Agentic AI and Human–AI Collaboration in Auditing: Roles, Benefits, and Risks in the Korean Audit Market
Bibliographic record
Abstract
<p>Artificial intelligence (AI) is rapidly transforming the auditing profession (Dong et al. 2023; Gu et al. 2024). Early pplications primarily focused on automating routine tasks, such as journal entry testing or anomaly detection. More recently, advances in AI have given rise to agentic AI—systems capable of autonomous goal setting, iterative reasoning, and adaptive task execution (Li et al. 2025). Unlike traditional decision-support tools, agentic AI can actively participate in audit processes by identifying risks, proposing procedures, and generating documentation. This shift raises fundamental questions regarding auditor responsibility, professional judgment, and accountability. These issues are particularly salient in jurisdictions with strong regulatory oversight and high legal exposure for auditors. The Korean audit market provides a distinctive and informative setting in this regard. Korea is characterized by stringent auditor liability under the External Audit Act, intensive regulatory inspections by the Financial Supervisory Service (FSS), and a growing emphasis on audit process documentation and consistency. This commentary examines how agentic AI may be integrated into auditing through a human–AI collaboration framework, focusing on the roles, benefits, and risks of such collaboration in the Korean audit environment.</p>
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.003 | 0.008 |
| Scholarly communication | 0.007 | 0.007 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.005 | 0.004 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".