Quantifying Gender Bias in Large Language Models: When ChatGPT Becomes a Hiring Manager
Name
gerszberg-ninager-meng-eecs-2024-thesis.pdf
Description
Thesis PDF
Size
8.67 MB
Format
Adobe PDF
Checksum (MD5)
e267d17686e460bd7d7b40ad273d9a73
Author(s)
Gerszberg, Nina R.
Advisor(s)
Lo, Andrew
Date Issued
May 2024
Publisher
Massachusetts Institute of Technology
Abstract
The growing importance of large language models (LLMs) in daily life has heightened awareness and concerns about the fact that LLMs exhibit many of the same biases as their creators. In the context of hiring decisions, we quantify the degree to which LLMs perpetuate biases originating from their training data and investigate prompt engineering as a bias-mitigation technique. Our findings suggest that for a given resumé, an LLM is more likely to hire a candidate and perceive them as more qualified if the candidate is female, but still recommends lower pay relative to male candidates.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
In Copyright - Educational Use Permitted
Copyright retained by author(s)
Persistent DSpace Link
