What Employers Should Know About Automated-Decision System (ADS) Regulations Starting Oct 1

September 04, 2025 | From HRCalifornia Extra

by James W. Ward, J.D.; Employment Law Subject Matter Expert/Legal Writer and Editor, CalChamber

The California Civil Rights Council (CRC) recently adopted new regulations addressing automated-decision systems (ADS) and artificial intelligence (AI) in employment practices. Now that the Office of Administrative Law has approved these regulations, they are set to take effect on October 1, 2025.

California’s Fair Employment and Housing Act (FEHA) protects job applicants and employees from discrimination in employment based on certain protected characteristics such as race, gender, age, religious beliefs and disabilities, among many others. And the CRC has the legal authority to issue regulations further defining and clarifying what may be unlawful discrimination under the FEHA — including regulations related to ADS and AI technologies.

While the CRC’s new ADS and AI regulations do not expand the FEHA’s protected classes, they do address employers’ use of technologies with automated computational processes in employment practices, such as screening applicants while recruiting or analyzing employees’ performance, and attempting to clarify when using those tools may be discriminatory.

While using ADS tools may make certain employment practices faster and easier, employers should exercise caution and ensure their use complies with California law. Here's what employers should know about the new regulations effective October 1.

Defining ADS

The CRC’s new regulations define ADS as well as several other terms like “algorithm,” “artificial intelligence” and “machine learning.” ADS is broadly defined as a “computational process that makes a decision or facilitates human decision-making regarding an employment benefit.” Amongst other data processing techniques, these systems may be derived from or use the following:

  • AI: A machine-based system that infers, from the input it receives, how to generate outputs, including predictions, content, recommendations or decisions.
  • Machine learning: The ability for a computer to use and learn from its own analysis of data or experience and apply this learning automatically in future calculations or tasks.
  • Algorithms: A set of rules or instructions a computer follows to perform calculations or other problem-solving operations.

Notably, the regulations provide a non-exhaustive list of tasks that an ADS may perform in the employment context, such as:

  • Using computer-based assessments or tests such as questions, puzzles, games or other challenges to:
    • Make predictive assessments about an applicant or employee;
    • Measure an applicant’s or employee’s skills, dexterity, reaction time and/or other abilities and characteristics;
    • Measure an applicant’s or employee’s personality traits, aptitude, attitude and/or cultural fit; and/or
    • Screen, evaluate, categorize and/or recommend applicants or employees.
  • Directing job advertisements or other recruiting materials to targeted groups.
  • Screening resumes for specific terms and patterns.
  • Analyzing facial expression, word choice and/or voice in online interviews.
  • Analyzing employee or applicant data from third parties.

Employers should carefully review any technologies they use to accomplish the above-listed tasks to determine if they fall within the regulations’ definition of an ADS.

While the regulations’ ADS definition focuses on computational processes, the regulations do exclude certain types of processes from the ADS definition — such as word processing and spreadsheet software, information technology security software, calculators, databases and other computational processes — as long as those processes don’t make decisions regarding an employment benefit.

Potential Liability When Using ADS

The CRC’s revised regulations primarily confirm that existing anti-discrimination laws apply to using an ADS in employment decisions. The regulations expressly state “[i]t is unlawful for an employer or other covered entity to use an automated-decision system or selection criteria (including a qualification standard, employment test or proxy (characteristic closely correlated with a protected category)) that discriminates against an applicant or employee or a class of applicants or employees on a basis protected by the [FEHA].” In other words, it is unlawful to use an ADS if it results in a person or group of people being negatively impacted in hiring and employment decisions based upon a protected characteristic.

In addition to a general rule prohibiting discriminatory ADS use, the CRC combed through the FEHA regulations and added ADS language to numerous provisions to clarify that any specifically prohibited practice includes when such practice is accomplished using an ADS. This may help employers further identify areas in which they should exercise caution when using an ADS.

For example, regarding employment applications, the revised regulations specify that using an ADS that measures an applicant’s skill, dexterity, reaction time or other abilities/characteristics may discriminate against individuals based on protected characteristics. To avoid that, employers may need to provide reasonable accommodations based on an applicant’s disability or religious beliefs. Similarly, an ADS that analyzes an applicant’s tone of voice, facial expressions, or other characteristics or behavior during an interview may discriminate against individuals based on race, national origin, gender, disability or other protected characteristics, for which employers may need to provide reasonable accommodations.

