analiza aiqbench
This study reveals that Large Language Models used for resume screening exhibit severe intersectional bias, systematically disadvantaging candidates with Black-associated names and female-associated identifiers. The findings demonstrate that embedding models trained on general corpora fail to mitigate historical employment prejudices when applied to high-stakes hiring decisions. These results underscore the critical need for rigorous auditing of AI tools before their deployment in automated recruitment workflows to prevent algorithmic discrimination.
Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval
arXiv:2407.20371v3 Announce Type: replace-cross Abstract: Artificial intelligence (AI) hiring tools have revolutionized resume screening, and large language models (LLMs) have the potential to do the same. However, given the biases which are embedded within LLMs, it is unclear whether they can be used in this scenario without disadvantaging groups based on their protected attributes. In this work, we investigate the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection. Using that framework, we then perform a resume audit study to determine whether a selection of Massive Text Embedding (MTE) models are biased in resume screening scenarios. We simulate this for nine occupations, using a collection of over 500 publicly available resumes and 500 job descriptions. We find that the MTEs are biased, significantly favoring White-associated names in 85.1\% of cases and female-associated names in only 11.1\% of cases, with a minority of cases showing no statistically significant differences. Further analyses show that Black males are disadvantaged in up to 100\% of cases, replicating real-world patterns of bias in employment settings, and validate three hypotheses of intersectionality. We also find an impact of document length as well as the corpus frequency of names in the selection of resumes. These findings have implications for widely used AI tools that are automating employment, fairness, and tech policy.
Abstract:Artificial intelligence (AI) hiring tools have revolutionized resume screening, and large language models (LLMs) have the potential to do the same. However, given the biases which are embedded within LLMs, it is unclear whether they can be used in this scenario without disadvantaging groups based on their protected attributes. In this work, we investigate the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection. Using that framework, we then perform a resume audit study to determine whether a selection of Massive Text Embedding (MTE) models are biased in resume screening scenarios. We simulate this for nine occupations, using a collection of over 500 publicly available resumes and 500 job descriptions. We find that the MTEs are biased, significantly favoring White-associated names in 85.1\% of cases and female-associated names in only 11.1\% of cases, with a minority of cases showing no statistically significant differences. Further analyses show that Black males are disadvantaged in up to 100\% of cases, replicating real-world patterns of bias in employment settings, and validate three hypotheses of intersectionality. We also find an impact of document length as well as the corpus frequency of names in the selection of resumes. These findings have implications for widely used AI tools that are automating employment, fairness, and tech policy.
Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
arXiv:2407.20371 [cs.CY]
(or arXiv:2407.20371v3 [cs.CY] for this version)
https://doi.org/10.48550/arXiv.2407.20371 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.1609/aies.v7i1.31748 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 332 cuvinte
Subiecte conexe
Unpacking AI safety for enterprises
Cohere · 4 iun. 2026
Fairness Is Not Enough: Auditing Competence and Intersectional Bias in AI-powered Resume Screening
arXiv · 29 iul. 2026
Intersectional Fairness in Large Language Models
arXiv · 15 sept. 2026
A Computational Audit of Demographic Association Encoding in ClinicalBERT Language Predictions
arXiv · 15 iun. 2026