Research on the Reconstruction of Government Audit Functions and the Logic of Intelligent Transformation Driven by Artificial Intelligence

Abstract

Against the backdrop of the continued development of digital government, new-generation information tech nologies such as artificial intelligence are profoundly transforming the operation of public governance and exerting a systemic influence on the functional structure and operational logic of government audit. Drawing on theories of technological empowerment and algorithmic governance, this paper constructs an analytical framework of “technology–institutions–power” interactions and systematically examines the theoretical foundations, internal logic, practical risks, and transformation pathways of the reconstruction of government audit functions under intelligent conditions. Artificial intelligence is extending the functions of government audit from ex post supervision to whole-process supervision, deepening them from outcome evaluation to process diagnosis, and expanding them from single-actor supervision to collaborative governance. At the same time, the transformation faces risks at the technological, institutional, and governance levels, including al gorithmic opacity, institutional lag, and power imbalances. Accordingly, intelligent transformation should be advanced under the guidance of functional reconstruction. Institutional norms should be improved, account ability mechanisms clarified, and governance coordination strengthened to ensure the standardized operation of functional reconstruction and enhance its overall effectiveness.