Making it clear that it will be scrutinizing the use of an ADS in all areas of employment law, the CRC included similar ADS language in provisions related to:

  • Pre-employment inquiries and recruiting practices;
  • Consideration of criminal history;
  • National origin and ancestry discrimination;
  • Sex discrimination;
  • Pregnancy, childbirth and related medical conditions;
  • Religious creed;
  • Disability; and
  • Age discrimination.

As with any FEHA-related claim, such as race or sex discrimination, an employer may defend against it by providing evidence that the decision or employment practice was based on a legitimate, nondiscriminatory reason. Specific to an ADS, the regulations provide that relevant evidence to any FEHA-related claim or defense will include evidence — or lack thereof — of “anti-biasing testing or similar proactive efforts to avoid unlawful discrimination, including the quality, efficacy, recency and scope of the effort, the results of such testing or other effort and the response to the results.” The regulations don’t require employers to engage in anti-bias testing, but it’s important to remember that such testing is relevant to discrimination claims.

Agent Liability

Liability may not be just limited to the potential or actual employer but include any agent of the employer. The regulations define “agent” to include “any person acting on behalf of an employer, directly or indirectly, to exercise a function traditionally exercised by the employer, which may include applicant recruitment, applicant screening, hiring, promotion or decisions regarding pay, benefits, or leave, including when such activities and decisions are conducted in whole or in part through the use of an automated decision system.”

This definition is rooted in a California Supreme Court case, Raines v. U.S. Healthworks Medical Group, which, as previously reported, extended liability under the FEHA to third-parties who perform employment-related activities on behalf of an employer. This potentially means developers of an ADS used for employment-related activities may also be directly liable if that ADS violates the FEHA.

Updated Recordkeeping Requirements

The new regulations also update employers’ recordkeeping requirements under the FEHA. Employers must now keep for four years any personnel or other employment record created or received dealing with any employment practice and affecting any employment benefit of any applicant or employee. Records include:

  • Applications;
  • Personnel records;
  • Membership records;
  • Employment referral records;
  • Selection criteria;
  • ADS data; and
  • Any other records related to the employment practice or affected employment benefit.

Notably, the regulations define ADS data as:

  • Any data used in or resulting from the application of an ADS, such as data provided by or about individual applicants or employees, or data reflecting employment decisions or outcomes; and/or
  • Any data used to develop or customize an ADS for use by a particular employer or other covered entity.

Other Related ADS Developments

In addition to the CRC, other government entities are interested in regulating the use of AI and other computational processes. Recently, the California Privacy Protection Agency (CPPA) that enforces the California Privacy Rights Act (CPRA) approved its own regulations governing the use of “Automated Decisionmaking Technology.” The regulations are currently under review with the Office of Administrative Law, which must approve them before they go into effect.

The California Legislature also has active ADS-related bills, including AB 1018 and SB 7, which CalChamber has identified as Cost Drivers under CalChamber’s 2025 Affordability Agenda. These bills may broadly impact an employer’s use of an ADS for employment purposes if enacted as currently written.

Employer Takeaways

Even with increased scrutiny in this area, using an ADS for employment activities is lawful and can still have a positive impact for employers depending on their circumstances. Employers must, however, use caution and make deliberate efforts to ensure their ADS usage is for legitimate, nondiscriminatory purposes. Employers using an ADS should consider the following tips:

  • Identify the ADS tools already in place. If you haven’t already, identify the tools that may qualify under the regulations as an ADS — keeping in mind ADS’ broad definition. One way to approach this may be to consider the ADS employment tasks that the CRC identified in its regulations, such as processes that screen applicants, measure performance, direct job advertisements to targeted groups, and others noted above. Identify the tools you use in these processes to determine whether they qualify as an ADS.
  • Consider how an ADS should objectively support the organization. ADS tools are designed to streamline and improve efficiency in time-consuming tasks, and using these tools should be tailored towards these goals. Evaluate the data used and results provided to ensure these tools meet the goals.
  • Audit the performance of ADS tools. Although the CRC regulations don’t require anti-bias testing, the regulations are clear that such testing — or lack thereof — is relevant evidence in discrimination claims. When using ADS tools, you should have someone evaluate and audit the tools to ensure no negative bias is in the results.
  • Make sure to keep proper records. Under the revised regulations, employers must keep ADS records for four years. While important for general compliance purposes, these records will also be important in the event the employer faces an ADS-based discrimination claim in the future. Plaintiffs will likely argue that an employer’s lack of ADS records is evidence of discrimination.
  • Consult with legal counsel on best practices for maintaining ADS tools, including relationships with vendors that provide these tools. Although the regulations expand liability to third-party agents, employers are always responsible for discriminatory conduct towards applicants and employees.