Workday, Artificial Intelligence, and the Future of Employment Discrimination

 Employment Testing: Failing to Make the Grade — Ten Years Later, Part 4

In 2013, this blog warned that commercially developed employment assessments could create widespread legal exposure when used by large numbers of employers.

In 2016, its final post addressed the need to measure the consequences of artificial intelligence and automated decision-support systems.

A decade later, those concerns have emerged in a major federal employment discrimination lawsuit.

Mobley v. Workday, Inc., No. 3:23-cv-00770, pending in the U.S. District Court for the Northern District of California, challenges the use of automated applicant-screening technology on grounds that include alleged discrimination based on age, disability, race, and sex.

The litigation raises questions about employment screening that extend well beyond Workday.

The Allegations

The lead plaintiff, Derek Mobley, alleges that he repeatedly applied for positions through employers using Workday's applicant-screening technology without obtaining employment.

He contends that the company's screening systems unlawfully disadvantaged applicants based on protected characteristics.

Mobley alleges that he is over 40 and has depression and anxiety. His disability-related allegations are particularly relevant to concerns this blog raised more than a decade ago about screening systems that may disadvantage people with mental health conditions.

Workday disputes the allegations.

The case remains pending, and no court has determined that Workday's screening technology actually engaged in unlawful employment discrimination.

Can a Software Vendor Be Liable?

One of the most significant developments occurred in July 2024.

The district court permitted certain discrimination claims against Workday to proceed under a theory that the company could qualify as an employer's agent when performing screening functions delegated by employers.

The decision did not establish that Workday was legally responsible for discriminatory hiring decisions.

It did, however, reject the proposition that the company's status as a third-party software vendor necessarily insulated it from liability.

That distinction is important.

As employment screening becomes increasingly automated, responsibility for applicant selection may be divided among employers, software vendors, consultants, and other entities.

The law must determine which entities exercise sufficient control over screening functions to bear responsibility under the relevant statutes.

This was one of the central concerns raised in this blog's August 2013 post, When the First Domino Falls.

The Scale of the Litigation

In May 2025, the court conditionally certified a nationwide collective action involving applicants aged 40 and older under the Age Discrimination in Employment Act. By October 2026, more than 15,000 individuals had opted into that collective action.

In September 2026, plaintiffs separately moved to certify additional classes, including one involving applicants with disabilities. 

Conditional certification does not establish that any applicant was discriminated against. It permits claims to proceed collectively while issues of eligibility, proof, and liability remain unresolved.

The case illustrates the potential scale of litigation involving a widely used applicant-screening platform.

A single product may be involved in hiring processes conducted by numerous unrelated employers, making its design, implementation, and actual effects important subjects of discovery.

What About Mental Health Conditions?

In June 2026, the court allowed a disability-discrimination claim involving alleged disability-related proxy indicators to survive a motion to dismiss. 

This ruling permits the claim to proceed but does not certify a disability-discrimination class or establish that Workday discriminated against applicants with disabilities. 

The separate request to certify a disability class remains pending.

This issue deserves particular attention.

An automated system may not explicitly ask whether someone has depression, anxiety, or another mental health condition. Nevertheless, it may evaluate information—such as certain employment gaps—that could be associated with disability.

Whether such criteria violate the ADA depends on the facts and applicable legal standards, including whether the screening practice unlawfully excludes qualified individuals with disabilities.

The existence of a correlation alone does not establish discrimination.

But it does provide a reason to examine what the system measures, why those measures were selected, and how they affect applicants.

These are substantially the same questions this blog raised when examining personality assessments, statistical prediction, and employment discrimination in 2013.

The Role of Federal Regulators

The Equal Employment Opportunity Commission issued technical assistance in 2022 and 2023 discussing the application of existing employment discrimination laws to AI-assisted hiring.

Those materials were removed from the agency's website in early 2025 following changes in federal administrative policy.

