AI-Driven Asymmetry in Supply Chain Concentration Risk: Upstream and Downstream Comparative Analysis

AI technology is widely used in contemporary supply chain management; however, its asymmetric effects on upstream and downstream supply chain concentration risks remain underexplored. Based on information asymmetry theories, this study empirically examines the relationships between corporate AI technology and asymmetric supply chain concentration risks. Specifically, Python 3.8.10-based word frequency analysis is used to measure corporate AI technology levels, and the Ordinary Least Squares (OLS) regression technique is applied to analyze panel data from Chinese listed firms from 2013 to 2025. The results reveal a significant asymmetric effect: AI reduces upstream supplier concentration risk but increases downstream customer concentration risk. This divergent pattern breaks the symmetry of risk distribution between upstream and downstream supply chain partners. As a result, AI adoption lowers firms’ total supply chain management risk. Further analysis shows how information asymmetry critically moderates these effects. Among firms facing greater information asymmetry, AI strongly mitigates upstream supplier and overall supply chain management risk. Nevertheless, its moderating effect on downstream customer concentration risk is not significant. This conditional asymmetry demonstrates that AI’s risk impacts are not fixed but depend on the symmetric state of information exchange in supply chains. As such, this study clarifies AI’s asymmetric governance role in supply chains and reveals the broken symmetry in upstream and downstream risk evolution driven by AI. It defines conditions for AI-enabled risk optimization and enriches information asymmetry theories by explaining how information imbalance reshapes the symmetric properties of supply chain risk. The findings offer practical guidance for firms to tailor AI strategies and balance upstream and downstream risk management.