What question did this study set out to answer?

The aim is to review the environmental effects of artificial intelligence, uncovering key trends and networks.

March 21, 2026Open Access

The Environmental Effects of Artificial Intelligence: A Bibliometric Perspective

Key Points

The aim is to review the environmental effects of artificial intelligence, uncovering key trends and networks.
Analyzed English journal articles from 2015 to 2025 in the Web of Science Core Collection.
Conducted bibliometric and knowledge mapping analyses using CiteSpace and VOSviewer software.
Examined collaboration networks and research trends over time.
Collaboration networks are multi-centered but display low density, indicating room for international cooperation.
Research has significantly increased since 2019, transitioning from exploratory to practical applications.
Focus from 2023 to 2025 centers on utilizing AI to combat climate change, emphasizing technologies like deep reinforcement learning.
Hotspots include AI's role in energy management and sustainable development, with rising interest in explainable AI and edge computing.

Abstract

HRMARS - Purpose: This paper aims to provide a comprehensive review of the environmental effects of Artificial Intelligence (AI), identifying key research forces, collaboration networks, and evolving trends. Design/methodology/approach: This study focuses on English journal articles published between 2015 and 2025 in the Web of Science Core Collection. Bibliometric and knowledge mapping analysis were conducted using CiteSpace and VOSviewer software to analyze the data. Findings: First, the collaboration network exhibits a multi-center structure. However, the network density remains low, suggesting potential for increased international collaboration. Second, research has grown rapidly since 2019, evolving from exploratory studies to practical applications. From 2023 to 2025, the focus shifted toward addressing climate change using AI, with advanced technologies such as deep reinforcement learning and explainable AI gaining prominence. Third, research hotspots include core AI technologies, AI’s impact on energy and carbon management, and its role in sustainable development. Finally, the scope of AI applications is expanding into new areas, such as smart cities and clean energy, alongside growing attention to explainable AI and edge computing. Research limitations/implications: Researchers should strengthen international collaboration, explore advanced techniques like explainable AI and edge computing, and address the dual challenge of maximizing AI’s environmental benefits while minimizing its energy footprint. Practical implications: Practitioners should integrate AI tools such as deep learning and intelligent energy management systems to optimize energy efficiency, reduce carbon emissions, and support green transformation initiatives. Originality/value: This paper contributes by offering a large-scale, systematic, and visual analysis of AI’s environmental effects. The results clarify the current research landscape, highlight hotspots, and map evolving trends.

Demander à l'IA

Bookmark

View Full Paper