Large Language Models (LLMs) are proving to be a powerful toolkit for management and organizational research. While early work has largely focused on the value of these tools for data processing and replicating survey-based research, the potential of LLMs for theory building is yet to be recognized. We argue that LLMs can accelerate the pace at which researchers can develop, validate, and extend strategic management theory. We propose a novel framework called Generative AI-Based Experimentation (GABE) that enables researchers to conduct exploratory in silico experiments that can mirror the complexities of real-world organizational settings, featuring multiple agents and strategic interdependencies. This approach is unique because it allows researchers to unpack the mechanisms behind results by directly modifying agents’ roles, preferences, and capabilities, and asking them to reveal the explanations behind decisions. We apply this framework to a novel theory studying strategic exploration under uncertainty. We show how our framework can not only replicate the results from experiments with human subjects at a much lower cost, but can also be used to extend theory by clarifying boundary conditions and uncovering mechanisms. We conclude that LLMs possess tremendous potential to complement existing methods for theorizing in strategy and, more broadly, the social sciences.
Authors
Matteo Tranchero, Cecil-Francis Brenninkmeijer, Arul Murugan, Abhishek Nagaraj
- Acknowledgements & Disclosure
- Corresponding author: Abhishek Nagaraj. We thank participants at the Macro Research Lunch at UC Berkeley-Haas and the 2024 Academy of Management for useful feedback. We acknowledge support from OpenAI in the form of computing credits to use their GPT class of models. All errors are our own. Nagaraj received research support in the form of computing credits from OpenAI. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
- DOI
- https://doi.org/10.3386/w33033
- Pages
- 39
- Published in
- United States of America
Table of Contents
- Introduction 3
- LLMs as tools for Management Research 7
- How Do Researchers Use LLMs? 7
- Management Theorizing With LLMs 8
- A Framework for Simulating Experiments 10
- Implementing GABE in Management Research: An Application 12
- The Theory of the Streetlight Effect 12
- The Online Lab Experiment: Searching Mountains for Hidden Gems 13
- Comparing Human Subjects and AI Agents 15
- Going Beyond the Original Experiment 17
- Varying the Experimental Setup 17
- Varying Group Size 17
- Varying Choice Landscape 18
- Varying Payoff Magnitude 18
- Relaxing theoretical assumptions 20
- Manipulating agent preferences and objectives 22
- Comparing GABE to Direct Elicitation 24
- Discussion 26
- Deeper Dive Into Generative-AI Based Experimentation (GABE) 32
- Key Parameters for Experiments 32
- Excerpt of Script Given to AI Agents 33
- Additional Figures and Tables 35
