Auditing using artificial intelligence: a systematic literature review with scientometric and topic modeling insights through a socio-technical systems (STS) perspective

Purpose This study aims to examine the existing literature on the use of artificial intelligence (AI) in auditing in terms of its diversity, evolution over time, and dynamics. Design/methodology/approach The authors integrate the systematic literature review, the preferred reporting items for the systematic reviews and meta-analyses framework, scientometric analysis, and topic modeling within socio-technical systems theory. The corpus comprises 236 peer-reviewed papers from the Scopus database, published between 2014 and 2025, and ranked by the Association of Business Schools and Australian Business Deans Council. Findings The authors identify nine latent topics: (1) big data analytics, (2) data mining, (3) robotic process automation, (4) machine learning, (5) natural language processing, (6) complex AI technologies, (7) legal and regulatory environment, (8) auditor’s skills and knowledge, and (9) audit quality. Topics 1–6 represent the technical subsystem, whereas Topics 7, 8, and 9 characterize the joint optimization of social and technical subsystems. Research limitations/implications This study acknowledges limitations related to its reliance on the Scopus database, the selected search terms, the time period, and the inclusion of only peer-reviewed papers. Practical implications The study offers significant benefits for regulators, standard setters, practitioners, and accounting educators, fostering the efficient and effective use of AI in auditing. Social implications The findings suggest that AI technologies are reshaping auditors’ roles and requiring new competencies and robust ethical and regulatory frameworks. Originality/value The findings emphasize the necessity of joint optimization of social and technical subsystems through a novel approach to identifying research gaps in AI-based auditing.