AI Hiring Tools Face a Patchwork of Rules as Federal Standards Stay Put
Employers using artificial intelligence to screen, rank, or evaluate job candidates are caught between federal rules that haven't changed and state restrictions that keep multiplying. Recent executive orders have signaled a lighter federal touch on AI regulation, but that doesn't mean compliance requirements are easing.
The compliance burden remains real. Organizations can't wait for regulatory clarity to settle-they need to understand how their AI tools work, identify potential risks, and establish governance now.
Federal law hasn't moved
Title VII of the Civil Rights Act, the Americans with Disabilities Act, and the Age Discrimination in Employment Act apply to AI-assisted hiring decisions the same way they apply to decisions made by a hiring manager. If an algorithm produces a disparate impact on a protected group, the legal analysis is identical.
Over the past year, executive actions have characterized disparate impact theory as constitutionally inconsistent and directed federal agencies to take a narrower approach to AI oversight. The Equal Employment Opportunity Commission has scaled back certain disparate impact investigations. A separate executive order signaled willingness to challenge state AI regulations viewed as overly burdensome.
But disparate impact remains a viable legal theory under federal civil rights law. Claims can proceed through private litigation or state enforcement channels. The Department of Labor has made clear that employers cannot rely on automated systems to satisfy obligations that still require human oversight and accountability.
AI tools used for scheduling, productivity monitoring, or leave administration remain subject to wage-and-hour and leave requirements. Algorithmic errors at scale can quickly become systemic violations, even without intent to violate the law. Employers must ensure AI-driven scheduling and timekeeping systems don't shave compensable time, misclassify hours, or deny leave requests in ways that create liability.
States are writing their own rules
Federal actions don't preempt state law. Until a court invalidates a specific statute, state and local requirements remain fully enforceable. A single AI tool may be subject to multiple, sometimes inconsistent legal standards for employers operating across jurisdictions.
Some requirements are already in force. New York City's Local Law 144 requires bias audits and public disclosures for certain automated employment decision tools. Illinois mandates notice to applicants and employees when AI is used in hiring, along with testing requirements. California has expanded its civil rights framework to address AI-driven employment decisions and requires extended record retention.
Other states are moving cautiously. Colorado, once positioned as a leading model for comprehensive state AI regulation, delayed its AI Act and is considering whether to repeal or significantly revise portions of it.
Employers should assess AI tools against the most demanding applicable state requirements, then implement controls such as bias testing, clear documentation, and defined governance protocols.
Courts aren't waiting for rules to clarify
In January, a class action against an AI hiring platform alleged that scoring and profiling practices implicate the Fair Credit Reporting Act and analogous state laws. This shows compliance considerations may extend beyond traditional discrimination frameworks.
Early cases addressing AI-driven employment decisions suggest that existing liability frameworks will apply. In Mobley v. Workday Inc., a federal court allowed disparate impact claims to proceed against a vendor whose software was alleged to screen out applicants based on race, age, and disability. The court rejected treating automated decision-making systems differently from human decision-making, noting that doing so would undermine established anti-discrimination protections.
One consistent point across cases: Employers remain responsible for employment decisions, even when third-party technology informs them. Using a vendor doesn't shift legal accountability.
A practical path forward
Waiting for regulatory clarity isn't a neutral strategy. A comprehensive federal framework for AI in employment may take years. Yet employers that delay risk missing operational benefits. The question is not whether to deploy these tools but how to do so responsibly.
Start by understanding where AI is already embedded in workforce processes-screening, evaluation, scheduling, compensation. These uses should be visible to legal and human resources stakeholders.
Evaluate how tools function, whether they produce disparate outcomes, and how decisions informed by them are documented and reviewed. Pre-deployment testing, periodic audits, and clear escalation paths for concerning outcomes are practical steps.
Vendor relationships matter. When AI tools integrate into broader platforms, responsibility for how they operate isn't always clearly defined. Contracts should address data use, audit rights, and accountability for compliance, especially for tools influencing hiring or other high-stakes decisions.
Ultimately, employers are responsible for the decisions they make. AI may change how those decisions are informed, but it doesn't change the obligation to ensure they are lawful, explainable, and consistently applied.
For HR professionals implementing or evaluating AI tools, understanding AI for Human Resources and developing organizational strategy is essential. CHRO-level guidance on AI strategy for human resources leadership can help align implementation with compliance and business goals.
Your membership also unlocks:
