Recommended Citation
Daniel S. Alcorn,
Artificial Intelligence and Automated Decision-Making in Employment: The Future of the Embattled Disparate Impact Theory of Discrimination Under Title VII,
17 Case W. Res. J.L. Tech. & Internet
A7
(2026)
Available at:
https://scholarlycommons.law.case.edu/jolti/vol17/iss1/7
Abstract
Artificial intelligence now makes core employment decisions—from resume screening and video interviews to promotion and termination—yet Title VII’s disparate impact doctrine was built for paper tests, not black-box models. This Article argues that, while under attack by the Trump Administration and a minority of the U.S. Supreme Court, disparate impact remains indispensable but increasingly inadequate without adaptation. We show how algorithmic opacity frustrates causation, vendor delegation diffuses liability, predictive-accuracy claims distort “business necessity,” and trade-secret barriers impede proof of less-discriminatory alternatives. Drawing on emerging U.S. and comparative regimes, we propose a practical toolkit: (1) treat the algorithmic system as the “specific employment practice” where warranted; (2) shift limited disclosure and validation burdens once plaintiffs show significant disparities; (3) require meaningful human-in-the-loop review (no solely automated adverse actions) with logs, explanations, and appeal rights; and (4) integrate periodic bias/impact assessments and re-audits after material model changes. We outline regulatory safe harbors and contract terms that operationalize these reforms while preserving efficiency. Without such adjustments, AI risks entrenching discrimination behind a facade of objectivity. With them, Title VII can remain a viable check on systemic bias in the algorithmic workplace.