Workday Wins Partial Victory in AI Hiring Bias Lawsuit—But Employer Accountability Risk Remains
Key Takeaways
- ▸Workday's internal bias-testing data is protected by attorney-client privilege and will not be produced in discovery, but the company must disclose its EEO-1 and OFCCP compliance filings.
- ▸Plaintiffs retain access to customer applicant data and employment outcome statistics, allowing them to build a disparate impact case without Workday's proprietary development secrets.
- ▸Employers using AI hiring tools face heightened accountability: they must understand and disclose the automation logic and bias-risk tolerance of tools they deploy, or risk their own Title VII, ADEA, and ADA liability.
Summary
On May 29, 2026, Magistrate Judge Laurel Beeler issued a discovery order in Mobley v. Workday, Inc. that shields Workday's internal bias-testing data behind attorney-client privilege, preventing plaintiffs from accessing the company's own evaluation of potential discrimination risks in its AI hiring tools. However, the ruling requires Workday to produce its EEO-1 and OFCCP filings and does not shield customer applicant data, keeping the core claims alive. Derek Mobley and four other plaintiffs allege that Workday's AI tools—including Candidate Skills Match, Spotlight, and Fetch—have systematically rejected hundreds of qualified applicants through proxy variables that reproduce and amplify historical hiring disparities based on race, age, and disability status.
The case highlights a critical tension in AI hiring regulation: while tool vendors like Workday can protect their development secrets, employers who deploy these systems without understanding or disclosing their automation logic bear direct legal exposure under Title VII, ADEA, and ADA. The discovery order does not insulate Workday from liability on the merits; rather, it narrows the evidentiary battleground to employment outcomes and algorithmic transparency at the customer level.
This ruling signals that courts will not grant blanket protection to AI hiring vendors based on privilege claims alone. Employers using these tools must now demonstrate they have conducted their own due diligence on bias risk and made informed choices about deployment—failure to do so could expose them to disparate impact liability even if Workday's bias-testing protocols remain confidential.
- The lawsuit alleges Workday's AI tools (Candidate Skills Match, Spotlight, Fetch) use proxy variables that systematically exclude African American, over-40, and disabled applicants—a pattern that could indicate algorithmic bias rather than neutral selection criteria.
