SECURING AUDIT INTEGRITY: BALANCING AI EFFICIENCY AND HUMAN JUDGMENT VIA THE CO-PILOT FRAMEWORK AND MANDATORY TRANSPARENCY STANDARDS

Abstrak

The rapid advancement of digital technology has transformed the auditing profession, shifting traditional manual methods to proactive, real-time analyses leveraging Artificial Intelligence (AI) and Machine Learning (ML). While AI significantly enhances efficiency and accuracy (e.g., achieving a 95.7% accuracy rate in classifying over 14 million financial records), its integration introduces critical challenges, including algorithmic opacity (black-box nature) and the risk of over-reliance on automated tools, which threaten professional skepticism and accountability. This qualitative study, utilizing a Systematic Literature Review, is distinct from prior technical research, focusing instead on investigating how AI-driven auditing frameworks influence auditors' professional behavior, ethical considerations, and perceptions of reliability to secure audit integrity. The research confirms three essential propositions: First, AI must function as a 'co-pilot' to maximize efficiency gains while actively mitigating risks posed by opacity. Second, despite confirmed improvements in efficiency and accuracy, auditors' professional judgment and skepticism remain essential for safeguarding audit quality due to inherent limitations in transparency and the risk of algorithmic errors. Finally, cross-jurisdictional regulations, ethics, and cultural contexts profoundly influence AI adoption, underscoring that consistent global ethical and regulatory guidance is critical for ensuring fairness and sustainable audit quality worldwide. These findings offer timely practical insights for regulators formulating consistent ethical standards for responsible AI implementation.