New California Regulations Address Employment Discrimination Through AI and Other Automated-Decision Systems

By Swerdlow Florence Sanchez Swerdlow & Wimmer

August 8, 2025

The California Civil Rights Counsel (“Counsel”) recently received the necessary final approval for its proposed regulations governing the use of Artificial Intelligence (“AI”) and other “automated-decision systems” in employment decisions. These regulations go into effect on October 1, 2025.

Purpose:

The regulations clarify how existing anti-discrimination laws in the workplace work with new and emerging technologies like algorithms and AI. Overall, these changes apply to the application, selection, interview, hiring, and employment processes for all applicants and employees.

Changes:

More than 20 sections of the California Code of Regulations have been modified by the Counsel regarding the use of automated-decision systems. Some of the important highlights include:

•  New California Code of Regulations, Title 2, Section 11008.1 provides certain definitions such as “automated-decision system” and “algorithm” that apply to related statutes.
•  California Code of Regulations, Title 2, Section 11009, which already provides relief for unlawful employment discrimination, has now been modified with additional language to expressly state that it is unlawful “to use an automated decision-system or decision criteria” that discriminates against applicants and/or employees (whether individually or as a class) based on a category protected under FEHA.
•  California Code of Regulations, Title 2, Section 11013 has been modified to require employers to keep and maintain, not just personnel or other employment records, but also all applications, employment-referral records, selection criteria, automated-decision system data and other similar records created or received by the employer.
•  California Code of Regulations, Title 2, Section 11016, which generally prevents discrimination during recruitment, the application process, and interviews, has been modified to regulate the related use of automated-decision systems. In particular, when screening applicants, to avoid unlawful discrimination when using systems that e.g., analyze an applicant’s tone of voice, facial expression, physical characteristics or other behavior that may lead to discrimination based on race, gender, disability or other FEHA protected categories, an employer may need to provide a reasonable accommodation.
•  Regulations governing the consideration of criminal history, and protections based on national origin, ancestry, sex, pregnancy, childbirth (or other related medical conditions), and other FEHA-protected categories, now include specific requirements for the use of “automated-decision systems.”

These new regulations are separate and distinct from the proposed legislation found in Assembly Bill 1018, which involves a comprehensive attempt to regulate AI in California and would impose significant compliance burdens on employers who use AI-driven hiring and workforce management tools.

Takeaways:

California is just beginning to address AI through regulations and legislation. Employers can expect their use of AI to become more regulated over the coming years. Employers who currently use algorithms, AI, machine learning or automated-decision systems when making employment decisions, whether through their own systems or through an agent or other third-party, should be careful to ensure that such use is not discriminatory. Because the definitions included in these regulations are highly technical, a detailed analysis of the use of these systems should be undertaken with your legal counsel.

www.swerdlowlaw.com