14 State AGs Reaffirm Disparate Impact Enforcement — With Algorithmic Hiring, Lending and Tenant-Screening Tools Squarely in Scope

14 State AGs Reaffirm Disparate Impact Enforcement — With Algorithmic Hiring, Lending and Tenant-Screening Tools Squarely in Scope

On September 17, 2026, the attorneys general of fourteen states — California, Delaware, Hawaii, Illinois, Maryland, Massachusetts, Michigan, Minnesota, Nevada, New Mexico, New York, Oregon, Vermont and Washington — issued joint guidance "reaffirming that disparate impact liability remains lawful, constitutional, and enforceable under both state and federal law¹," responding directly to the federal government's effort to narrow or eliminate disparate-impact enforcement (Ogletree Deakins analysis, Sept. 24).

The federal/state divergence it pushes back against: Executive Order 14281 directs agencies to stop enforcing disparate-impact theories, and on June 9, 2026 DOJ's Office of Legal Counsel issued a memorandum opinion "concluding that the EEOC's disparate impact guidelines are inconsistent with Title VII and constitutionally suspect."

Why this matters for AI enforcement specifically — the guidance and the Ogletree read of it:

• "The guidance cites recent actions against algorithmic bias in lending and tenant screening, housing ordinances that disproportionately affect domestic violence survivors, and language access barriers."
• It "points to recent court decisions allowing disparate impact claims against algorithmic screening and hiring tools to proceed, and to state AG settlements targeting AI-related practices."
• It includes "a sector-by-sector roadmap covering employment, housing, lending, language access, disaster preparedness, siting, and education" — signaling what these AGs "will look for in enforcement actions."
• Ogletree's advice to employers: "Auditing AI and algorithmic tools used in employment decisions for potential disparate impact, particularly in the signatory states where enforcement and litigation targeting these tools are already underway."

What it means: Even as federal disparate-impact enforcement recedes (EO 14281, the OLC opinion), fourteen states representing most of the largest AI deployment markets have formally committed to enforcing the same theory against the same employer data² — including AI/algorithmic hiring, screening and lending tools. The same employment decision can now face a federal intentional-discrimination/DEI inquiry and a state disparate-impact inquiry simultaneously. For vendors and deployers of algorithmic hiring tools, the guidance is a preview of state enforcement theories where no AI-specific statute exists — the exact "existing law applied to AI" pattern this topic tracks.