Is It Ethical to Allocate So Much Water to AI Data Centers? Water Scarcity and Allocation in the Southwest

Abstract

This study examines the relationship between everyday use of artificial intelligence (AI), indirect water consumption, and the ethical implications of allocating water resources to AI infrastructure in water-scarce regions. In the southwestern United States, where water scarcity is a growing concern, increasing reliance on AI and data centers raises questions about how limited resources are distributed and who bears the environmental cost. While prior research has explored the water demands of AI infrastructure, less is known about how individual AI use contributes to this demand and how people perceive the fairness of water allocation to these systems. This study used a mixed-methods approach, combining quantitative estimates of AI-related water consumption among college students with survey responses about perceptions of fairness, sustainability, and water allocation priorities. The results suggest that while individual AI use consumes relatively small amounts of water, the cumulative demand raises ethical concerns about prioritizing technological infrastructure over community needs, especially in areas that already experience water shortages. These findings highlight the importance of considering ethical and sustainability factors when planning AI infrastructure and managing water resources in arid regions. This study is significant because it translates everyday AI usage into estimated indirect water consumption and combines these findings with perceptions of water allocation to examine the environmental and ethical implications of AI infrastructure in arid regions.