The global diffusion of artificial intelligence exposes an efficiency–transformation gap
Environmental Science and Ecotechnology · 2026-11 · 34 · 100784
DOI: 10.1016/j.ese.2026.100784 ↗Kurz gefasst
Fachartikel in der Zeitschrift Environmental Science and Ecotechnology, 2026-11. Verfasst von Meiling Wu und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Management Science, Chengdu University of Technology, Chengdu, 610059, China. Die Arbeit nennt 77 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Global development agendas increasingly rely on artificial intelligence to accelerate sustainability transitions. As automated systems integrate into power grids and industrial operations, policymakers assume these technological upgrades will naturally translate into broader ecological gains. Yet it remains unclear whether algorithmic efficiency drives the deep structural changes needed for global climate action, or merely optimizes the status quo. Here we show an inherent efficiency-transformation gap in how artificial intelligence affects global sustainability. Analyzing 69 countries from 2000 to 2023, we find that technological diffusion robustly improves operational capacities-such as clean energy access (SDG 7) and industrial infrastructure (SDG 9)-driven by measurable gains in human capital and reduced national energy intensity. However, this techno-efficiency does not extend to systemic environmental governance. Artificial intelligence yields no robust progress in responsible consumption (SDG 12) or climate action (SDG 13). Instead, the rapid deployment of digital infrastructure temporarily crowds out green innovation, exposing a short-term friction between operational optimization and ecological transition. Furthermore, scenario projections to 2100 indicate that even under highly optimistic sustainability pathways, technological growth will not automatically erase regional development divides. Realizing the full promise of artificial intelligence requires policies that explicitly align computational power with structural climate action, rather than relying on efficiency gains to solve ecological crises.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Management Science, Chengdu University of Technology, Chengdu, 610059, China.
- School of Economics and Management, Nanchang University, Nanchang, Jiangxi, 330031, China.
- Graduate School of Environmental Studies, Tohoku University, Sendai, 980-8572, Japan.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Wu, M., Liu, W., He, Z., & Zhang, Z. (2026). The global diffusion of artificial intelligence exposes an efficiency–transformation gap. Environmental Science and Ecotechnology, 34, 100784. https://doi.org/10.1016/j.ese.2026.100784
Vancouver
Wu M, Liu W, He Z, Zhang Z. The global diffusion of artificial intelligence exposes an efficiency–transformation gap. Environmental Science and Ecotechnology. 2026;34:100784. doi:10.1016/j.ese.2026.100784
BibTeX
@article{Wu2026,
author = {Wu, Meiling and Liu, Wei and He, Ziqiong and Zhang, Zhengyang},
title = {The global diffusion of artificial intelligence exposes an efficiency–transformation gap},
journal = {Environmental Science and Ecotechnology},
year = {2026},
volume = {34},
pages = {100784},
doi = {10.1016/j.ese.2026.100784}
}
EUVIMED