Managers who rely on generative AI (Gen-AI) tools such as ChatGPT for everyday decisions are gradually losing the practical judgment built through experience, according to a University of Bath–led study published online Aug. 24 in the Academy of Management Review.
Researchers link Gen-AI reliance to a decline in managerial judgment
The study, A Process Model of Managerial Phronesis in the Age of Generative AI, was conducted by researchers at the University of Bath along with academics from Ohio, Lausanne, and Cardiff. It examines what happens to “managerial phronesis,” the practical, experience-built wisdom that managers need to make nuanced calls on people and problems.
The researchers found that managers who habitually turn to GenAI for idea generation or problem-solving might come to rely less on their own judgment over time, a process they term “epistemic de-skilling.”
Professor Dirk Lindebaum of Bath’s School of Management said this creates “a significant risk” for organizations, according to Personnel Today. As managers “outsource their thinking to Gen-AI,” he said, they risk relying “less on their own judgment” over time.
The researchers found that this shows up in specific behaviors. For example, managers ask fewer probing questions, seek out fewer outside perspectives, and draw less on real-world experience. Instead, they lean on AI-generated answers that lack the context and moral judgment complex decisions require.
Not all AI use is equally risky
The researchers also examined the conditions under which AI could strengthen managerial judgment rather than weaken it, a process they call “epistemic upskilling.” AI can sharpen judgment, they found, when managers use it to test their own assumptions and reasoning rather than accepting its answers at face value.
Lindebaum said organizations need to “design roles, responsibilities, and workflows” deliberately, so employees “continue developing the human skills that AI cannot replicate.” Simply introducing AI tools, he said, will not automatically improve decision-making or organizational performance.
The judgment gap on the floor
Frontline supervision is the function BPO clients watch most closely. Team leads make dozens of calls daily, de-escalating customer issues, flagging fraud, coaching underperformers, and determining whether a client needs a call rather than an email. These are exactly the skills the study warns can erode with unchecked reliance on AI.
The fix isn’t restricting AI, which is already woven into QA and coaching workflows industry-wide. It’s using it deliberately, considering AI as a sparring partner, not a decision-maker, with managers still accountable for the call.
Meanwhile, clients should ask providers how they develop supervisors alongside AI. Providers that deliberately build judgment will be better positioned to demonstrate that advantage to clients as adoption deepens.
Read more Unity Communications and industry news on our main BPO News page.
Faragher, J. (2026, September 4). Managers ‘outsourcing thinking’ to AI creates risks, study finds. Personnel Today. Retrieved from https://www.personneltoday.com/hr/managers-outsourcing-thinking-to-ai-creates-risks-study-finds/
Lindebaum, D., Balasubramanian, N., Ashraf, M., & Haack, P. (2026, August 24). A PROCESS MODEL OF MANAGERIAL PHRONESIS IN THE AGE OF GENERATIVE AI. Academy of Management Review. Retrieved from https://journals.aom.org/doi/epub/10.5465/amr.2024.0582
