Does artificial intelligence enhance the circular bioeconomy's decarbonization potential? A quasi-experimental evidence

Amid accelerating climate change and resource constraints, integrating artificial intelligence (AI) with the circular bioeconomy presents a promising yet underexplored decarbonization pathway. This study provides the first causal evidence on AI's carbon reduction potential within circular bioeconomy systems using a quasi-experimental triple-difference design. Results demonstrate that AI innovation capacity—proxied by patent activity—is associated with a 3.53% reduction in carbon emissions in circular bioeconomy pilot cities relative to non-pilot cities, statistically significant at the 1% level. Heterogeneity analysis reveals that biomass resource endowment and environmental regulation intensity significantly moderate policy effectiveness. We quantify welfare impacts through economic valuations ($13.05–$761.25 million) and mortality prevention (155–327 avoided deaths over 80 years). Theoretically, we identify five mechanism pathways—resource efficiency optimization, product lifecycle extension, intelligent waste management, demand forecasting, and circular business model enablement—explaining how digital intelligence amplifies material circularity. These findings demonstrate that AI-enabled circular bioeconomy interventions deliver welfare benefits comparable to established climate policies while offering superior cost-effectiveness across jurisdictions.