| 胡化广,刘海建,周枫凯.人工智能的“稳链”效应——来自供应链波动的经验证据[J].数量经济技术经济研究,2025,(10):90-111 | | 人工智能的“稳链”效应——来自供应链波动的经验证据 | | Stabilizing Effect of Artificial Intelligence on Supply Chains: Empirical Evidence from Volatility Dynamics | | | | DOI： | | 中文关键词: 人工智能 稳链 供应链波动 供应链韧性 | | 英文关键词: Artificial Intelligence Stabilize the Chain Supply Chain Volatility Supply Chain Resilience | | 基金项目: | | | 中文摘要: | | 增强供应链稳定性和韧性关乎新质生产力的释放和价值实现, 是统筹发展与安全的必然要求。综合上市公司数据和全国企业税收调查数据，在创新性构建供应链波动测度指标的基础上，考察了企业人工智能应用对供应链波动的影响，研究发现，人工智能，尤其是具有通用人工智能潜力的人工智能，能够有效降低企业的供应商波动和客户波动，提高了供应链稳定性。机制检验表明，人工智能应用不仅有助于缓解被上下游“卡脖子”问题，推动供应链多元化突破，还畅通了供应链信息流，升级了供应链实物流，进而发挥了“稳链”效应。异质性检验发现，人工智能“稳链”效应在对跨边界创新需求强、面临的供需变化不确定性高、机器人嵌入潜力大的企业中更显著，说明“稳链”效应与人工智能独特的自适应能力、预测能力以及智能实体应用有关。进一步研究发现，人工智能应用缓释了外部供应链冲击引致的供应链剧烈波动，增强了供应链韧性。人工智能主要抑制了供应链过度波动，而不会使供应链陷入僵化困境。本文为深入推进“人工智能+”供应链行动提供了理论支撑，更为加快构建新安全格局和培育新质生产力提供了重要政策启示。 | | 英文摘要: | | The security and stability of the supply chain are the cornerstone for establishing a new development paradigm. In recent years, the severe and complex domestic environment in China and the international situation have posed substantial threats to the stability and continuity of supply chain operations for Chinese firms. Therefore, the issue of how to enhance supply chain stability has emerged as a strategic concern commanding high-level national attention. The transformation and upgrading of the supply chain depend on the adoption of advanced technologies. Artificial intelligence (AI), as the most emblematic digital and intelligent technology, exerts a profound influence and is driving a new wave of industrial revolution.By integrating data from publicly listed companies and China’s National Enterprise Tax Survey, this study constructs a novel supply chain stability index and examines how firms’ adoption of AI affects it. The research findings indicate that AI, particularly that with the potential for general-purpose intelligence, can effectively reduce fluctuations in both supplier and customer relationships, thereby enhancing supply chain stability. The mechanism analysis indicates that the application of AI helps alleviate the “bottleneck” constraints imposed by upstream and downstream partners and promotes breakthroughs in supply chain diversification. In addition, it facilitates smoother information flows and upgrades the physical logistics of the supply chain, thereby reinforcing its stabilizing effect. The heterogeneity analysis reveals that the supply chain stabilizing effect of AI is more pronounced in firms with strong cross-boundary innovation demands, high supply–demand uncertainty, and substantial potential for robotic integration. These findings reveal that the stabilizing effect is closely related to AI’s distinctive capabilities in self-adaptation and predictive analytics, as well as its applicability in intelligent physical systems. Further analysis reveals that the application of AI mitigates abnormal supply chain fluctuations induced by external risk shocks and enhances overall supply chain resilience. Finally, AI primarily curbs excessive supply chain volatility while preserving the flexibility and adaptability of the supply chain, thus avoiding the risk of rigidity.Based on the above conclusions, this study proposes the following recommendations: First, deepen the “AI + Supply Chain” initiative to provide support for the establishment of a new security framework. Second, AI technology should be utilized to enhance the autonomy and controllability of the supply chain, thereby laying a solid foundation for cultivating new quality productive forces. Third, improve the “risk-opportunity” dual-module integrated supply chain system driven by AI and coordinate development and security.The contributions of this study are as follows: First, it extends research on the measurement of supply chain stability by introducing a new index. While recent research has focused on the role of digital technologies, such as the Internet of Things (IoT), big data, and cloud computing, in stabilizing supply chains, the unique contribution of AI to supply chains has not been sufficiently explored. This study empirically demonstrates the relationship between AI applications and supply chain stability, highlighting the distinction between AI-driven supply chain empowerment and other digital technologies, such as IoT and big data, thus deepening the understanding of AI’s stabilizing effects. Second, it enriches research on the economic impacts of AI applications. While existing empirical studies on AI’s economic impact mainly address areas such as labor employment, product innovation, and business performance, only a few have examined AI’s effects within the broader supply chain context. This study fills this gap by providing direct empirical evidence of AI’s influence on supply chain stability, thus expanding research on AI’s economic impacts. Third, using data from listed companies and tax surveys, the study delves into the mechanisms through which AI impacts supply chain stability, including its effects on “diversification breakthrough,” “information flow unblocking,” and “logistics optimization.” This analysis deepens the understanding of AI’s “supply chain stabilizing” effect. Furthermore, this study finds that AI application not only alleviates excessive supply chain fluctuations but also prevents the loss of growth opportunities for enterprises. Fourth, from a dynamic perspective, this study demonstrates how AI enhances supply chain resilience by strengthening its absorptive capacity. Using the supply chain volatility indicator, the study indicates that AI mitigates external risk shocks, thus improving supply chain resilience. | | 查看全文 相关附件： 下载数据代码附录 |
| | |
