Is ChatGPT competent? Heterogeneity in the cognitive schemas of financial auditors and robots

The constraints of ChatGPT, as an intelligent conversational robot, in mimicking complex human activities have created doubts about its competence in the financial profession. Prior research has shown that the limitations of such robots result from differences between human and artificial cognitive structures, such as differences in structural approaches to interpreting information by organizing knowledge. However, explanations of how such differences are associated with human characteristics remain limited. This study focuses on one specific finance profession, that of financial auditors, to demonstrate how ChatGPT shows advances that mean it can be argued to imitate financial auditors with longer tenures. Our findings suggest possible applications and limitations of ChatGPT in light of these advances.