Artificial Intelligence in Employment: 2025 Regulatory Update

By Dave Schmidt and Sarah Layman

The integration of artificial intelligence (AI) into employment decision-making processes for organizations continues to accelerate, as does the evolution of the legal and regulatory environment. This post provides information on key happenings in state and federal regulatory activity as of June 2025.

State Activity Adds to the Patchwork of Requirements

Colorado, California, and Virginia have seen significant activity in 2025 related to the use of AI that employers need to know about:

Colorado
Status: Passed, In Effect on February 1, 2026

The Colorado Consumer Protections in Interactions with Artificial Intelligence Systems (CPIAIS) is currently the most comprehensive state law addressing the development and use of AI in high-impact contexts (including employment decision-making). Despite significant debate and last-minute calls by Governor Polis to delay or revise the legislation, including efforts to ease compliance burdens and refine the definition of “algorithmic discrimination,” the law is set to take effect on February 1, 2026.

CPIAIS applies to high-risk AI systems, defined as any AI that makes—or plays a substantial role in making—a “consequential decision,” such as hiring, promotion, or termination. Both developers and deployers of such systems are held to a duty of reasonable care to avoid algorithmic discrimination. If discrimination is discovered, it must be reported to the Colorado Attorney General within 90 days. Reasonable care is presumed if specific compliance steps are taken (e.g., Risk management programs, Annual impact assessments, Transparency and notices). For additional details, see our blog or download DCI’s Colorado Artificial Intelligence Act Cheat Sheet.

California
Status: Pending Approval by Office of Administrative Law

Although early efforts to pass broad, stand-alone AI laws in California stalled (see DCI’s blogs here and here), state regulators have instead moved forward with targeted amendments to the California Fair Employment and Housing Act (FEHA). The new regulations, titled “Employment Regulations Regarding Automated-Decision Systems” were finalized in March 2025 and could take effect as early as July 2025 (pending approval from the Office of Administrative Law). Most notably, the amended text provides:

• Clear definitions of “automated-decision systems (ADS),” “algorithm,” “artificial intelligence,” and “machine learning”;
• A mandate to retain all relevant employment records, including ADS data and documentation, for at least four years;
• Notice requirements obligating employers to inform applicants and employees when an ADS is used in employment decision-making;
• A requirement to provide reasonable accommodations for individuals with disabilities or religious needs, particularly when ADS tools evaluate attributes (e.g., reaction time, vocal tone) that may be impacted;
• Explicit language stating that practices resulting in adverse impact are unlawful unless the criteria used are job-related, consistent with business necessity, and there is no less discriminatory alternative available that would serve the employer’s goals as effectively;
• Recognition that, in the event of a legal claim, evidence (or lack thereof) of proactive anti-bias testing and mitigation efforts—including the quality, effectiveness, scope, and response to such testing—will be relevant.

Virginia
Status: Vetoed by Governor

This spring Virginia’s legislature narrowly passed House Bill 2094, the High-Risk Artificial Intelligence Developer and Deployer Act, —bore a close resemblance to Colorado’s CPIAIS. However, Governor Glenn Youngkin subsequently vetoed the bill, citing its “burdensome” regulatory framework. The bill could technically still become law via a two-thirds override in both chambers, but given the initial vote tallies, this outcome is unlikely.

Federal Updates

There have also been developments related to AI use and oversight at the federal level in early 2025,including new guidance from the Office of Management and Budget (OMB) for federal agencies and the withdrawal of previously provided guidance by some federal agencies.

OMB Guidance for Federal Agencies

At the federal level, the White House Office of Management and Budget issued two significant memoranda in April of 2025 addressing AI use and procurement across federal agencies, officially rescinding and replacing Biden-era guidance.

The primary directive, OMB Memo M-25-21-"Accelerating Federal Use of AI through Innovation, Governance, and Public Trust", establishes comprehensive standards for the governance, risk management, and oversight of AI in government operations. According to the White House Fact Sheet, these policies mark a fundamental shift from the previous administration by emphasizing rapid and responsible AI adoption (e.g., “Agency Chief AI Officer roles are redefined to serve as change agents and AI advocates, rather than overseeing layers of bureaucracy”), efficient and effective acquisition of American AI systems (the focus of OMB Memo M-25-22), and practical applications that “work for the American people.”

While these requirements are centered on federal agencies, their reach is likely to extend to private-sector vendors and contractors providing AI tools and services to the government, shaping wider expectations for responsible AI use and procurement.

Withdrawn Guidance

Under the Trump Administration, a range of previously issued AI-related guidance documents have been formally removed from federal websites including:

• Artificial Intelligence and Worker Well-being: Principles and Best Practices for Developers and Employers (DCI Blog)
• Guide on Artificial Intelligence and Equal Employment Opportunity for Federal Contractors (DCI Blog)
• Joint Statement on Enforcement of Civil Rights, Fair Competition, Consumer Protection, and Equal Opportunity Laws in Automated Systems (DCI Blog)

It is yet to be seen if revised guidance documents related to the use of AI in employment decision-making will be issued.

Looking Ahead: Preparation and Monitoring Needed

As the legal landscape surrounding AI in employment evolves, it is prudent for employers to not only track legislative changes, but also proactively create robust governance teams and frameworks to address the challenges and risks associated with using AI-driven hiring processes. DCI will continue to monitor developments and provide timely updates with practical insight for employers.

For more information, register for DCI’s upcoming webinar covering the key updates in AI regulations for employers.