Their removal did not repeal the ADA, the Age Discrimination in Employment Act, or Title VII.

Nor did it eliminate employers' obligations to use lawful selection procedures.

The underlying statutory questions remain, regardless of the technology used to evaluate applicants.

Ten Years Later, the Questions Remain

The Workday litigation has not yet answered whether the company's particular screening systems violate employment discrimination laws.

But it demonstrates why the concerns raised in Employment Testing: Failing to Make the Grade remain relevant.

The same underlying issues persist:

Can a screening system disadvantage people with disabilities without explicitly identifying their conditions?

Can software trained on historical employment information reproduce existing bias?

How should employers establish that an assessment measures legitimate job qualifications?

And who bears responsibility when a third-party system allegedly discriminates against applicants across numerous employers?

Artificial intelligence has made employment screening faster and potentially more sophisticated.

It has not eliminated the need for lawful, transparent, and properly evaluated hiring practices.

Ten years later, the question is not whether computers can help employers make hiring decisions. It is whether those decisions are fair, legally defensible, and genuinely related to an applicant's ability to perform the job.

That remains the test employment screening must pass.

When the First Domino Falls: The Risks of Outsourcing Hiring Decisions

 Employment Testing: Failing to Make the Grade — Ten Years Later, Part 3

In August 2013, this blog published When the First Domino Falls: Consequences to Employers of Embracing Workforce Assessment Solutions.

The post raised a concern that has become even more significant with the rapid expansion of artificial intelligence in employment screening.

What happens when thousands of employers rely on the same commercially developed assessment system—and that system turns out to violate employment discrimination laws?

The potential consequences are not necessarily limited to one employer, one workplace, or one group of job applicants.

The same problem may be reproduced across thousands of hiring decisions and numerous employers.

The Attraction of Outsourcing

Employers have understandable reasons for using third-party screening products.

Automated systems promise faster processing, consistent evaluation standards, reduced administrative costs, and improved identification of promising candidates.

For companies receiving thousands of applications, those benefits can be substantial.

But outsourcing a hiring function does not necessarily transfer the employer's legal responsibilities to the software company.

Employers remain responsible for complying with applicable employment discrimination laws, even when they rely on vendors to evaluate applicants.

That raises a basic question:

What does the employer actually know about the system it is using to decide who receives an employment opportunity?

Validation Is Not Simply a Vendor's Responsibility

A software company may advertise that its assessment has been scientifically validated or that its algorithms predict successful employees.

But validity is not necessarily universal.

A screening procedure developed using one employer's workforce may not perform in the same way for another employer, occupation, or applicant population.

Likewise, a test that predicts one measure of workplace performance may be less effective—or inappropriate—for evaluating a different position.

Employers should therefore examine whether a selection procedure is suitable for its intended use and whether available evidence supports its application to the particular jobs being filled.

They should also consider whether the system creates discriminatory effects or accessibility barriers.

Accepting a vendor's assurances without examining these questions may leave important legal and practical risks unaddressed.

The Problem of Shared Liability

The 2013 post also questioned whether the widespread use of common assessments could produce systemic legal exposure.

That concern extends beyond employers.

Software developers, assessment companies, consultants, and other vendors may face legal claims depending on their conduct, their relationships with employers, and the applicable laws.

The allocation of liability is not identical under every employment discrimination statute. Nor does the existence of a vendor agreement automatically establish the vendor's responsibility to applicants.

But the growing reliance on shared screening systems raises questions about who designs the selection criteria, who controls the decision-making process, who possesses relevant records, and who can correct discriminatory results.

These questions become especially important when employers cannot fully explain why a particular applicant was rejected.

Can Indemnification Solve the Problem?

The original blog also examined contractual indemnification—agreements under which screening companies may promise to reimburse employers for certain legal claims.

Such agreements can be useful.

But indemnification does not prevent discrimination, erase statutory obligations, or necessarily compensate every employer for every loss.

