Governing Artificial Intelligence Through Auditing

The rapid diffusion of artificial intelligence (AI) across sectors raises concerns about transparency, accountability, and ethical compliance. Traditional regulatory frameworks often lag behind technological change, making auditing a critical governance mechanism. This study conducts a systematic literature review to examine the role of auditing in AI governance, highlighting its benefits and future prospects. Findings reveal that audits extend beyond technical evaluation to embed accountability, independence, and legitimacy. Independent and third-party audits mitigate conflicts of interest, while lifecycle and risk-based approaches enable proportional and continuous oversight. Methodological innovations, such as hybrid white-box and black-box audits, provide assurance even where access to proprietary systems is restricted. Auditing fosters institutional accountability by embedding ethical standards, deterring malpractice, and building public trust. Standardisation and accreditation enhance credibility and reduce audit-washing.