The rapid expansion of artificial intelligence is reshaping industries, but at what environmental cost? A new study by Megan Rand Wheeler, an information science researcher at the University of North Texas, Denton, reveals that AI’s green potential is not as straightforward as it seems. Published in *Clean Technologies* (translated from *Tecnologías Limpias*), Wheeler’s research introduces a decision-making framework to help clean technology practitioners navigate AI’s environmental paradox.
Wheeler’s team analyzed 73 peer-reviewed studies from 2021 to 2025, uncovering a critical pattern: AI’s environmental impact follows an S-curve, where early benefits often give way to rebound effects before long-term optimization. “We found that regions with renewable energy and abundant water see net benefits within two to three years,” Wheeler explains. “But in areas reliant on fossil fuels and facing water scarcity, AI deployment may never offset its environmental costs.”
The study highlights staggering figures—global AI infrastructure investment is nearing $500 billion annually, while data center electricity consumption could surpass 1,000 terawatt-hours by 2030. Wheeler’s framework categorizes AI’s impacts across energy optimization, resource conservation, and supply chain extraction, emphasizing that strategic interventions—such as renewable energy integration and advanced cooling technologies—are essential to mitigate harm.
For energy sector leaders, this research underscores a pressing commercial reality: AI’s long-term sustainability hinges on infrastructure preparedness. Wheeler’s work suggests that without proactive measures, the environmental trade-offs of AI could outweigh its benefits, reshaping investment strategies and regulatory priorities in the clean tech space.