A vendor might dispute whether a claim falls within an indemnification provision. Coverage may be subject to limitations or exclusions. Insurance may be insufficient for unusually widespread litigation.

And reputational harm, lost applicants, and disruption to hiring operations may be difficult to repair.

The better approach is to evaluate the screening system before problems arise and continue monitoring its operation.

A Warning That Has Become More Relevant

The risks discussed in 2013 have grown alongside the reach of automated employment systems.

A questionable assessment used at a single workplace might affect dozens or hundreds of applicants.

A similar assessment integrated into a widely used hiring platform could affect vastly more people.

That does not mean every shared screening system is discriminatory or that all customers of a vendor automatically face identical liability.

It does mean the consequences of a poorly designed or inadequately evaluated system may extend well beyond its original developer.

The first domino may be a single applicant's challenge. The larger question is how many other hiring decisions depended on the same system.

Next: The Workday litigation and the questions courts are now confronting about automated hiring and vendor responsibility.

When Algorithms Discriminate Without Asking About Disability

 Employment Testing: Failing to Make the Grade — Ten Years Later, Part 2

Suppose an employer uses an algorithm to rank job applicants. The system does not ask about race, age, gender, or disability. It uses employment history, education, work experience, and other seemingly neutral information.

Does that mean the system cannot discriminate?

Not necessarily.

One of the central concerns raised in this blog more than a decade ago was that employment-screening systems can disadvantage protected groups without explicitly identifying—or intentionally targeting—those groups.

The problem becomes particularly important when automated systems use information that may serve as a proxy for a protected characteristic.

The Hidden Meaning of Neutral Data

Consider employment gaps.

An applicant may have been out of the workforce while receiving treatment for depression, recovering from a psychiatric hospitalization, or managing another disability. Another may have taken time away to care for a family member.

An automated screening tool might treat an extended employment gap as a negative indicator, even when the applicant possesses all the qualifications necessary for the position.

Similarly, an assessment might penalize applicants whose work histories involve frequent job changes, unconventional educational paths, or periods of reduced employment.

These characteristics may have legitimate relevance to some positions. But their relevance should be established rather than assumed.

The central concern is not whether an algorithm uses a particular data point. It is whether the criterion is meaningfully related to job performance and whether its use unlawfully disadvantages qualified applicants.

Learning From Yesterday's Employees

In its August 2013 discussion of Kroger and Kronos, this blog examined an assessment system that reportedly used neural networks to identify characteristics associated with an employer's successful employees.

The problem was straightforward: What happens when an algorithm learns from a workforce shaped by past hiring practices?

If an employer historically favored particular educational backgrounds, employment patterns, personality characteristics, or demographic groups, a system trained on those employees may reproduce those preferences.

The resulting recommendations may appear objective because a computer generated them.

But mathematical consistency is not the same as fairness.

A system can apply the same rule to every applicant and still produce discriminatory outcomes.

Different Legal Questions

Not every statistical correlation involving a protected group constitutes unlawful discrimination.

Under Title VII, an employment practice producing a disparate impact on a protected group may require the employer to demonstrate that the practice is job-related and consistent with business necessity. Other requirements and defenses may also apply.

The ADA has separate provisions addressing qualification standards and selection criteria that screen out individuals with disabilities, subject to the applicable statutory standards.

And an algorithm that indirectly disadvantages applicants with disabilities does not necessarily constitute a prohibited medical examination.

These distinctions matter. The legality of an automated screening system depends on its design, operation, justification, and actual effects.

What Should Employers Be Asking?

Before using an automated applicant-screening tool, employers should understand:

• What information does the system evaluate, and why?

• How were its criteria selected and validated?

• Do those criteria accurately measure the abilities required for the particular job?

• Could they disadvantage applicants with disabilities or other protected characteristics?

• Can applicants obtain appropriate accommodations?

