Environmental Impact of Generative and Agentic AI: An in-Depth Analysis and Green Solutions

2026-09-26

AI Summary

This study addresses the critical environmental challenges—including energy consumption, carbon emissions, and water usage—arising from the proliferation of generative and agentic AI, noting that existing research has yet to comprehensively cover their full life-cycle impacts. Employing order-of-magnitude estimation based on publicly available data, combined with life cycle assessment and multidimensional environmental modeling, this work systematically evaluates ecological footprints across all stages from hardware manufacturing to end-of-life disposal. It proposes a cross-stage taxonomy and the SAFIA assessment framework, revealing key findings such as inference energy demands potentially exceeding those of training and the exponential energy consumption patterns inherent in agentic workflows. Furthermore, the authors outline a research roadmap through 2035 and advocate for the establishment of mandatory disclosure standards to bridge existing gaps in the field.

Abstract

The proliferation of generative and agentic artificial intelligence (AI) systems has introduced computational demands whose environmental consequences are substantial yet underexamined. This paper examines the environmental footprint of modern AI systems across energy consumption, carbon emissions, water usage, and electronic waste over the full lifecycle of large language models, multimodal foundation models, and agentic workflows, from hardware fabrication and training through fine-tuning and inference to end-of-life disposal. This work provides a conceptual analysis, utilizing order-of-magnitude estimations based on published data, without conducting original physical measurements. We contribute a lifecycle taxonomy that crosses lifecycle phases with five impact dimensions and emission scopes; a Sustainability Assessment Framework for AI Systems (SAFIA) comprising nine indicators; a comparative analysis of traditional, generative, and agentic AI; seven open challenges; policy recommendations for regulators, cloud providers, hardware manufacturers, and AI developers; and a research roadmap to 2035. The analysis indicates that inference can rival or exceed training energy over a deployment lifetime, that agentic workflows can multiply the energy of equivalent single-pass inference by one to several orders of magnitude depending on the number of model and tool calls, that indirect water use and embodied carbon are systematically underreported, and that measurement tools, disclosure practices, and regulation have not kept pace with agentic deployment. These findings call for agent-aware energy measurement, lifecycle-based carbon and water accounting, and mandatory disclosure for large-scale AI training and deployment.

Problem

• Generative AI
• Agentic AI
• Environmental impact
• Lifecycle assessment
• Carbon footprint

Innovation

• Generative AI
• Agentic AI
• Lifecycle Assessment
• Sustainability Framework
• Environmental Impact