Digital Transformation in Public Sector Organization: Examine Machine Learning Adoption in Accounting and Auditing Practices with Moderating effect of Regulatory Framework in Khyber Pakhtunkhwa

Abstract

Digital transformation has become a critical priority for public sector organizations, aiming to improve efficiency, transparency, and accountability in financial management. Among emerging technologies, machine learning a key tool of artificial intelligence has the potential to transform accounting and auditing practices by enabling automated data analysis, pattern recognition, predictive forecasting, and anomaly detection. This study examines the adoption of machine learning in accounting and auditing practices within public sector organizations of Khyber Pakhtunkhwa, with a focus on the moderating effect of the regulatory framework. Grounded in the Technology Acceptance Model and Institutional Theory, the study investigates how machine learning enhances financial accuracy, audit quality, and decision making, and how regulatory structures influence its effective implementation. A quantitative research design was employed, collecting primary data from 562 accounting and audit personnel across diverse public sector organizations using a structured questionnaire. Data analysis involved descriptive statistics, correlation analysis, structural equation modeling, and multiple regression analysis. The findings indicate that machine learning adoption significantly improves accounting and auditing practices, while a strong regulatory framework strengthens this positive relationship. The study highlights the critical role of regulatory oversight in ensuring ethical, transparent, and accountable use of machine learning in public financial management. These results provide empirical evidence to guide policymakers, regulators, and public sector administrators in promoting digital transformation initiatives. The study concludes that integrating machine learning with robust regulatory governance can significantly enhance financial management, operational efficiency, and institutional accountability in public sector organizations of Khyber Pakhtunkhwa. Recommendations for effective implementation, capacity building, and employee training are provided, while noting limitations such as the focus on a single province and cross sectional data; future research could extend to other regions and explore additional artificial intelligence technologies in both public and private sectors.