Generative AI and Quality Illusion in Knowledge Creation: AI Hype and the Role of Prestige
30 Apr 2026
11:00 -12:30
Times are shown in local time
Open to: All
Room W2.02 (Cambridge Judge Business School)
Trumpington St
Cambridge
CB2 1AG
United Kingdom
Join our Operations and Technology Management seminar
Speaker: Professor Volodymyr Babich, McDonough School of Business
About the seminar topic
We study the quality illusion, defined as the systematic overestimation of substantive quality that arises when observers rely on signals whose informational content has degraded. Historically, producing polished output required effort correlated with talent, enabling observers to use polish as a proxy for substance. Generative AI disrupts this inference by commoditising polish, severing the connection between surface and substance that gave the fluency heuristic its validity. We develop a model in which authors choose substantive quality and discretionary format effort, while readers decide on evaluation intensity and endorsement. Before AI, format sustains a separating equilibrium: high-talent authors signal through costly polish, and readers screen without evaluation. After AI reduces format costs, separation collapses and pooling emerges. In the short run, belief inertia generates a hype phase in which all market participants perceive AI as beneficial. In the long run, multiple equilibria can arise: a gaming one in which low-talent authors exploit imperfect screening, a self-disciplining one in which reputational concerns induce voluntary quality compliance, and a self-disciplining one in which readers always screen. We characterise how evaluation costs, reputational penalties, and prestige weight influence equilibrium selection, showing that prestige, often viewed as zero-sum status competition, serves a socially valuable coordination function by pushing markets toward self-discipline. Our theoretical predictions align with recent large-scale empirical evidence documenting that writing complexity has become uninformative for AI-assisted academic papers. The analysis yields actionable guidance for institutional design in academic publishing and other knowledge markets.
Speaker bio
Volodymyr Babich is the Heisley Family Chair Professor and Professor of Operations and Analytics at the McDonough School of Business, Georgetown University. He is currently a visiting Professor at London Business School. In the past, he has been a visiting scholar at the Booth School of Business at the University of Chicago, a visiting scholar at the Wharton School at the University of Pennsylvania, and an Assistant Professor at the University of Michigan, Ann Arbor. He earned his PhD in Operations Research from Case Western Reserve University, Weatherhead School of Management. He holds MS degrees in Management Science and in Mathematics.
Professor Babich’s research interests are the interface of operations and finance, supply risk management, supply chain management, energy and sustainability, entrepreneurship, and innovative technologies. His research has been supported by the National Science Foundation and various university and industry grants. His papers have been published in leading journals in Operations Research, Operations Management, and Industrial Engineering. Babich serves as an Associate Editor for Management Science, Manufacturing & Service Operations Management, and Naval Research Logistics, and as a Senior Editor for Production and Operations Management journals. He has served as the Department Editor for Special Issues of Manufacturing & Service Operations Management and as an Editor for the 2024 INFORMS TutORials in Operations Research. He has served twice as the Chair of the MSOM Special Interest Group on the Interface of Finance, Operations, and Risk Management (iFORM).
Register
No registration required. If you have any questions about this seminar, please email Bet Brooke.