AI Alignment Amplifies the Role of Race, Gender, and Disability in Hiring Decisions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Ze, Shen, Guobin, Thaler, Michael
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917493364228096
author Wang, Ze
Shen, Guobin
Thaler, Michael
author_facet Wang, Ze
Shen, Guobin
Thaler, Michael
contents Humans increasingly delegate decisions to language models, yet whether these systems reproduce or reshape human patterns of discrimination remains unclear. Here we run a large-scale study to analyse whether language models use demographic information in hiring decisions. We show, across 27 models and 177 occupations, that language models give female and Black candidates hiring advantages relative to otherwise-comparable male and white candidates, while giving disabled candidates disadvantages. The differences are meaningful in magnitude: the role of race, gender, and disability status is comparable to six months to one year of additional education. Post-training alignment is the primary driver: relative to matched pre-trained models, alignment amplifies advantages for female and Black candidates by 325% and 330%, and disadvantages for disabled candidates by 171%. Compared with previous human correspondence studies, language models reverse the direction of racial discrimination, attenuate the disability penalty, and amplify the female advantage by 190%. Alignment changes how models use qualification signals: alignment increases returns to skills and work experience overall, but relatively more so for female and Black candidates. Meanwhile, the absence of qualification signals harms marginalised groups more, particularly for disabled candidates, differences that may explain the asymmetry of alignment effects across groups we observe.
format Preprint
id arxiv_https___arxiv_org_abs_2605_13866
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle AI Alignment Amplifies the Role of Race, Gender, and Disability in Hiring Decisions
Wang, Ze
Shen, Guobin
Thaler, Michael
Computers and Society
General Economics
Economics
Humans increasingly delegate decisions to language models, yet whether these systems reproduce or reshape human patterns of discrimination remains unclear. Here we run a large-scale study to analyse whether language models use demographic information in hiring decisions. We show, across 27 models and 177 occupations, that language models give female and Black candidates hiring advantages relative to otherwise-comparable male and white candidates, while giving disabled candidates disadvantages. The differences are meaningful in magnitude: the role of race, gender, and disability status is comparable to six months to one year of additional education. Post-training alignment is the primary driver: relative to matched pre-trained models, alignment amplifies advantages for female and Black candidates by 325% and 330%, and disadvantages for disabled candidates by 171%. Compared with previous human correspondence studies, language models reverse the direction of racial discrimination, attenuate the disability penalty, and amplify the female advantage by 190%. Alignment changes how models use qualification signals: alignment increases returns to skills and work experience overall, but relatively more so for female and Black candidates. Meanwhile, the absence of qualification signals harms marginalised groups more, particularly for disabled candidates, differences that may explain the asymmetry of alignment effects across groups we observe.
title AI Alignment Amplifies the Role of Race, Gender, and Disability in Hiring Decisions
topic Computers and Society
General Economics
Economics
url https://arxiv.org/abs/2605.13866