Is AI Everywhere (in Leadership Research)? Demand Effects and an Intervention Against AI Hype
Type
conference paper
Date Issued
2025
Author(s)
Research Team
PLAID Lab
Abstract
Research on Artificial Intelligence (AI) and its use has proliferated, matching the pace of technological advances in the field of AI and its presence in business and science. While encouraging, I propose that this rapid, recent, and far-reaching interest in AI is also affecting ostensibly unrelated research. Integrating the idea of “noise” (i.e., unwanted variance) from behavioral economics (Kahneman et al., 2021), I theorize and test if this “AI hype” has profound, unintended effects on aspects of research we have not yet considered by adding noise and clouding our conclusions. In a mixed methods program of research, I first introduce the idea of AI demand effects (i.e., participants infer the response they think researchers expect from them and behave in line with these expectations—or at least differently than they otherwise would have) as a discovery in leadership humor research (N = 402, Study 1) in which leaders were described as using an AI tool for time management (i.e., RescueTime) in their annual evaluations. I then replicate similar effects in a study of leadership and motherhood (Ntotal = 384, Study 2) which used an AI tool (elai) to generate videos of ostensible job applicants to be hired. To conclude, I consider key predictors and moderators of these effects, as well as a theory-based intervention. This work contributes to the science of leadership science as well as modern methods in social science more broadly.
Language
English (United States)
Keywords
Artificial Intelligence
AI
KI
demand effects
social science research
intervention
methods
HSG Classification
contribution to scientific community
Refereed
Yes
Event Title
Interdisciplinary Perspectives on Leadership Symposium
Event Location
Greece
Event Date
7-10 May 2025
Subject(s)
Division(s)