Automated-decision systems in employment (Civil Rights Council FEHA regulations)
FEHA ADS Regulations
Artificial intelligence · Employees
Regulations approved June 27, 2025 and effective October 1, 2025 that confirm using an automated-decision system (including AI) in hiring, promotion, or other employment decisions can violate California's anti-discrimination law if it harms people based on protected traits. They require keeping automated-decision data with other employment records for four years and treat some AI assessments as unlawful medical inquiries.
- Where
- California
- Citation
- Cal. Code Regs. tit. 2, 11008 et seq. (amendments to FEHA employment regulations)
- Status
- In force
- In force since
- 2025-10-01
- Enforced by
- California Civil Rights Department
- People can sue
- Yes
- Penalties
- FEHA remedies (after an administrative complaint): damages, back pay, injunctive relief, attorney's fees.
- Applies to
- Employers with five or more employees, employment agencies, and their agents covered by the Fair Employment and Housing Act
Practices it requires
- Do not use an automated-decision system or selection criteria that discriminates against applicants or employees on a protected basis.Cal. Code Regs. tit. 2, 11008 et seq.
- Automated assessments (tests, questions, puzzle games) that elicit disability information may be unlawful medical inquiries.Cal. Code Regs. tit. 2 (medical inquiry provisions)
Other duties
- Keep employment records, including automated-decision data, for at least four years.Cal. Code Regs. tit. 2 (recordkeeping provisions)
Sources
Checked against these sources on 2026-09-25 by research agent (Claude), primary sources.
Unverified: The final-text PDF is a scanned image and could not be text-extracted, so section-level citations (11008, 11013, etc.) are not verified; obligations are summarized from the CRD press release.
Research reference, not legal advice.