| Question | Short answer | What it means |
|---|---|---|
| Is it decided? | No | Case is in active discovery — no ruling on the merits yet |
| Who's covered | Applicants 40+ | Rejected via Workday's platform since Sept 24, 2020 |
| Legal theory | Disparate impact | Doesn't require proof anyone intended to discriminate |
| Latest ruling | March 2026 | Court confirmed the ADEA protects applicants, not just employees |
| Can I join? | Maybe | Talk to an employment attorney — this article isn't legal advice |
| Does this affect me? | Likely yes | Most large employers use some form of automated resume screening |
01What Mobley v. Workday actually alleges
Derek Mobley applied to more than 100 jobs that used Workday's applicant screening platform and says he was rejected from every one — including some rejections that landed at 1:30 in the morning, which is the kind of detail that makes you suspect no human was involved yet. He was over 40 at the time. He's now the lead plaintiff in a nationwide collective action that could reshape how companies screen resumes.
He alleges Workday's screening tools — its Candidate Skills Match and HiredScore AI features — systematically score down applicants over 40, violating the Age Discrimination in Employment Act (ADEA) and California's Fair Employment and Housing Act (FEHA). The legal twist that makes this case different: he isn't only suing the employers who used Workday's software. He's suing Workday itself, arguing the vendor acted as an "agent" of every company that used its screening tool — a theory that, if it holds, would extend discrimination liability to AI vendors, not just their customers.
A Northern District of California judge dismissed the intentional discrimination claim early on. But she let the disparate impact claim proceed — the doctrine from Griggs v. Duke Power Co. (1971) that allows a facially neutral practice to be challenged if it disproportionately harms a protected group, regardless of intent. As the Miami Law Review breaks down, this is the framework that makes the case viable at all — plaintiffs don't need a smoking-gun email proving intent, just a statistically skewed outcome.
"Disparate impact," in plain English
Nobody has to prove Workday's algorithm was built to discriminate. Mobley only has to show it produced a statistically skewed outcome against workers 40+. Then the burden shifts to Workday to prove the practice was job-related and necessary. That's a much lower bar than proving intent — which is exactly why this case is being watched so closely.
02Where the case stands right now (July 2026)
This isn't a closed story. Coverage that stops at the May 2025 certification is already out of date — the case has moved through two more major rulings since then.
| Date | What happened | |
|---|---|---|
| Feb 2023 | Mobley files suit against Workday | |
| May 2025 | Court grants preliminary certification of a nationwide collective — everyone 40+ who applied via Workday since 9/24/2020 and was rejected can join | |
| Mar 2026 | Court rejects Workday's argument that the ADEA doesn't cover job applicants, only employees — a major procedural win for plaintiffs | |
| Jun 2026 | Court denies plaintiffs' motion to compel Workday's internal bias-testing data (ruled attorney-client privileged) but orders Workday to hand over EEO-1 and OFCCP compliance filings | |
| Jun 2026 | Court grants in part, denies in part Workday's motion to dismiss — the FEHA claim survives | |
| Now | Case is in active discovery, fighting over access to the actual scoring logic behind Workday's tools |
No court has ruled that Workday's algorithm did discriminate — only that the claim is allowed to proceed, and that enough applicants share a common enough experience to sue as a group. But a federal discrimination claim against an algorithm surviving this many rounds of dismissal attempts is itself the story, as outlets like The Talent Architect and CDF Labor Law have both tracked closely.
03How algorithmic bias actually reaches your resume
This is the part most legal coverage skips, and it's the part that matters most if you're job searching. Nobody has to write "reject anyone over 40" into the code. Bias creeps in through proxies — data points that correlate with age without ever mentioning it.
Common proxy signals
- Graduation year — recency-weighted models can quietly penalize older dates
- Unexplained employment gaps — read as risk by models trained on continuous career paths
- Senior historical titles or long tenure — can read as "overqualified," a common proxy for age
- Missing the newest named tools/frameworks — favors recently-trained candidates by default
Why it happens without malice
- Models are trained on "successful hire" data from the past
- That historical data already skews toward who companies favored before
- The model doesn't need to know your age — only signals that correlate with it
- Nobody has to intend the outcome for disparate impact to apply
04What this means if you're job searching right now
You can't audit Workday's model, and you can't wait for a lawsuit to resolve before you need a job. What you can control is what signals your resume actually sends.
- Standardize your dates — consistent MM/YYYY formatting reduces the chance of a parser misreading tenure or flagging inconsistency
- Frame gaps instead of hiding them — a one-line context note reads as intentional to both a parser and a human; an unexplained gap reads as risk to both
- Lead with skills that map to the job description, not just a chronological list of past titles
- Consider trimming pre-2010 education dates if the role doesn't require them — graduation year is one of the clearest age proxies on a resume, and omitting it isn't misrepresentation
- Don't assume silence means "unqualified" — a 1:30 a.m. auto-rejection may reflect a scoring model's blind spot, not your fit for the role
This is not legal advice
If you believe you've experienced discrimination in an automated hiring process, the resume steps above are not a substitute for speaking with an employment attorney. This article explains the case — it doesn't replace legal counsel.
05What this means for employers and universities
For any organization using algorithmic screening — including the university career centers in our own partner network — this case is the clearest signal yet that "the vendor's algorithm did it" is not a liability shield. Regular audits of pass-through rates by age band, documented job-relatedness for any automated scoring criteria, and vendor contracts that explicitly assign responsibility for bias are moving quickly from best practice to baseline expectation.
That's a real consideration for career services offices rolling out AI-assisted tools for students: the same disparate-impact framework at issue in Mobley applies to any automated system that screens or scores people, not just corporate ATS platforms. Institutions evaluating AI hiring or career tools should be asking vendors directly how their models are audited for exactly this kind of proxy bias.
06Frequently asked questions
Is Mobley v. Workday decided? Did Workday lose?
No. As of mid-2026 the case is in the discovery phase. Courts have allowed the disparate impact claim and the nationwide collective to proceed past multiple dismissal attempts, but there has been no ruling on the merits — no finding that discrimination occurred.
Can I join the Mobley v. Workday lawsuit?
The certified collective covers individuals 40 and older who applied for jobs through Workday's platform since September 24, 2020 and were not selected. Whether you qualify and how to join is a legal question best directed to an employment attorney.
Does this mean AI hiring tools are illegal?
No. The case challenges specific alleged outcomes of specific tools under existing anti-discrimination law — it doesn't outlaw automated screening. It does signal that "we used an algorithm" is increasingly not a defense against a discrimination claim.
How do I know if my resume is being screened by AI?
Most mid-size and large employers now use some form of automated screening — industry estimates put Fortune 500 ATS adoption near 98% and AI-assisted screening across large employers above 80%. See our ATS statistics breakdown for the full data picture. If you're applying through a large company's careers portal, assume a scoring model touches your resume before a person does.
What can I actually do about algorithmic bias in hiring?
You can't audit or change the model. You can control the clarity, consistency, and framing of the signals your resume sends it, and you can pursue legal recourse through an employment attorney if you believe you've experienced unlawful discrimination.
You can't audit the algorithm. You can control what it sees.
Get a free ATS score that flags the same formatting and structural issues screening algorithms use as proxies — gaps, inconsistent dates, keyword mismatch — before a bot decides you're not a fit.
Get your free ATS score →This article is for informational purposes only and does not constitute legal advice. Case details are current as of July 2026 and may change as litigation proceeds. If you believe you have experienced employment discrimination, consult a licensed employment attorney.