnostics than physicians. However, AI fails to properly gather the initial patient information, which can affect the result of diagnoses. AI serves as a tool for pattern recognition, but it cannot explore the unknown without proper context and training.
In the end, AI can contribute to society in a meaningful way, but there is always a cost.
