Gender Bias in LLM Hiring Decisions: Evidence from a Japanese Context and Evaluation of Mitigation Strategies
For practitioners deploying LLMs in non-Western hiring contexts, this work identifies candidate name as the primary channel of gender bias and reveals a practical challenge with name anonymization due to safety filter incompatibility.
This study confirms a significant pro-female gender bias in LLM hiring decisions within a Japanese context, replicating Western findings. A name-reliance analysis shows that removing the candidate name from the prompt nearly eliminates the bias, while a prompt-level gender-neutrality instruction fails to reduce it.
Large language models (LLMs) are increasingly deployed in hiring workflows, yet most research on gender bias in LLM hiring decisions has focused on English-language, Western-format resumes. This study examines whether pro-female gender bias extends to a Japanese corporate context and evaluates two practical mitigation strategies. Using a counterfactual resume design with 60 Japanese rirekisho-format resumes, 12 name pairs selected on linguistically grounded gender-signal criteria, and five state-of-the-art LLMs (Claude Sonnet 4.6, GPT-4o, DeepSeek-V3, Gemini 2.5 Flash, Llama 3.3 70B), we conducted 43,200 API calls across baseline, prompt instruction, and privacy filter conditions. A crossed random-effects linear mixed model confirms a significant pro-female bias across all five models, replicating Western findings in a non-Western context. A prompt-level gender-neutrality instruction produces no meaningful reduction in bias. A name-reliance analysis formally identifies the candidate name as the primary gender channel: removing the name from the prompt reduces the female effect by nearly its full magnitude. An unexpected incompatibility between the privacy filter and GPT-4o's content safety filter, resulting in a 42% refusal rate, highlights a practical deployment challenge for name anonymization in LLM-assisted recruitment pipelines.