AuditGPT: Automated Financial Auditing and Regulatory Compliance Checks using LLMs
Bibliographic record
Abstract
Financial auditing and regulatory compliance are essential for maintaining transparency, accountability, and adherence to legal standards. However, traditional auditing processes often rely heavily on manual effort and lack the scalability and explainability required in complex financial environments. This paper introduces AuditGPT, a novel framework that leverages Retrieval Augmented Generation (RAG) and large language models (LLMs) to automate and enhance financial auditing and compliance verification. AuditGPT processes a wide range of financial documents including audit reports, sales records, and regulatory texts by embedding them into a vector store for efficient semantic retrieval. When queried, the system generates evidence backed assessments with structured verdicts: Compliant, Non Compliant, or Requires Review. These outputs are supported by traceable references to the original source documents, ensuring transparency and verifiability. An experimental evaluation on real world audit and sales datasets demonstrates that AuditGPT outperforms standard LLM based approaches across multiple metrics, including compliance accuracy, factual consistency, and retrieval relevance. The results highlight AuditGPT’s potential to significantly reduce manual workload, improve auditing accuracy, and offer greater explainability in decision-making. By addressing the limitations of conventional tools, AuditGPT establishes a scalable and trustworthy foundation for intelligent financial compliance monitoring.
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 | 0.003 | 0.018 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.005 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.003 |
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".