AI Series: Balancing Data Infrastructure Resource Demands

Previous episodes in our AI series have focused on the evolving features of artificial intelligence itself: its potential to democratize education and to improve city planning and weather forecasting. In this final installment, we examine its costs: the accelerating resource demands of AI and other data-intensive technologies.  

Maya Chari, this year’s Ten Across + APM Research Lab data journalism fellow, recently investigated the true water and energy costs associated with data center facilities in the Phoenix metro area— now on track to become the second largest market in the U.S. Though granular industrial data can be difficult to come by, Maya located a report submitted by Microsoft to City of Goodyear officials, stating that one of their proposed data centers would use as much potable water each year as 670 homes. Amplified across the 140 other data centers currently dotting the state of Arizona alone, the scale of such consumption becomes clearer.  

As data centers rapidly multiply in response to market demand around the world—often preferring arid places like the water-stressed U.S. Southwest—critical questions are pressed about whether and how such development can be sustained.    

In back-to-back conversations in this episode, we’ll hear from experts involved in managing and reducing the impact of the physical infrastructure behind our digitized lives. Bobby Olsen, chief planning, strategy, and sustainability executive at the Arizona electric and water utility Salt River Project, describes planning to meet staggering levels of projected energy demand. And Dr. Kerri Hickenbottom, principal investigator at University of Arizona’s Hickenbottom Environmental Research Lab, discusses working in concert with the public and private sectors to improve water reuse strategies and overall efficiency of data center operations.  

To support our I-10 neighbors' disaster recovery in Central Texas this week:  Kerr County Flood Relief Fund  

Related articles and resources:  

“At Amazon’s Biggest Data Center, Everything is Supersized for A.I.” (The New York Times, June 2025)  

“Are Data Centers Depleting the Southwest’s Water and Energy Resources?” (American Public Media Research Lab, February 2025)

 “Thirst for power and water, AI-crunching data centers sprout across the West” (Stanford University, April 2025)  

“’I can’t drink the water’ —life next to a US data center” (BBC, July 2025)  

“Meta is