Artificial Inteligence’s (AI) Adoption in Accounting: Managing Large Language Models (LLMs)
Main Article Content
Abstract
Introduction: This paper examines the use of artificial intelligence applications to support accounting decisions and it tries to answer the following question: can artificial intelligence assist accounting professionals and what role it has compared to human expertise?
Literature review/research gap: This research contributes to the extant literature in a threefold way. First, it expands the debates on how accountants can leverage AI and are influenced by it. Second, it identifies the main trust issues that have so far blocked their extensive adoption. Third, it argues that, in order to advance theories in this area, the accounting field needs empirical studies that allow policymakers and managers to make informed decisions concerning organizational challenges and necessary adaptations.
Research method: This study considers the tool of Large Language Models (LLMs), that allow companies to automate tasks and make better decisions given their natural language processing capabilities (Rowdur, 2023), and is informed by recent advances in artificial intelligence-based publications, including quantum information research.
Findings: this paper presents how AI technology has influenced the accounting profession so far and in what ways this discipline could be affected in the future, to understand the impact of knowledge-based systems on human users’ knowledge acquisition and retention and considering that, as the development of LLMs continues and despite their impressive abilities, organizations face trust issues when adopting these tools.
Theoretical and practical implications: the present study can inform both the theory and the practice and considering that in the near future professional hybrids will emerge, the area of AI in accounting will certainly benefit from interdisciplinary research (Hasan, 2021). Thus, accounting and information science scholars have to collaborate with data scientists to find theoretical frameworks and choose the consequent adequate algorithmic solutions (Kemper and Kolkman, 2019), including information-theoretical concerns with regard to the data needed and how to guarantee a widespread diffusion of AI.
Australian Academy of Business Research, Volume 1, issue 1, September 2023, pp 20-31
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.