AI Screening Tools Rate White Men Without Degrees Lowest, Study Finds

White men without undergraduate degrees receive the lowest average ratings from large language models, according to a paper by Maxim Chupilkin. The study ran full-factorial vignette experiments with 18 models from 12 developer groups, varying gender, race, age, citizenship, and education while holding stated financial or occupational circumstances constant within each setting. Each model evaluated all 32 profiles ten times per setting, yielding 17,280 ratings. Averaging over models, age, and citizenship, White men without degrees came out lowest among eight gender-race-education groups: 75.87 in credit, 92.71 in hiring, and 86.62 in rental housing on a 0-100 scale. Black women with degrees received the highest average ratings, with corresponding gaps of 2.94, 3.66, and 3.56 points. Separate attribute effects favored women, Black applicants, and degree holders in all three settings, and White men without degrees had the lowest or second-lowest mean in 46 of 54 model-scenario combinations (85.2%), a pattern the paper connects to evidence of growing economic and health vulnerabilities among White men without degrees — highlighting a group whose disadvantages can be obscured by broad racial or gender categories.