Workday AI bias lawsuit in California expands as FEHA nexus is recognized

A federal judge in San Francisco has ruled that California's Fair Employment and Housing Act (FEHA) can apply to nonresident job applicants in the ongoing Workday AI discrimination lawsuit. The decision allows state-level discrimination claims to proceed against Workday, widening the potential scope of liability for AI-driven hiring tools used nationwide.

What happened

The lawsuit, filed as a proposed class action in February 2023, alleges that Workday's applicant screening technology discriminates based on age, race, and disability under the Age Discrimination in Employment Act (ADEA), Title VII of the Civil Rights Act, and the Americans with Disabilities Act (ADA). Later amendments added claims under FEHA. The lead plaintiff, Derek Mobley, said that since 2017 he had applied for over 100 positions at companies using Workday tools and was rejected for every one, often with immediate or automated rejections.

The collective action was approved in May 2025, despite Workday's concern that the class could be in the hundreds of millions of people. Judge Rita F. Lin rejected that argument, stating that "allegedly widespread discrimination is not a basis for denying notice."

Workday argued that FEHA should not apply to nonresident class members, but Judge Lin found a "sufficient nexus" because the AI tools are "designed, developed, maintained, and controlled" from the company's California headquarters, and the screening and rejection processes originate in California. Workday did not challenge those assertions.

The court allowed ADA claims to proceed for one plaintiff, connecting her rejections to screening based on medical-related leave and treatment patterns. However, the judge dismissed a race-based disparate impact claim and a claim that Workday is liable as an employer for using the challenged procedures.

Why AI builders should care

This ruling is significant for any team building or deploying AI recruitment tools, especially those with operations or customers in California. The decision establishes that state anti-discrimination laws can reach AI tool vendors even when those vendors are not the direct employer, provided their software development and operation occur within the state.

For AI builders, the key takeaway is that jurisdictional defenses are narrowing. If your AI screening tool is designed, trained, or operated from California, you may face FEHA claims from applicants anywhere in the country. The nexus standard does not require the applicants to be California residents.

Practical implications

For vendors, the ruling signals that state-level compliance must be factored into AI screening product design. Human oversight and transparent, job-relevant screening criteria are no longer just best practices; they are legal requirements under active litigation.

For employers using AI hiring tools, the case underscores the need to document your oversight processes and ensure that protected traits are not used as proxies in automated screening. Workday maintains that its "AI recruiting tools don't make hiring decisions" and that "customers maintain full control of their hiring processes," but the court has allowed claims to proceed regardless.

The potential scope of the class action is notable. If the class grows to include hundreds of millions of applicants, as Workday speculated, the financial and operational consequences for AI hiring tool providers could be enormous.

Caveats

This is an ongoing litigation. The ruling is a motion to dismiss decision, not a final judgment. Workday has denied the allegations and stated it looks forward to addressing them in court. Additional appellate review could alter the claims that survive, and the case may be settled before trial.

The race-based disparate impact claim and the employer liability claim were dismissed, narrowing some theories of liability. However, the core ADA and FEHA claims remain active, and the nexus standard could influence future cases.