S. 4727 — Artificial Intelligence Environmental Impacts Act of 2026

This section finds that artificial intelligence has rapidly increased computational power and data center energy demand, which has tripled over the last decade and is projected to double or triple again by 2028. It states that hyperscale facilities will consume 16 to 33 billion gallons of water annually by 2028, electricity prices rose as much as 13 percent in 2025 and are projected to rise another 25 percent in some areas over five years, and e-waste and environmental harms will grow, while noting potential positive AI applications and disparate community impacts.

This section establishes definitions for the Act, including the Administrator of the Environmental Protection Agency, artificial intelligence (as defined in 15 U.S.C. 9401), an artificial intelligence data center (any facility or group of facilities exceeding 50 megawatts that houses AI inference, training, or deployment infrastructure), an artificial intelligence model, and an artificial intelligence system.

This section directs the Administrator of the Environmental Protection Agency, in collaboration with the Secretary of Energy, the Director of the National Institute of Standards and Technology, the Director of the Office of Science and Technology Policy, the Secretary of the Interior, and the Federal Energy Regulatory Commission, to conduct a comprehensive study on the environmental and energy-related impacts of artificial intelligence and related infrastructure. The study must examine full lifecycle impacts, design and location decisions, disaster resilience, positive and negative effects, and updated data center results, with forecasts for the 2-year, 5-year, and 10-year periods after completion, and incorporate public comments and regional hearings before the report is submitted to Congress and made publicly available within 1 year of enactment.

This section directs the Director of the National Institute of Standards and Technology, in consultation with the Administrator and the Secretary of Energy, to convene a consortium of stakeholders from Indian Tribes, local governments, academia, civil society, and industry. The consortium must develop measurements, methodologies, and standards to report the full lifecycle environmental and energy impacts of artificial intelligence models and systems; create open-source tools for such measurements; and issue recommendations on mitigation and positive use cases. It requires equal representation across stakeholder groups and at least three representatives each from Tribal communities and cities or counties hosting artificial intelligence data centers.

This section requires operators of artificial intelligence data centers to annually report to the Environmental Protection Agency the full range of environmental impacts of those facilities. It directs the Administrator, in consultation with the National Institute of Standards and Technology, the Department of Energy, and others, to develop detailed public reporting requirements covering energy consumption and intensity, water use, air and water pollution, electronic waste, noise and light pollution, and other impacts; to solicit public comment before finalizing those requirements; to publish the reported data online except for trade secrets; and to seek corrective action or impose administrative penalties scaled to facility size and duration of noncompliance after six months without a required report.

This section requires the Administrator, the Secretary of Energy, and the Director of the National Institute of Standards and Technology to jointly submit to Congress, no later than two years after development of reporting requirements, a report on the findings of a consortium studying artificial intelligence, information submitted under those requirements, recommendations for legislative or administrative actions to mitigate negative AI impacts, and recommended environmental use cases for artificial intelligence based on net-benefit assessments.