LARGE LANGUAGE MODELS AND THE TRANSFORMATION OF PROFESSIONAL ACCOUNTING PRACTICE: FROM DATA PLUMBING TO STRATEGIC INTERPRETATION
International Journal of Advanced Research · 31.03.2026 · 14 (03) · 794-798
DOI: 10.21474/ijar01/22999 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research, online erschienen am 31.03.2026. Verfasst von Francis Melaragni.
Abstract (Originalsprache)
Large language models (LLMs) are reshaping the professional landscape of accounting and finance, shifting practitioners from labor-intensive data preparation toward higher-order analytical reasoning. This paper examines how LLMs can be strategically deployed across core accounting workflows - from financial document analysis and regulatory compliance to audit risk assessment - while identifying the governance structures necessary to ensure professional integrity. Drawing on recent empirical benchmarks, industry case studies, and the emerging retrieval-augmented generation (RAG) architecture, we argue that the critical variable is not whether LLMs are used, but whether they are deployed within a framework that preserves professional judgment, data privacy, and regulatory fidelity. Implications for accounting educators, practitioners, and firms undertaking phased AI adoption are discussed.
Zitieren
APA
Melaragni, F. (2026). LARGE LANGUAGE MODELS AND THE TRANSFORMATION OF PROFESSIONAL ACCOUNTING PRACTICE: FROM DATA PLUMBING TO STRATEGIC INTERPRETATION. International Journal of Advanced Research, 14(03), 794-798. https://doi.org/10.21474/ijar01/22999
Vancouver
Melaragni F. LARGE LANGUAGE MODELS AND THE TRANSFORMATION OF PROFESSIONAL ACCOUNTING PRACTICE: FROM DATA PLUMBING TO STRATEGIC INTERPRETATION. International Journal of Advanced Research. 2026;14(03):794-798. doi:10.21474/ijar01/22999
BibTeX
@article{Melaragni2026,
author = {Melaragni, Francis},
title = {LARGE LANGUAGE MODELS AND THE TRANSFORMATION OF PROFESSIONAL ACCOUNTING PRACTICE: FROM DATA PLUMBING TO STRATEGIC INTERPRETATION},
journal = {International Journal of Advanced Research},
year = {2026},
volume = {14},
number = {03},
pages = {794--798},
doi = {10.21474/ijar01/22999}
}
EUVIMED