With the improvement of enterprise business complexity and the surge in data scale, traditional audit risk identification methods are facing bottlenecks such as insufficient full-data processing and delayed risk response. This paper focuses on the construction of an audit risk identification model based on big data technology, combining the advantages of big data in full-volume analysis, real-time processing and intelligent mining, and clarifies the core logic and technical support for model construction. Firstly, it sorts out the core elements of audit risk identification and the adaptability of big data technology, then constructs a model framework from the data layer, technology layer and application layer, and focuses on expounding the design points of multi-source data integration mechanism, intelligent risk feature mining and dynamic risk assessment system. The research shows that the model can break through the limitations of traditional experience-driven methods, realize accurate identification, dynamic tracking and forward-looking early warning of audit risks through data-driven approach, and effectively improve the efficiency and accuracy of audit risk identification. The research of this paper provides a feasible model reference for the digital transformation in the audit field, and is of great significance for promoting the development of audit risk identification towards intelligence and systematization.