ARTIFICIAL INTELLIGENCE IN TAX ENFORCEMENT: THE ROLE OF PERCEIVED AI CAPABILITY IN SHAPING TAX EVASION INTENTION

This study examines the effect of perceived artificial intelligence (AI) capability on tax evasion intention among corporate taxpayers in Indonesia. As digitalization and the adoption of AI in tax administration continue to expand, understanding how taxpayers cognitively respond to advanced technological surveillance has become increasingly important, particularly in developing country contexts. Using a quantitative explanatory design, data were collected through an online structured questionnaire administered to corporate tax decision-makers, yielding 278 valid responses. Hypotheses were tested using Partial Least Squares–Structural Equation Modeling (PLS-SEM) with SmartPLS. The empirical results indicate that perceived AI capability has a positive and significant effect on tax evasion intention, suggesting that the hypothesized negative relationship is not empirically supported. This finding implies that higher perceptions of AI-based surveillance capability do not automatically deter tax evasion intentions. Instead, they may encourage more adaptive and strategic responses in corporate tax planning. Corporate taxpayers appear to respond to sophisticated monitoring technologies by engaging in more complex risk evaluations rather than uniformly increasing compliance. The study contributes to the tax behavior literature by integrating perceived AI capability as a technology-based psychological factor within the behavioral intention framework. From a practical perspective, the findings suggest that the implementation of AI in tax administration should be accompanied by policies emphasizing transparency, legal certainty, and clear risk communication to prevent strategic behavioral adaptation by corporate taxpayers.