AI&SupplyChain
Description
This dataset contains the publicly shareable data used in the study “AI Innovation Spillovers along the Supply Chain: Evidence from Contractual Relationship Stability.” The deposited files include the annual-report text corpus collected from publicly available corporate disclosure documents through web crawling. The annual reports are used to support the identification and verification of supplier–customer relationships, and to supplement the extraction and validation of patent-related information. This dataset does not contain the full replication data. The main empirical data are drawn from CSMAR, and CNRDS, which are subject to institutional subscription and licensing restrictions and therefore cannot be redistributed by the authors. In addition, the validation process involved discussions with experts and managers to reduce measurement noise caused by strategic patents, defensive patents, patent thickets, and low-quality patent accumulation. The related validation records, expert comments, manually screened patent lists, and firm-specific feedback involve specific listed companies and industry confidential information. They are covered by non-disclosure agreements and therefore cannot be made publicly available.