AI Data Centers and the Water Use Feedback Loop
For policymakers and data center operators, it quantifies the community-scale water burden of AI data centers, highlighting a critical sustainability feedback loop.
The paper formalizes the Water and AI Feedback Loop, showing that AI data centers' water consumption for cooling can consume up to 134% of host utility capacity across ten US sites, with burden spanning three orders of magnitude.
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.