Auditing Artificial Intelligence-Driven Financial Systems: Accountability, Transparency, and Auditor Liability in Algorithm-Based Decision Making

Abstract

Purpose: This study examines how the growing use of Artificial Intelligence (AI) in financial reporting and auditing affects audit reliability, accountability, and transparency. It focuses on key challenges such as AI’s “black box” nature, outdated auditing standards, limited auditor expertise, and unclear legal responsibility. Method: The study uses a conceptual and literature-based approach by reviewing prior research, auditing standards, and regulatory developments related to AI, financial reporting, and audit assurance. Findings: The study finds that although AI can improve risk assessment and audit efficiency, its complexity and lack of transparency may increase audit risk. Current standards, such as ISA 315 and ISA 500, are not fully suitable for algorithm-based decision-making. The study also highlights a shortage of auditors with data science skills and uncertainty over legal accountability between auditors, companies, and AI software providers. Implications: The study proposes the Assurance for Ethical and Governed AI Systems (AEGIS) framework, which emphasizes system review, AI explainability, and continuous monitoring. It recommends that standard setters, including the IAASB, develop AI-specific audit guidance, strengthen auditor training in data analytics and AI governance, and create a fairer legal responsibility framework. Without these changes, the audit profession may struggle to remain relevant in an AI-driven reporting environment.