What type of study is this?

August 16, 2025

Propensity factors of artificial intelligence technology adoption by public sector auditors

Key Points

Data mining adoption in supreme audit institutions identified key enabling and inhibiting factors, pointing to low implementation rates despite its potential benefits.
Important metrics revealed include stakeholder pressures for sophisticated technology adoption to enhance auditing performance and accountability.
This research applied structural equation modeling in a dual-factor approach to address the complexities of technology acceptance and resistance.
Findings outline essential considerations for overcoming barriers to AI implementation in public auditing, supporting necessary digital transformation efforts.

Abstract

Purpose Data mining (DM) represents an example of artificial intelligence (AI) technology aimed at discovering patterns, trends, correlations or other useful information in large sets of data. The potential of this technology can be applied in different fields, including auditing. DM helps in activities such as fraud detection, data extraction, querying, manipulation analysis, risk assessment, data summarization, high-risk transaction and unusual item detection, continuous monitoring, data analytics, key performance indicator tracking and identifying trends in transactions (Amani and Fadlalla, 2017; Cardoni et al., 2020; Gray and Debreceny, 2014; Papík and Papíková, 2020). Yet, despite the usefulness of DM, its adoption is low (De Almeida and Pedrosa, 2011). This study aims to investigate the drivers of DM adoption by auditors within supreme audit institutions (SAIs). The latter face a lot of pressure from stakeholders to adopt sophisticated technologies and thereby improve the performance of their auditing activities (Otia and Bracci, 2022). Design/methodology/approach To address this topic, this paper adopts an explorative approach. We use a dual-factor perspective, which investigates both the enabling and inhibiting factors that lead to technology adoption or resistance. We combine the technology acceptance model and status quo bias theory. We collected a total of 206 questionnaires and analyze the results by applying structural equation modeling. Findings This paper contributes to the literature by identifying various elements that explain resistance to the introduction of DM in SAIs. This is useful in practice for accelerating digital transformation (Otia and Bracci, 2022) and addressing the demands for greater accountability in SAIs (Cordery and Hay, 2019). Originality/value The results contribute significantly to defining the key elements necessary to reduce resistance toward AI implementation in this field. By addressing both the enablers and barriers, the study provides a foundational framework for understanding how AI can be effectively integrated into public auditing, supporting the move toward digital transformation.

Ask AI

Helpful

Bookmark