Are AI Power Plants Destroying the Planet?
As artificial intelligence (AI) usage becomes normalized in everyday life, the demand for data storage increases, leading corporations to speed up the production of data centers in order to keep up with the demand. This has led to concerns about the amount and type of energy that these plants use and the effects they can have on the environment.
Data centers are large groups of networked computer servers that are used for storage of large amounts of information. AI takes information from these data centers in order to generate responses to the user. Dr. Mark O’Gorman, professor of political science at Maryville college, explained the impact of these centers.
“AI software can basically tap into that data center, grab [information] off the cloud and within a second the AI can churn it out,” said O’Gorman. As AI usage accelerates, so does data center production.
“More and more data centers are being created because AI organizations want to have these data centers created, meaning there’s going to be many more of them and the increased demand for energy is going to be bigger,” said O’Gorman.
Concerns arise from the source of energy that the data centers use. According to the International Energy Agency, a ChatGPT search uses 10 times more energy than a Google search. Not only is more energy needed to search the cloud for information through AI software, like ChatGPT, but even a search on Google now utilizes AI in the top generated response.
Often, AI data centers are powered by energy sources coming from the extraction of natural resources, leading to further issues of carbon emissions.
Jan Taylor, professor of Languages and Literature at Maryville College, said that another concern regarding environmental effects of data centers is their immense usage of water.
“Data centers supporting AI operations consume much water for cooling purposes, which can strain water resources,” said Taylor.
Electronic companies launch new versions of their products almost every year, with a prime example being smartphones. Most people change their phone every two to three years, creating a new source of waste and contamination of the environment.
O’Gorman explained that he has had six phones and seven computers.
“If I’m the average user of half a dozen cell phones and half a dozen computers, times 300 million Americans, that’s a lot of waste” he said. “It’s the electronic waste that has been the fastest growing category of waste in the last 25 years.”
Alongside personal technology development, these data centers will need to upgrade as well in order to keep up with new demands of a growing industry.
“The data centers of 2025 will have to significantly upgrade in 10 years to take on AI,” said O’Gorman. “So what happens with the old stuff? You chuck it away– more waste–and that’s the concern,” he added.
However, this is not a hopeless situation. With an acceleration of the use of AI in recent years, there is the hope that AI could be used to discover new methods of environmental sustainability or protection.
“While AI consumes significant resources, it also is used for environmental protection activities like monitoring climate change and conservation efforts, optimizing energy consumption, and developing sustainable agriculture,” said Taylor. “As with most technologies, the net impact depends on a balancing of factors.”
O’Gorman explained why there is a delay in environmental concerns becoming a focus when it comes to new technology
“It’s always a lag,” he said. “You usually make the technology and later you’re like, ‘oh, we weren’t thinking about that when we were making that.’ And then the concern becomes, ‘how do we manage that?’”
AI brings new opportunities, but has significant impacts on our environment that are only gaining public awareness now. The last few years have revolutionized artificial intelligence while also creating new environmental issues to be aware of.