LLMs Show Reversal in Hiring Bias: 2024+ Models Favor Black Candidates

A recent study published on arXiv examined fourteen large language models (LLMs) for racial bias in resume screening. The research found that models released in 2023 exhibited a pro-White callback gap, mirroring real-world labor market discrimination. However, all models released in 2024 and later demonstrated either no bias or a significant pro-Black reversal. This pattern was also observed along gender lines, indicating a shift in algorithmic hiring bias across different model generations. AI

IMPACT Recent LLMs show a significant shift towards reducing or reversing racial bias in hiring, suggesting improved fairness in AI-driven recruitment tools.

RANK_REASON The cluster is a research paper published on arXiv detailing findings about LLM bias. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• Kline, Rose, and Walters
• large language models
• ScienceCast

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →