Why Artificial Intelligence Should Never Be Used in Managerial Positions: A Critical Analysis of Automated Decision-Making
0 viewsAbstract
This study critically examines why artificial intelligence should not be granted managerial authority, even though AI demonstrably improves performance in narrow, well-defined tasks. Management requires contextual judgment, ethical reasoning, and accountability, capacities that automated systems do not possess. The study synthesizes three bodies of evidence: the behavioral literature on algorithm aversion (as cited in Dietvorst et al., 2015), the organizational literature on algorithmic control and its consequences (Kellogg et al., 2020), and documented case evidence of algorithmic failures in hiring, criminal justice, and social services . A qualitative, multi-source analysis classifies documented failures by the stage of the decision process in which they occur, revealing that failures originate systematically in biased training data, misaligned optimization objectives, and the absence of accountability mechanisms. The findings show that algorithmic management produces opaque, difficult-to-remedy errors and triggers resistance and loss of trust among employees. The study concludes that AI should serve as a decision-support tool under human oversight, a position consistent with emerging regulation such as the European Union’s Artificial Intelligence Act EuropeanCommission,2021. The theoretical contribution extends agency theory by demonstrating that accountability, rather than accuracy, is the binding constraint on automation in managerial roles.
Keywords
Accountability, Algorithmic Bias, Algorithmic Management, Artificial Intelligence, Human Oversight, Managerial Decision-Making