Ke (Kay) Fang (BA ’24) (he/him)
Leverage Large Language Models as GPT4 for Social and Behavioral Study
In this project, Kay explores the potential for social and behavioral scientists to harness advanced artificial intelligence models, specifically large language models (LLMs) like GPT-4, in their research. He critically examined the emerging literature on LLMs’ capacity for human-like judgment and the feasibility of replacing human participants with these models. Utilizing GPT-4, he generated and validated research materials for his own empirical study. The project concludes with a discussion of future research directions and more generally important considerations for integrating LLMs like GPT-4 into social and behavioral science research.