Problem
Research questions and friction points this paper is trying to address.
AI data centers
water use
water scarcity
feedback loop
water consumption
AI Data Centers and the Water Use Feedback Loop
🤖 AI Summary
This study addresses the bidirectional feedback between AI data center water consumption and regional water scarcity, a relationship lacking a systematic quantitative framework. The work formalizes this feedback mechanism for the first time, developing a coupled water–AI model and introducing a “Water Impact Index” to quantify the pressure exerted by data centers on local water supply systems at the community scale. Integrating a systematic literature review, indicator-based modeling, and empirical analysis across ten U.S. locations, the research reveals that data center water burdens vary by up to three orders of magnitude, accounting for 0.2% to 134% of local water supply capacity. These findings underscore the pronounced spatial heterogeneity of potential water resource risks associated with AI infrastructure.
📝 Abstract
AI data centres consume water for cooling, water scarcity constrains siting, and AI tools can improve water system efficiency. These dynamics are studied separately yet form a feedback loop. This review formalises the Water and AI Feedback Loop, introduces the Water Consumption Impact index to quantify community-scale utility burden, and demonstrates across ten US sites that burden spans three orders of magnitude, from 0.2% to 134% of host capacity.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Water and AI Feedback Loop
Water Consumption Impact index
AI data centers
water scarcity
community-scale utility burden
🔎 Similar Papers
No similar papers found.
