Big data audit is a novel auditing mode that applies emerging information technology to auditing business activities. In recent years, with increasing attention to big data audit, exploring the value creation mechanisms and pathways of big data audit mode has become an important topic in both academic and practical realms. However, existing research on big data audit has not yet formed a clear theoretical system and lacks a systematic framework. Therefore, this paper reviews literature on big data audit published in authoritative journals domestically and internationally from 2011 to 2023. Firstly, it systematically expounds on the origin, conceptual connotation, and comparative analysis of big data audit. Secondly, based on the analytical framework of “core link–key element–basic guarantee–fundamental purpose”, it comprehensively sorts out the research dynamics of big data audit thinking, process, technology, and application, revealing the internal logic of big data audit and supplementing relevant evidence with empirical research. Finally, it constructs a theoretical framework for big data audit research and proposes future prospects from different perspectives such as audit forensics mode, audit early-warning mechanism, audit risk-prevention mechanism, and audit effectiveness, aiming to provide reference for further promoting research in big data audit.