Mitigating Biases in Managerial Decision Making with Assistance from Large Language Model Tools: The Case of Artificial Intelligence Auditing

Artificial intelligence, Auditing, Business, Bias, Decision making, Regulation, entrepreneurship, LLM

Managers have historically relied on data for decision-making. The purpose of this study is to explore the potential enhancement of managerial decision-making capabilities through the usage of Large Language Models (LLMs). The study aims to investigate how LLM-based tools could provide cues to managers, influencing their decision-making process and contributing to better organizational decision-making. The research employed a theoretical approach to examine the role of LLMs in managerial decision-making. It assessed how LLM tools offered decision choices to managers and explored the convergence divergence between LLM suggestions and human intelligence. The study emphasized the phenomenon of artificial intelligence avoidance and underscored the need for artificial intelligence auditing at both the planning and operational phases to ensure effective decision-making. The findings highlighted that LLM-based tools could significantly impact managerial decision-making. When LLM suggestions aligned with managers' human intelligence, decision certainty could be enhanced. However, if there was a divergence, managers might experience uncertainty due to artificial intelligence avoidance. The study emphasized the importance of auditing LLM tools, particularly in the context of business organizations, to address biases and ensure reliable decision support for managers. This study contributes to the literature by providing insights into artificial intelligence auditing in the context of business organizations. It underscores the originality of exploring the convergence and divergence between LLM suggestions and human intelligence, shedding light on the nuanced dynamics of managerial decision-making in the presence of artificial intelligence tools.