The Largest Study of AI Hiring Tools Ever Conducted Found Real Racial Bias. Here’s What That Means for You

With the proliferation of AI tools in nearly every sector, researchers have taken a keen interest in how it impacts employment. Recently, researchers at Stanford, Chapman, and Northeastern universities completed the most comprehensive independent study of AI hiring algorithms to date. The results revealed that more than one in four applications submitted by Black job seekers were sent to positions where the screening algorithm produced outcomes that trigger federal discrimination scrutiny under Title VII. These harrowing results show how pervasive racial bias and discrimination really are during the hiring process. If you have continually been rejected for positions after being screened using an AI-powered tool, you should contact an attorney to discuss what this may mean for you. 

Disclaimer: This article is for information purposes only. McCormack Law Firm is not involved in this study.

What Did the Stanford Study Find?

The research team analyzed more than 4.1 million job applications submitted by 3.4 million people to 1,746 positions across 156 employers. The researchers derived their data from Pymetrics using applications that were submitted between December 2018 and December 2022. During this period, the platform used online games to score applicants.

The research team found that 26% of Black applicants and 15% of Asian applicants applied to jobs where the algorithm’s selection rate for their group fell below 80% of the rate for the most-favored group. Relevant to these findings is the threshold the EEOC uses to flag adverse impact under its “four-fifths rule.” When researchers examined each job posting individually rather than pooling results across the entire dataset, 10.62% of the 1,746 positions showed a measurable adverse impact against Black applicants specifically.

Additionally, the study also revealed something the researchers refer to as “algorithmic monoculture.” Essentially, employers who use the same hiring vendor end up sharing correlated outcomes, meaning a candidate rejected by one company’s algorithm often gets rejected by other companies relying on that same underlying model. The findings suggest that if Black and Asian candidates advanced at the same rate as the most-favored group, roughly 40,000 more candidates would have moved forward to the next stage of hiring.

Is There a Lawsuit Behind This Kind of Bias?

Yes. A closely watched case, Mobley v. Workday, Inc., is currently working through federal court in California and raises very similar issues. Derek Mobley alleges that Workday’s AI-driven applicant screening tools discriminated against him and other job seekers based on race, age, and disability. In June 2026, a federal judge allowed the core discrimination claims to move forward.

Disclaimer: This article is for information purposes only. McCormack Law Firm is not involved in this case.

The Mobley case tests a question the Stanford study raises indirectly: can a company that builds AI hiring software, not just the employer using it, be held legally responsible when that software produces discriminatory outcomes? Courts are still working that question out, and the answer will impact how employees pursue these claims going forward.

Does Getting Rejected by an Algorithm Count as Discrimination?

Not automatically, but it can. Title VII doesn’t require proof that a human being intended to discriminate. Adverse impact claims focus on outcomes, not intent. This means that a facially neutral tool, an algorithm that never considers race directly, can still violate the law if it produces a discriminatory result.

That is precisely the legal theory that the researchers built their analysis around. The four-fifths rule they applied isn’t an academic invention. It’s the same standard the EEOC uses when evaluating whether an employer’s practices disproportionately screen out a protected group.

What Should You Do If You Suspect an AI Tool Screened You Out Unfairly?

There are a few steps that you can take to help preserve your ability to challenge what happened.

• Save the job posting, application confirmation, and any rejection notice you received.
• Note which platform or hiring software the employer used, if that information was disclosed.
• Track how many positions you applied to through the same platform and what happened with each.
• Keep records of your qualifications relative to the position’s stated requirements.

Many people brush this off, worried they’re exaggerating or making something out of nothing. An algorithm rejecting you a dozen times in a row isn’t nothing. Reach out to an employment lawyer before you assume there’s nothing you can do about it.

Contact a San Francisco Employment Lawyer

AI hiring tools are becoming the default at large employers, and the Stanford study shows exactly why that shift deserves scrutiny rather than blind trust. If you believe an algorithm screened you out of job opportunities because of your race or another protected characteristic, McCormack Law Firm can help you understand your options. Contact our discrimination attorneys at (415) 925-5161 for a free consultation. There is no fee unless we win your case, and past results do not guarantee a similar outcome in yours.