AI Hiring Tools Yield Racial Bias and Systemic Rejection; 26% Black & 15% Asian

A large-scale study of 3.4 million job applicants found that a widely-used AI hiring tool systematically discriminates against Black and Asian candidates, with 26% of Black applicants and 15% of Asian applicants facing bias in positions where they applied. The research also revealed that reliance on a single vendor's algorithm creates "systemic rejection," where applicants rejected by one employer using the same AI tool are more likely to be rejected across multiple employers, with 10% of applicants submitting four applications being rejected by all of them. The study suggests that while bias in the pooled data appears minimal, examining individual job positions reveals substantial racial disparities that violate U.S. employment law under the EEOC's "four-fifths rule."