Artificial intelligence, institutional quality, and carbon neutrality: a pathway analysis of OECD nations

This study examines how institutional quality affects the relationship between artificial intelligence (AI) adoption and carbon neutrality across 35 OECD countries from 1990 to 2020. Using a dynamic panel approach and the Augmented Anderson–Hsiao (AAH) estimator, it investigates whether AI reduces emissions and if its benefits depend on governance strength. The results reveal that AI adoption alone is positively associated with carbon emissions, underscoring its energy-intensive nature. However, the interaction between AI and institutional quality has a significant negative effect on carbon dioxide (CO2) emissions, highlighting the vital role of strong institutions in steering AI toward sustainable results. Additionally, globalization has had limited but positive effects on carbon emissions, while urbanization and the energy transition have shown mixed outcomes. Overall, the study underscores the importance of institutional frameworks in aligning technological innovation with climate goals and provides evidence that AI can support achieving carbon neutrality when backed by effective governance.