EEOC Scraps Affirmative Action Guidelines, Retreats From Disparate Impact Theory
Not legal advice. This article discusses legal and regulatory developments and is generated with AI assistance under human editorial review. It is provided for informational purposes only and does not constitute legal advice. Laws and regulations vary by jurisdiction and are subject to change. You should consult your own qualified legal counsel before taking or refraining from any action based on the content of this article. Read our AI Transparency statement.
The News
Three moves in rapid succession are reshaping US employment discrimination enforcement. On 30 June, the EEOC voted to rescind its affirmative action interpretive guidelines and related compliance manual sections — guidance that had stood for roughly 40 years. On 9 June, the Department of Justice's Office of Legal Counsel issued a memorandum concluding that disparate impact liability under Title VII is unconstitutional, arguing it violates equal protection principles. And on 4 June, the EEOC published its National Enforcement Plan for fiscal years 2025-2029, prioritising 'cases involving repeated or overt discrimination' — a deliberate pivot away from the systemic disparate impact cases that have defined AI hiring enforcement. Employment Law Worldview notes that the EEOC has also reduced staff by approximately 10% and, in February 2025, dismissed six of its own cases involving sexual orientation and gender identity discrimination. The Commission separately voted on 1 July to release a draft strategic plan for FY 2026-2030 for public input.
Why It Matters
This matters for TA leaders because disparate impact — not intentional discrimination — is the legal theory under which most AI hiring tool challenges are brought. When an algorithmic screening tool produces a selection rate for a protected group that is substantially lower than for other groups, that's a disparate impact claim. The EEOC's 2024-2028 Strategic Enforcement Plan had explicitly prioritised 'algorithmic fairness.' Now the agency's posture has reversed: the focus is on overt, intentional discrimination and merit-based hiring, and the DOJ has declared the entire disparate impact framework unconstitutional. However — and this is the operational reality — state and local enforcement isn't following the federal retreat. New York City's Local Law 144, California's ADS regulations, Illinois's Human Rights Act amendments, and the newly signed Colorado SB 26-189 all maintain or strengthen disparate impact-style protections for AI hiring. Multi-state employers now face an environment where federal enforcement is pulling back precisely as state enforcement is ramping up. The compliance burden hasn't disappeared — it's just moved to a patchwork of state-level requirements.
Taking Action
All action is situated in your corporate context. These are starting points, not prescriptions — choose what fits.
Connect
Who to talk with in your enterprise
Your employment counsel: Ask directly how the DOJ's disparate impact memorandum affects your organisation's exposure — and whether your current AI hiring tool bias audits would hold up under state-level scrutiny even if federal enforcement retreats. Your head of TA operations: If you operate in multiple states, now is the time to map which jurisdictions have active AI hiring laws (NYC, Illinois, California, Colorado) and confirm you're meeting each one's specific requirements. The EEOC pulling back doesn't mean state AGs are.
Inspect
What to take a closer look at
Review your most recent AI hiring tool bias audit. Does it test for disparate impact across race, sex, age, and intersectional categories? Does it use the four-fifths rule or a comparable standard? If the audit was conducted by the tool vendor rather than an independent auditor, it likely doesn't satisfy NYC Local Law 144. Check whether you have current bias audits for every jurisdiction where you use automated screening — the federal retreat doesn't exempt you from state requirements. Also review your EEO-1 reporting practices: the EEOC is phasing out EEO-1 reporting, but several states have their own pay and demographic data reporting mandates that remain in force.
Reflect
What to think about more deeply
If the federal government stops asking whether your AI hiring tools discriminate against protected groups, do you still ask the question yourself? Because your candidates' lawyers almost certainly will — and they'll bring the case under state law instead.