Primary Investigator (PI) Name
Md Abdullah Al Hafiz Khan
Department
CCSE – Computer Science
Abstract
AI-driven automated hiring tools are reshaping how companies find talent, but they often reproduce the hidden biases embedded in their training data. Our project, PRISM (Proxy Recognition and Inclusion Scoring Method), investigates how subtle demographic signals, specifically first names associated with gender and race, influence AI resume screening even when candidates have identical qualifications. We built a controlled dataset of resumes that are identical in every way except for the applicant's first name, with each resume using a racially neutral surname to isolate how first names alone affect scoring. We tested these resumes against job postings in technology, healthcare, and law to capture different professional contexts. Our research compares two AI systems: a Sentence-BERT model that measures how well resumes match job descriptions, and a large language model designed to imitate how a human recruiter would score candidates. We look at how scores, rankings, and pass/fail cutoffs vary across demographic groups, using statistical tests to measure whether the differences are significant and meaningful. Specifically, we examine whether AI systems use names as proxies for identity, potentially leading to scoring gaps between candidates with identical qualifications. These tools are already being used to screen job applicants, making this issue directly relevant to people's career prospects. PRISM goes beyond identifying problems by testing solutions like masking demographic proxies to see what actually reduces bias. We believe responsible AI innovation must prioritize equity and transparency. As AI hiring tools spread rapidly across industries, PRISM serves as an initial investigation into embedded bias. By examining how names influence algorithmic scoring, this research raises vital awareness about whether AI hiring advances fairness or simply automates existing discrimination.
Disciplines
Artificial Intelligence and Robotics | Computational Engineering | Data Science
PRISM Powerpoint
Included in
Artificial Intelligence and Robotics Commons, Computational Engineering Commons, Data Science Commons
PRISM (Proxy Recognition and Inclusion Scoring Method)
AI-driven automated hiring tools are reshaping how companies find talent, but they often reproduce the hidden biases embedded in their training data. Our project, PRISM (Proxy Recognition and Inclusion Scoring Method), investigates how subtle demographic signals, specifically first names associated with gender and race, influence AI resume screening even when candidates have identical qualifications. We built a controlled dataset of resumes that are identical in every way except for the applicant's first name, with each resume using a racially neutral surname to isolate how first names alone affect scoring. We tested these resumes against job postings in technology, healthcare, and law to capture different professional contexts. Our research compares two AI systems: a Sentence-BERT model that measures how well resumes match job descriptions, and a large language model designed to imitate how a human recruiter would score candidates. We look at how scores, rankings, and pass/fail cutoffs vary across demographic groups, using statistical tests to measure whether the differences are significant and meaningful. Specifically, we examine whether AI systems use names as proxies for identity, potentially leading to scoring gaps between candidates with identical qualifications. These tools are already being used to screen job applicants, making this issue directly relevant to people's career prospects. PRISM goes beyond identifying problems by testing solutions like masking demographic proxies to see what actually reduces bias. We believe responsible AI innovation must prioritize equity and transparency. As AI hiring tools spread rapidly across industries, PRISM serves as an initial investigation into embedded bias. By examining how names influence algorithmic scoring, this research raises vital awareness about whether AI hiring advances fairness or simply automates existing discrimination.