Opinion: Learning about AI comes at a cost. Georgia feels it, too. -

An opinion piece by Mercer University student Kaylee Grace Buchanan '29 argues that the seemingly invisible nature of artificial intelligence conceals significant environmental costs, particularly in Georgia. Buchanan highlights that AI systems rely on massive data centers, which demand constant electricity, consume vast amounts of water for cooling, and are built with hardware derived from environmentally destructive mining. Georgia has emerged as a rapidly expanding hub for data infrastructure in the Southeast, a growth driven by relatively low energy costs and available land. These data centers operate continuously, consuming as much electricity as tens of thousands of homes each, with global AI-driven data center electricity demand projected to quadruple by 2030. Given Georgia's reliance on fossil fuels, this expansion directly links AI's growth to rising greenhouse gas emissions, with large language models alone producing substantial carbon footprints. Beyond energy, data centers are water-intensive, using millions of gallons daily for cooling, a critical concern during drought seasons in the Southeast. Buchanan also points out that the minerals required for AI hardware, like lithium and cobalt, are often sourced under ethically troubling conditions. She suggests that efficiency gains in AI technology do not necessarily reduce overall consumption but can instead encourage more use, overlooking the broader environmental footprint. Buchanan urges universities, including Mercer, to demand transparency from tech companies, prioritize vendors using renewable energy, and integrate environmental accountability into technology policies. She also calls on Georgia lawmakers to require clearer reporting on data center energy and water use, invest in renewable infrastructure, and ensure technological growth does not harm local communities. The article concludes by emphasizing that ignoring the true cost of AI is no longer an option, particularly for residents of a state increasingly tied to digital infrastructure.