As artificial intelligence (AI) becomes part of everyday life, it’s designed to feel insignificant. Yet, behind every effortless prompt lies a significant environmental cost.
AI programs are powered by massive data centers, warehouses of servers that process enormous amounts of information. These facilities consume electricity at rates that the implementation of renewable energy sources into the electric grid can’t keep up with. This leaves data centers heavily reliant on fossil fuels, tying AI’s rapid growth to the ever-worsening climate crisis.
Elsa A. Olivetti, a professor in the Department of Materials Science and Engineering at MIT, explains how it’s harde for users to understand the full impact of their searches.
“When we think about the environmental impact of generative AI, it is not just the electricity you consume when you plug the computer in. There are much broader consequences that go out to a system level and persist based on actions that we take,” Olivetti said.
Additionally, the scale of AI’s energy demands is staggering. According to the Organization for Economic Cooperation and Development, electricity use from data centers is expected to reach 1,050 terawatt-hours by 2026. That would make them the fifth-largest consumer of electricity worldwide when stacked up against other nations.
These problems extend beyond electricity. Data centers generate extreme heat to process all of the information from different servers, so they use massive amounts of water to cool down. The Environmental and Energy Study Institute reports that larger data centers can consume up to 5 million gallons of water a day, equivalent to the daily needs of 50,000 people.
AI tools also require a lot more energy than traditional search engines. University of Maine’s professor of new media Jon Ipollito developed an app to compare the amount of energy AI servers draw and found that a simple AI prompt requires 23 times more electricity than a Google search, while more complex prompts consume 210 times as much.
Despite these impacts, most users remain unaware of AI’s environmental footprint. The easy-to-use design of popular systems encourages heavy use and keeps the resources required to sustain them a mystery to the average user.
The electricity demand also drives up carbon emissions. According to the Nature Conservancy, ChatGPT emits 8.4 tons of carbon dioxide every year. The U.S. Government Accountability Office (GAO) has suggested that the federal government could encourage developers to build more efficient data centers to reduce energy and water consumption. However, the GAO fears that the industry would resist this change in such a fast-moving industry, especially when the long-term consequences of AI are still not fully understood.
Without stricter regulations and more investment in renewable energy, AI could become one of the most significant new drivers of climate change. How quickly governments and tech companies act could decide if AI is remembered as a tool of progress or a horror story for the human race.