• Are the results monitored over time, including after software changes?

These are not merely questions for software developers. They are questions for employers, lawyers, human resources professionals, and executives responsible for hiring practices.

In 2016, this blog's final post called attention to the need to measure and assess artificial intelligence and decision-support systems.

Ten years later, that responsibility is no less important.

An algorithm's failure to ask about disability does not establish that its decisions are free from disability discrimination.

Next: Why using a third-party screening vendor may create legal risks far beyond a single employer.

The New Technology, the Same Old Discrimination: AI Hiring and Mental Health

Employment Testing: Failing to Make the Grade — Ten Years Later, Part 1 

Ten years ago, the final post on this blog addressed the growing influence of artificial intelligence and decision-support systems on employment, education, and other important opportunities. The concern was not simply whether computers could make decisions more efficiently than people. It was whether anyone could adequately measure the consequences of those decisions. 

A decade later, that question has become considerably more urgent.

 

Employers increasingly use automated applicant-screening systems, predictive assessments, and AI-assisted recruiting tools to evaluate job candidates. These technologies promise greater efficiency, consistency, and objectivity. 

But they also raise an old question in a new form: Are these systems identifying people who can perform a job, or are they excluding qualified applicants because their characteristics do not fit an algorithm's definition of the ideal employee? 

The Problem Was Never Limited to Technology 

In August 2013, this blog examined employment assessments marketed by Kronos and its subsidiary Unicru in Kroger and Kronos: Chaos and Disorder. Among other concerns, the post questioned whether assessments using personality characteristics and historical employee data could disadvantage applicants with mental health conditions. 

The underlying problem remains relevant today.

Consider an applicant who has experienced major depression but is fully qualified for a position. Another applicant may have post-traumatic stress disorder or bipolar disorder that has been successfully treated. A third may be autistic and possess exceptional technical skills but communicate differently from other candidates. 

An employment assessment that rewards a narrow range of emotional, social, or behavioral characteristics may disadvantage these applicants without reliably measuring their ability to perform the essential functions of the job. 

An algorithm does not have to ask whether someone has a psychiatric diagnosis to create such a problem. 

The ADA Still Matters 

The Americans with Disabilities Act (ADA) restricts disability-related inquiries and medical examinations before an employer makes a conditional job offer. It also prohibits certain qualification standards and selection criteria that unlawfully screen out qualified individuals with disabilities. These are related but distinct protections. 

In Karraker v. Rent-A-Center, 411 F.3d 831 (7th Cir. 2005), the court held that the use of the Minnesota Multiphasic Personality Inventory in an employment selection process constituted a prohibited medical examination because the test was designed, at least in part, to reveal mental illness. 

That decision does not mean every personality assessment is unlawful. Nor does it mean that every algorithm that evaluates behavior is a medical examination. But it does establish that calling something a personality test does not necessarily determine its legal status. 

Similarly, an assessment that is not a medical examination may still raise ADA concerns if it unlawfully screens out qualified applicants with disabilities. 

Artificial Intelligence Does Not Eliminate Human Bias 

An algorithm may identify patterns in employees previously labeled successful. But what if the employer's definition of success reflects longstanding hiring preferences, organizational culture, or discriminatory practices? 

The system may reproduce those preferences on a much larger scale. An applicant can be rejected without anyone knowing—or asking—whether a disability affected the assessment. 

The technology may be new. The underlying concern is not. 

Employers should have to demonstrate that their selection systems measure qualifications relevant to the job, rather than characteristics that unnecessarily exclude people with disabilities. That requires more than trusting the technology or accepting the vendor's assurances. It requires examining the actual selection process, its relationship to job performance, and its effects on applicants with protected characteristics. 

Ten years after this blog last examined artificial intelligence, the essential question remains: Are we using technology to identify talent—or to automate exclusion? Next: How seemingly neutral algorithms can disadvantage applicants without ever asking whether they have a disability.