Americans Wary of AI Data Centers Near Homes
A recent survey by Gallup has revealed that a significant number of Americans are against the idea of building artificial intelligence (AI) data centers close to their homes, even more so than their opposition to nuclear power plants. This reflects growing anxieties about the effects of AI technology on the environment and local communities.
In the survey, about 71% of U.S. adults expressed disapproval of AI data centers in their neighborhoods, with nearly half of those surveyed feeling very strongly against them. In contrast, 53% opposed having a nuclear power facility nearby. Gallup noted that disapproval of nuclear plants has remained consistently high since it started polling on the topic in 2001.
As AI companies ramp up their infrastructure to meet rising demands for computing power, many communities are voicing their concerns. Issues such as excessive water and electricity usage, increased traffic, and noise pollution have been highlighted by local residents.
Community Concerns About Resources and Quality of Life
In another part of Gallup’s survey, many respondents against data centers cited worries about the pressure these facilities could place on water and energy resources. About half of those opposing the data centers mentioned potential problems like losing farmland, while 22% expressed concerns over quality of life, such as more traffic and higher utility bills. Noise pollution was also a common complaint.
Wannie Park, CEO of PADO AI, suggests that much of the public’s resistance stems from a lack of understanding of how data centers function and their economic benefits. He remarked on the miscommunication surrounding these facilities, saying, “I think there’s just a lack of education.” Park believes that better outreach and information could change public perceptions.
Growing Demand for AI Infrastructure
As AI usage grows, so does the impact of data centers on resources. Currently, U.S. data centers account for about 4.4% of the nation’s electricity consumption, a significant increase from 1.9% in 2018. This figure is predicted to climb to 12% by 2028, with the overall demand for energy from data centers expected to rise from 80 gigawatts in 2025 to 150 gigawatts by 2028.
Moreover, studies suggest that AI could be responsible for emitting between 24 million and 44 million metric tons of CO₂ each year by 2030. In addition, data centers are expected to consume approximately 17 billion gallons of water in 2023—three times more than in 2014.
Despite the pushback, Park feels that bans on new projects may only relocate these developments rather than stop them altogether. He noted, “If you want to shut us down here, we’re gonna go somewhere else,” reflecting the ongoing demand for AI infrastructure. Even forecasting new projects shows that the need for computing power is so strong that facilities are already booked three years in advance.
As communities navigate the complexities of AI technology, conversations about its impact on daily life will likely continue to evolve.
