A San Francisco judge cleared a proposed class action to proceed, in a case described as the first to broadly target the algorithms behind AI screening software.
A US federal judge has ruled that Workday must face claims its AI-powered hiring software screened out job applicants at other companies in ways that allegedly broke California law and a federal ban on disability discrimination.
The decision, by US District Judge Rita Lin in San Francisco, keeps alive a case that has become a test of whether the vendors behind algorithmic recruiting tools can be held liable for what those tools do.
Lin rejected Workday’s argument that California anti-discrimination law should not reach it when its software screens applicants who are based outside the state and applying for jobs elsewhere, according to Reuters.
That jurisdictional point mattered to the company, because much of its defence rests on the idea that it merely supplies software and does not itself do the hiring, in California or anywhere else.
The case is Mobley v. Workday, a proposed class action first filed in 2023 by lead plaintiff Derek Mobley, who is over 40, African American, and has a disability. It is described as the first of its kind to broadly target the algorithmic decision-making behind AI screening software, rather than the conduct of any single employer that used it.
Mobley alleges he was rejected from scores of jobs routed through Workday’s platform.
Lin’s ruling was not a clean win for the plaintiffs. She dismissed a claim that the software discriminated against Asian American applicants, while letting separate allegations proceed that it disadvantaged Black applicants, women, and people older than 40.
The surviving claims remain allegations the plaintiffs must still prove; the ruling decides only that they may be argued, not that they are true.
The case had already cleared one significant bar. The court previously allowed it to proceed as a collective action covering applicants aged 40 and over who were turned down through Workday’s platform from 24 September 2020 onward, a window that potentially sweeps in a very large number of rejected job-seekers.
The latest decision widens the legal theories that can now be tested at trial.
Workday rejects the claims in full. The company says they are false, that its AI recruiting tools “do not make hiring decisions in California or anywhere else”, and that the technology “looks only at job qualifications, not protected traits like race, age, or disability”.
It adds that it rigorously tests its products under a Responsible AI programme, a defence that, if it holds, would draw a clear line between the vendor and the employers who configure and act on its software.
That line is precisely what is now in dispute. If a screening tool produces biased outcomes, the open question is whether the company that built it can be treated, in legal terms, as an agent of the employers who deploy it.
The stakes reach well beyond one firm. Algorithmic screening has spread quietly through corporate hiring, and an AI-on-AI war in recruitment has focused minds on jobs lost rather than candidates filtered out before a human ever reads their application.
Many of the largest employers in the US run applicants through one platform or another, often without disclosing that an algorithm, not a recruiter, made the first cut.
A finding that the vendor can be sued would put every one of those tools on notice. Scrutiny of how AI systems handle sensitive personal data has sharpened in parallel.
For now, the practical effect of Lin’s ruling is narrow but real: Mobley v. Workday proceeds, and the discovery phase will start to put numbers behind the allegations.
