Disparate Impact Shift Puts HR in a Compliance ‘Gray Area’

The Trump administration’s pivot away from disparate impact enforcement is creating uncertainty for employers, but HR leaders should not interpret the development as a green light to abandon adverse-impact analysis, validation studies, or other safeguards around employment decisions, experts said.

In June, the U.S. Department of Justice issued an opinion concluding that the Equal Employment Opportunity Commission’s (EEOC’s) longstanding interpretation of disparate impact liability is unconstitutional. In addition, the EEOC has proposed rescinding the Uniform Guidelines on Employee Selection Procedures (UGESP), the federal framework employers have relied on for 50 years to evaluate and validate employment tests and selection procedures.

The moves are not binding on federal courts, do not change the law nor eliminate the statutory basis for disparate impact claims. But the shift creates a complicated environment for organizations trying to balance business needs, legal compliance, and equal employment opportunity practices.

Clarifying that uncertainty will be the focus of a Nov. 16 session at BLUEPRINT for I&D in Nashville. The session “AI in Hiring: Adverse-Impact Law, Vendor Audits, and What Holds Up in Court,” will offer a practitioner perspective for HR leaders navigating the intersection of AI vendors, adverse-impact law, and the UGESP.

“We’re in one of those times where there is a lot of gray area,” said Lauren Winans, CEO and principal HR consultant at Next Level Benefits HR Consulting in Pittsburgh. “HR professionals could make changes to internal practices and unintentionally do something with its own consequences relating to discrimination.”

Lynn Clements, senior director of people insights at Berkshire Associates in Columbia, Md., posed the question, “Should employers change how they evaluate seemingly neutral employment practices, including tests and other selection procedures?”

At least for now, the answer is generally "no,” she said.

Seminar: U.S. Employment Law & Compliance

Enforcement Has Shifted, but the Law Has Not

Disparate impact involves a facially neutral employment practice that disproportionately excludes or disadvantages members of a protected group. Unlike disparate treatment, it does not require an employer to intentionally discriminate. The theory was established by the U.S. Supreme Court in 1971 and codified by Congress in the Civil Rights Act of 1991.

The UGESP, adopted in 1978, provided employers with recognized approaches for validating selection procedures. Its principles have been used for decades to evaluate cognitive and personality tests, physical ability tests, work samples, structured interviews, promotional procedures, and other selection tools, Clements explained.

The DOJ opinion — requested by EEOC Chair Andrea Lucas — argues that the agency’s historical approach places excessive emphasis on statistical disparities and insufficient emphasis on discriminatory intent. It contends that the framework can pressure employers into race-based decision-making and suggests a lower threshold for establishing that an employment practice serves a legitimate business purpose.

But employers remain subject to the law Congress enacted. “Although federal enforcement priorities have changed, the legal risks associated with disparate impact remain substantial,” said Cara Crotty, an attorney in the Columbia, S.C., office of Constangy. “Employers face a more complex environment than ever before, particularly as AI becomes embedded throughout the employment life cycle, and state and local jurisdictions increasingly regulate automated employment decision-making,” she said.

That distinction between enforcement priorities and governing law is critical for HR.

“Employers shouldn’t mistake reduced federal enforcement for reduced discrimination risk,” said Jocelyn King, CEO at VirgilHR, a compliance technology company. “Title VII and state and local laws still apply, so HR leaders need to navigate an increasingly fragmented compliance landscape, particularly around hiring, selection, compensation, and AI.”

Clements added that employers may continue to face private litigation based on disparate impact regardless of changes in the EEOC’s enforcement priorities.

“Private litigation does not disappear simply because an enforcement agency changes course,” she said. “Employers should expect plaintiffs’ attorneys to continue challenging selection procedures that produce statistically significant adverse impact, particularly where validation evidence is weak or outdated.”

Similarly, rescission of the UGESP would not eliminate disparate impact claims, Crotty said. “Employers would simply lose the long-standing framework that has helped them evaluate their employment practices,” she said.

State Laws Add Another Layer

The federal shift comes as states and local jurisdictions continue to develop their own approaches to disparate impact and AI-driven employment decisions. The result is a compliance landscape in which a practice that receives less scrutiny from federal regulators may still create significant exposure elsewhere.

New York and New Jersey recently expanded disparate impact liability under their existing employment discrimination laws, while Illinois and Maryland passed legislation codifying disparate impact liability. California, Colorado, Massachusetts and Minnesota also recognize disparate impact liability through statute or case law.

For multistate employers, simply following the direction of federal enforcement may therefore be insufficient.

Winans said HR leaders should “stay close to the changes, connect with their legal counsel, and take a closer look at their practices. HR will need to walk that fine line of doing what’s right for the business but also keeping employees in mind,” she said. “HR is caught between a rock and a hard place for the moment.”

AI Raises the Stakes

The implications are particularly significant as employers expand their use of AI in recruiting and talent management. Organizations increasingly use automated tools to screen resumes, score interviews, rank applicants, and identify candidates for promotion. Yet algorithms trained on historical employment data can reproduce or amplify existing disparities, while so-called black-box systems can make it difficult for employers to understand how recommendations are generated.

That creates a potentially dangerous disconnect. Federal enforcement may become less focused on disparate impact just as technology makes it easier for employment decisions to be made at scale.

Crotty said employers should not assume that outsourcing a selection tool outsources the legal risk. “If a hiring tool screens out applicants in protected groups at disproportionate rates, plaintiffs will generally focus on the employer who used the tool to make the challenged employment decisions, not the vendor or developer,” she said.

HR leaders should therefore understand how AI-enabled systems function, what data they rely on and whether independent validation has been conducted, Crotty said.

“Vendor contracts should also address legal compliance, documentation, and data retention obligations, cooperation during investigations and audits, and appropriate indemnification provisions,” she said.

Winans cautioned that the new federal enforcement direction “may open the flood gates to employers wanting to use AI tools without guardrails.”

She noted the tension HR leaders face: “Yes, these tools help with productivity but could also potentially create risk. They may create some consequences that an organization is not ready for.”

Toolkit: Using Artificial Intelligence for Employment Purposes

Validation Remains Valuable

Experts recommended not to abandon validation or workforce analytics.

Clements said employers should continue to review selection procedures, including assessments, interviews, screening tools, and AI-enabled systems, and determine whether existing validation studies remain current, job-specific, and adequately documented. Employers should also continue monitoring selection outcomes and documenting the business necessity of their practices, she said.

“The DOJ opinion itself acknowledges that employers still must be able to demonstrate legitimate business reasons for selection procedures and other neutral rules that have a disparate impact,” Clements said. “In practice, the strongest way to demonstrate business necessity remains the same approach employers have used for decades.”

The rationale extends beyond legal compliance. “These questions are important not only for legal defensibility but also for improving talent decisions and workforce outcomes,” Clements said. “Validation helps employers identify selection procedures that are not only fair but, even more importantly, also effective. Properly developed and validated selection tools increase the success of hiring people who have the necessary knowledge, skills, abilities, and competencies needed for success in the role. Validated tools also reduce costs associated with turnover and with hiring the wrong candidate.”