Pathways for Sustainable AI Servers in the USA
This resource has been reviewed by SubjectToClimate's climate scientists and verified for scientific accuracy and up-to-date information. Learn about our review process →
Provided by: Nature|Published on: April 28, 2026
Scientific ReportsScientist Reviewed
1112AP
Scientist Reviewed This resource has been reviewed by SubjectToClimate's climate scientists and verified for scientific accuracy and up-to-date information. Learn about our review process →
Synopsis
- This report explains the impacts of AI data centers on water, energy use, and climate change, and makes suggestions for minimizing these impacts.
- Students will learn about data limitations due to unreliable reporting, the influence of renewable energy deployment, and how the location of a data center can have an impact on its environmental footprint.
Subjects: Engineering
Authors: Tianqi Xiao, Francesco Fuso Nerini, H. Damon Matthews, Massimo Tavoni, and Fengqi You
Region: United States
Languages: English
Teaching Materials
Positives
- This report considers many factors that influence the environmental impacts of AI data centers.
- The many graphs, maps, and data visualizations support students in understanding the written content.
Prerequisites
- This report is best suited for students who have some prior knowledge of data centers' energy and water usage, climate change and its impacts, and water scarcity concerns.
Differentiation & Implementation
- After reading the article, students can use the Greenhouse Gas Equivalencies Calculator to compare emissions from AI data centers to something familiar, like driving a car.
- Some students may benefit from having a note sheet on hand to keep track of unknown terms and unfamiliar initialisms (such as PUE and WUE) that are used throughout the report.
- After reading, students can discuss options for incentivizing AI companies to work with third-party and governmental groups to monitor their impacts and prioritize public well-being over private profits.
This resource is a peer-reviewed article discussing the water and carbon footprint associated with AI servers across the US. The article also examines the efficacy of decarbonization and water-saving efforts. This is a topic that frequently comes up when discussing AI and the environment. The information presented is accurate, and this resource is recommended for teaching.
Teaching Tips
Standards
Resource Type and Format
Scientist Reviewed
This resource has been reviewed by SubjectToClimate's climate scientists and verified for scientific accuracy and up-to-date information. Our review process ensures that every resource in our library reflects the current state of climate science.
Learn about our review process →
