A new study has found that AI hiring tools discriminate against Black and Asian job applicants. The study, “Algorithmic Monocultures in Hiring,” was conducted by researchers from Stanford University, Chapman University, and Northeastern University. It highlighted how 90% of US employees use AI screening tools to find top applicants. But how does this affect people of colour looking for new roles?

AI hiring tools affecting Black and Asian job applicants

The study found that 26% of Black applicants and 15% of Asian applicants applied to roles where the AI tool discriminated against their racial group. This means that if the AI system had recommended Black and Asian candidates at the same rate as white applicants, 40,000 more of their applications would have advanced to the next stage.

Lawsuit against Workday’s AI hiring tool

This comes after a California judge recently upheld more recent claims against Workday in its anti-discrimination lawsuit. Judge Rita. FA Lin upheld the plaintiffs’ more recent claims and refused Workday’s effort to dismiss them, according to Reuters. This included Hughes’ disability discrimination claim, which alleged it could proceed under the Americans with Disabilities Act.

This is part of a collective-action lawsuit alleging that its AI-powered hiring tools discriminate against workers age 40 and older. The lawsuit was first filed in February 2023 by Derek Mobley, a Black, disabled man in his 40s, who claims he applied for over 100 jobs without securing a role, according to ITPro.

California Implements New Rules On Employers Using AI

On October 1, 2025, California will implement new rules on Automated Decision Systems (ADS). These regulations will enforce strict compliance obligations on employers that use AI in hiring, promotions, evaluations, and other employment decisions, according to Holland & Hart.

According to the regulations, any automated technology that significantly affects employment outcomes is subject to California’s Fair Employment and Housing Act (FEHA). Employers must:

• Ensure AI and algorithmic tools comply with FEHA’s anti-discrimination standards,
• Even if there is no intention to discriminate, take precautions against disparate impact
• Continue to be accountable for skewed outcomes from outside vendors who can be considered “agents” of the company.
• For four years, save the records of the bias audit, inputs, outputs, and decision logic.
• Build a possible affirmative defense by doing result monitoring and bias audits

Image: Vitaly Gariev

The post New Findings Highlight Racial Bias in AI-Powered Hiring Decisions appeared first on POCIT. Telling the stories and thoughts of the underrepresented in tech..