Artificial intelligence and green growth nexus: evidence from OECD countries using panel CS-ARDL and wavelet analysis

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Provides novel empirical evidence using CS-ARDL and wavelet methods for the AI-green growth nexus, useful for researchers studying the macro-level impact of AI on sustainability.

🏛政策担当者:Highlights that AI's environmental benefits depend on complementary policies like environmental taxes and renewable energy investments, informing integrated policy design.

This study explores the dynamic relationship between artificial intelligence (AI) and green growth across 38 OECD countries during 1996–2024. It assesses whether AI acts as a catalyst for sustainable economic transformation and environmental improvement, while accounting for complementary factors such as renewable energy consumption, environmental taxation, investment, and health expenditure. The study employs the Panel Cross-Sectionally Augmented Autoregressive Distributed Lag (CS-ARDL) model and Wavelet Coherence Analysis to examine both short- and long-run interactions, as well as time-frequency dynamics between AI and green growth. The dataset is balanced, covering 38 OECD countries, and uses AI-related publications per million population as a proxy for AI development. The empirical results indicate that AI significantly enhances green growth in the short run, and this effect strengthens cumulatively over the long term. Environmental taxes exhibit an immediate positive effect, while renewable energy consumption initially generates transitional costs before contributing positively in the long run. The wavelet analysis reveals that the AI–green growth relationship has intensified after 2010, coinciding with digital transformation policies and the diffusion of modern AI technologies. Causality tests confirm a unidirectional causality from green growth to AI, implying that sustainable development policies stimulate AI innovation through feedback effects. The findings highlight that AI's environmental benefits depend on complementary policy frameworks and institutional capacities. While the study relies on AI publication intensity as a proxy for technological advancement, it calls for future research incorporating industrial adoption metrics. Policymakers should integrate AI strategies with environmental taxation and renewable energy investments to strengthen the “Green AI” framework. This study is among the first to integrate Panel CS-ARDL and Wavelet Coherence methods to analyze the AI–green growth nexus for OECD countries. By bridging short- and long-term analyses in both time and frequency domains, it provides novel evidence supporting the “Green AI” paradigm, which emphasizes AI's role in fostering low-carbon, innovation-driven sustainable development.

• semanticscholar https://doi.org/10.1108/meq-08-2025-0576first seen 2026-06-10 05:28:42