AI could undermine managers' judgment unless used carefully, study warns

A new study warns that overreliance on generative artificial intelligence (Gen-AI) tools could undermine managers' ability to make sound judgments. Published in the Academy of Management Review, the research from the University of Bath explores how Gen-AI could erode managers' "managerial phronesis" - the practical wisdom they develop through real-world experience, reflection and human interaction.

While AI like ChatGPT can help complete tasks more quickly, it cannot replace the insights gained from firsthand experience. As managers increasingly outsource their thinking to AI, they may rely less on their own judgment, potentially reducing their ability to learn from experience, think critically and anticipate future needs.

The study identifies "epistemic deskilling," a process where people lose knowledge-related capabilities by outsourcing too much thinking to AI. This is most likely when managers are under time pressure and use AI as a shortcut, rather than engaging deeply with problems. To mitigate this, the researchers suggest using AI for reflection rather than replacement, a process they call "epistemic upskilling."

By using AI to challenge assumptions and test reasoning, managers can develop their judgment. However, this requires managers to fill explanatory gaps themselves, as AI systems often lack clear explanations for their outputs. The study emphasizes that simply introducing AI tools won't automatically improve decision-making or organizational performance.

Instead, organizations must design roles, responsibilities and workflows to ensure employees continue developing the human skills AI cannot replicate.