AI-Enabled Fraud Detection Readiness in Sarawak SMEs: A Quantitative TOE-UTAUT Model for Accounting Transformation
TL;DR
A quantitative research model for examining artificial intelligence (AI)-enabled fraud detection readiness among small and medium-sized enterprises (SMEs) in Sarawak, Malaysia and provides a replicable model for studying digitally underserved regional economies is developed.
Abstract
This manuscript develops a quantitative research model for examining artificial intelligence (AI)-enabled fraud detection readiness among small and medium-sized enterprises (SMEs) in Sarawak, Malaysia. Rather than treating AI adoption solely as a general technology acceptance issue, the paper positions adoption readiness as a multi-layered condition shaped by technological fit, organizational mobilisation, institutional facilitation, and user-level acceptance. Drawing on the Technology-Organization-Environment (TOE) framework and the Unified Theory of Acceptance and Use of Technology (UTAUT), the proposed model explains how relative advantage, compatibility, cost, security, task-technology fit, business innovativeness, top management support, government support, effort expectancy, and social influence may affect SMEs' intention to adopt AI in accounting. Trust and firm size are incorporated as boundary conditions to capture behavioral uncertainty and resource asymmetry in regional SME settings. The manuscript further specifies a survey-based methodological protocol involving purposive sampling, a seven-point Likert-scale questionnaire, pilot testing, SPSS-based data screening, and Partial Least Squares Structural Equation Modelling (PLSSEM) using SmartPLS. By emphasizing readiness, measurement logic, and empirical testability, the paper offers a methodological contribution to AI accounting research and provides a replicable model for studying digitally underserved regional economies.