Artificial intelligence is reshaping how companies hire, evaluate, and manage employees, but the legal framework governing these tools is fragmenting across state lines while the federal government actively blocks stricter rules. According to the Society for Human Resource Management's 2025 Talent Trends Report, 43% of all employers now use AI in HR tasks, nearly doubling within the past year. Yet only 18% of health care organizations report having a mature AI governance structure, creating significant exposure for employers navigating a patchwork of conflicting state and federal regulations.

Why Is AI in Employment Decisions So Risky?

AI tools are increasingly used to screen job applicants, recommend candidates for promotion, and evaluate employee performance. The problem is that these systems can inherit biases from their training data or reflect the prejudices of their designers. A pending class action lawsuit in California federal court, Mobley v. Workday, illustrates the stakes: a job applicant alleges that Workday's AI-based screening tools systematically rejected him and other applicants based on age discrimination. According to court filings, Workday's software rejected 1.1 billion applicants during the relevant time period, meaning the potential class could include hundreds of millions of members. The court granted conditional certification of Age Discrimination in Employment Act claims in May 2025.

Beyond age discrimination, AI hiring tools can disadvantage people with disabilities, different racial or ethnic backgrounds, or other protected characteristics. Existing employment laws like the Americans with Disabilities Act technically apply to AI systems, but federal guidance on how to enforce these protections remains sparse. This legal ambiguity is exactly what's driving employers to scramble for clarity.

Which States Are Regulating AI Employment Tools, and How?

In the absence of comprehensive federal law, states and cities have begun enacting their own requirements. These regulations typically mandate notice, transparency, consent, bias mitigation, auditing, and recordkeeping. Here's what's on the books:

• California: Amended its Fair Employment and Housing Act to expressly prohibit employers from using AI and automated decision systems in ways that cause discrimination, requiring proactive anti-bias testing and recordkeeping.
• Illinois: Amended the Illinois Human Rights Act to prevent discrimination and mandate notice and transparency when employers use AI in employment decisions.
• New York City: Passed Local Law 144, imposing bias evaluation and transparency obligations on employers and employment agencies using automated employment decision tools to screen candidates for hire or promotion.
• Texas: Takes a narrower, employer-friendly approach with the Texas Responsible Artificial Intelligence Act, which prohibits intentional discrimination using AI against protected classes under Texas or federal law.
• Colorado: Passed the Artificial Intelligence Act, effective January 1, 2027, requiring employers to provide notice before using automated decision tools, implement robust adverse action processes with meaningful human review, and retain records about AI use for a minimum of three years.

The Federal Government Is Fighting Back Against State Rules

Here's where the conflict intensifies. In December 2025, a federal Executive Order announced a preference for a "minimally burdensome" national AI framework, signaling potential preemption of state and local laws that impose significant burdens on employers using AI in employment functions. The Executive Order directs a federal task force to challenge state laws deemed inconsistent with this approach. In April 2026, the Department of Justice intervened in litigation challenging Colorado's AI law, further signaling the federal government's willingness to actively oppose state-level AI regulations it views as overly burdensome.

"The ongoing conflict between state and federal positions on this area of the law underscores the uncertainty employers face in navigating the uncharted waters of AI use in the employment space," noted legal experts at Hall Render.

Hall Render, Employment Law Firm

This federal-state clash leaves employers in legal limbo. A company operating across multiple states may need to comply with California's strict transparency rules, Colorado's human review requirements, and Texas's narrower anti-discrimination standard all at once. Meanwhile, the federal government is actively trying to weaken state protections, creating the possibility that rules could be struck down entirely.

How to Build AI Governance Before the Law Catches Up

Given the legal uncertainty, employment law experts recommend that organizations take proactive steps to mitigate risk and prepare for future regulation:

• Develop an AI governance framework: Create clear AI use policies that address both employee and HR use of generative AI tools, establishing who can use AI, for what purposes, and under what oversight.
• Implement approved AI tools with human oversight: Deploy AI systems that meet operational needs while reducing reliance on unmonitored platforms, and ensure human review is part of any adverse employment decision process to catch unintentional bias or discriminatory outcomes.
• Train employees on permissible AI use: Educate staff on what AI tools they can and cannot use, particularly those handling sensitive information like patient data, to prevent unauthorized or inappropriate applications.
• Monitor AI systems for bias: Continuously audit AI tools to identify impermissible bias and track how they're performing across different demographic groups.
• Stay informed on regulatory changes: Continuously monitor federal, state, and local developments related to AI regulation in employment decision-making, since the legal landscape is evolving rapidly.

The gap between AI adoption and legal clarity is widening. Employers are using these tools at scale, but the rules governing them remain fragmented and contested. Organizations that wait for federal clarity may find themselves exposed to liability, discrimination lawsuits, and regulatory penalties. Those that act now to build robust governance structures are positioning themselves to adapt as the law inevitably catches up.