New Automated Employment Decision Systems ( AEDS) Regulations – Employment Law Weekly

New Automated Employment Decision Systems ( AEDS) Regulations

New regulations, Title 2 CCR 11008 et.seq., promulgated under the California Civil Rights Department’s authority pursuant to AB 331 (the Automated Employment Decision-Making Law), establish requirements to promote fairness, transparency, and accountability in the use of automated tools for employment decisions.

They aim to prevent discrimination by mandating evaluations for adverse impacts on protected groups. These new regulations took effect on October 1, 2025, applying to private employers with 25 or more employees (or independent contractors) who deploy such systems in California. Initial assessments for pre-existing AEDS must be completed by January 1, 2026. CRD will issue FAQs and templates to aid compliance.

The purpose of these new regulations is to ensure automated employment decision systems (AEDS) do not perpetuate or exacerbate discrimination based on protected characteristics (e.g., race, gender, age, disability) in hiring, promotion, termination, or other employment decisions. Employers must conduct regular impact assessments to identify and mitigate biases.

An Automated Employment Decision System (AEDS) is defined as any computational process, including AI, machine learning, or statistical modeling, that analyzes data to make or assist in employment decisions. Excludes tools used solely for payroll, benefits administration, or non-decision-making functions.

Exemptions include Government entities. Use of AEDS for compliance with federal/state laws (e.g., EEO reporting). And temporary or one-time use without ongoing deployment.

Deployers of an AEDS must conduct an annual Impact Assessment for each AEDS, starting with the first deployment after October 1, 2025. Assessments are due by January 1 of the following year for initial use. Required Elements:

– – Description of the AEDS (purpose, data inputs, outputs, decision types).
– – Data analysis for adverse impacts: Compare outcomes across protected groups (e.g., selection rates, pass/fail rates) using statistical tests (e.g., 80% rule or four-fifths rule).
– – Identification of mitigations for any disparities found (e.g., adjusting algorithms, diversifying training data).
– – Notification to affected parties if adverse impact is detected.

The new regulations require Deployers to provide clear and conspicuous notices at least 45 days before deploying an AEDS. These must be provided to Individuals via job postings, application portals, or direct communication, including:

– – That an AEDS will be used.
– – Types of data collected and decisions affected.
– – Right to request accommodations or opt out (if feasible).
– – Contact info for questions.

Penalties include up to $100 per individual (for notice violations) or $200 per employee (for assessment failures). Injunctions my compel corrective actions, and damages for discriminatory outcomes. Willful violations can lead to enhanced penalties under FEHA. However there are Right to Cure provisions. First-time minor violations allow 30 days to correct before penalties are imposed.

New Automated Employment Decision Systems ( AEDS) Regulations

There are 0 comments

Share:

More Posts

Bills Signed by Governor (10/7/25)

AB 715 by Assemblymember Rick Chavez Zbur (D-Hollywood) — Educational equity: discrimination: antisemitism prevention (signing message) SB 48 by Senator Lena Gonzalez (D-Los Angeles) –

Send Us A Message

Skip to content