Does AB 2930 apply to AI vendors as well as employers?

Yes. AB 2930 imposes independent obligations on both employers and AI vendors. Vendors must publish bias audit summaries publicly, maintain technical documentation of training data and performance metrics, and ensure accuracy and fairness claims are substantiated by audit results. Vendor obligations exist independently — they cannot be contractually shifted to employers, and both entities face direct enforcement exposure. This means AI hiring tool vendors have direct legal obligations in California regardless of what their employer contracts say.

What must the annual bias audit include under AB 2930?

The annual bias audit must be conducted by an independent auditor (not employed by employer or vendor, no financial conflict of interest) and must calculate selection rates and impact ratios by sex, race, ethnicity, and intersectional combinations where sample size permits (minimum n=30 per group). Required elements: impact ratio calculation using the 4/5ths rule, demographic composition of training data, adverse impact analysis for each AEDT output, and full methodology documentation retained 3 years minimum. If adverse impact is identified, the audit must document remediation steps and track remediation status in each subsequent annual audit. Results must be published on a publicly accessible URL.

What advance notice is required before using an AEDT?

Employers must provide written notice at least 10 business days before using an AEDT in an employment decision. Notice must disclose: that an AEDT will be used, the type of AEDT and what it evaluates, the data collected and how it is used, and information about the opt-out process. Notice must be in the language of any job posting or employment communication. Candidates not yet in the formal application process must receive notice at the time they are first evaluated — not after the AEDT has already run.

What are the opt-out rights and what alternative process must employers provide?

Candidates may opt out of AEDT evaluation. Employers must provide a reasonable alternative selection process — the alternative cannot be materially inferior in decision timeline. Employers cannot deny employment solely because a candidate exercised opt-out rights. Documentation of opt-out requests, the alternative process offered, and the outcome must be retained. The alternative process does not need to be identical to the AEDT path, but it cannot effectively eliminate the candidate's chance of consideration by creating an unreasonably burdensome or slower process.

What is required when a candidate requests an explanation?

Upon written request, the employer must provide: the data collected by the AEDT about this candidate, and the score, ranking, or recommendation the AEDT produced with the principal factors contributing to it. Response required within 30 days. The explanation must be specific to this candidate — a generic model description does not satisfy the obligation. This requires per-decision records with inputs, outputs, and feature attribution captured at evaluation time, since re-running a current model on historical inputs may produce different results if the model has been updated after the original evaluation.

How does AB 2930 compare to New York City Local Law 144?

AB 2930 is broader than NYC LL 144. AB 2930 covers all employment decisions (LL 144 covers hiring and promotion only), applies statewide in California (LL 144 is NYC only), requires intersectional analysis (LL 144 uses binary sex and race/ethnicity), adds explanation rights absent from LL 144, requires opt-out with an alternative process (not required in LL 144), and imposes direct vendor obligations. Penalties are also higher — up to $10,000/day per violation vs $1,500 per violation for NYC. Companies operating in both jurisdictions must satisfy both frameworks simultaneously.

What documentation should employers and vendors maintain?

Employers: per-decision records with candidate inputs, AEDT outputs, and principal factors retained 2 years minimum; notice delivery records for each candidate; opt-out request records and alternative process documentation; published bias audit results and methodology retained 3 years. Vendors: training data documentation including demographics and preprocessing; independent bias audit results and methodology; performance metrics by demographic subgroup; documentation supporting any accuracy or bias claims in marketing materials. Both parties should retain records 3 years to cover the enforcement window — records assembled after an enforcement action begins have limited defensive value.

California AB 2930: What Employers and AI Vendors Must Do for Automated Employment Decisions

Q: What is an automated employment decision tool under California AB 2930?

An AEDT under AB 2930 is any computational process — including AI, machine learning, or algorithmic methods — that substantially assists or replaces discretionary decision-making for employment decisions. Covered decisions: hiring, promotion, demotion, transfer, termination, pay setting, performance evaluation, and work scheduling. A tool 'substantially assists' when it generates a score, ranking, or recommendation that materially influences who the human decision-maker considers — even if a human makes the final call. Tools used purely for scheduling administrative tasks (not performance evaluation), cybersecurity monitoring, and data processing without scoring or ranking are excluded.

California AB 2930 imposes obligations on employers and AI vendors using automated employment decision tools (AEDTs) in California. Covered employers must commission annual independent bias audits, provide 10-business-day advance notice to candidates, offer opt-out rights with an alternative selection process, and respond to per-candidate explanation requests within 30 days. AI vendors must publish bias audit summaries publicly and maintain technical documentation. Violations carry civil penalties up to $10,000 per day per violation.

Who AB 2930 Covers

AB 2930 applies to employers with California employees who use an automated employment decision tool (AEDT), and independently to vendors who sell or provide such tools. An AEDT is any computational process — AI, machine learning, or algorithmic method — that substantially assists or replaces discretionary decision-making for employment decisions including hiring, promotion, demotion, transfer, termination, pay setting, performance evaluation, and work scheduling. A tool substantially assists when it generates a score, ranking, or recommendation that materially influences who the human decision-maker considers — even if a human makes the final selection. Both employer and vendor have independent legal obligations that cannot be contractually shifted between them.

Employer Obligations: Bias Audit, Notice, Opt-Out, Explanation

Employers have four categories of obligation. Bias audit: commission an independent bias audit annually, covering selection rate disparities by sex, race, ethnicity, and intersectional categories — publish results on a publicly accessible URL. Advance notice: provide written notice at least 10 business days before AEDT evaluation, disclosing the type of tool, what it evaluates, data collected, and opt-out information, in the language of the job posting. Opt-out: provide a reasonable alternative selection process for candidates who opt out — the alternative cannot be materially inferior, and employment cannot be denied for exercising opt-out rights. Explanation: upon written request, respond within 30 days with the candidate's data collected by the AEDT and the principal factors contributing to the score or recommendation.

Vendor Obligations: Audit Publication and Technical Documentation

AI vendors who sell or license AEDTs to California employers have independent obligations. Audit publication: publish a summary of any bias audits conducted on the AEDT — including auditor identity, methodology, and demographic results — at no cost and within 60 days of completing each annual audit. Technical documentation: maintain records of training data sources and demographic composition, preprocessing steps, known limitations, and performance metrics across demographic subgroups. Claims substantiation: any accuracy, bias, or fairness claims in marketing must be substantiated by audit results consistent with the published audit summary — unsupported claims create concurrent exposure under FTC Section 5. Vendor obligations exist independently of what the employer contract says.

Bias Audit Requirements and the 4/5ths Rule

The annual bias audit must be conducted by an independent auditor — not employed by the employer or vendor, no financial conflict of interest. Required audit elements: selection rate analysis by sex, race, ethnicity, and intersectional combinations (minimum n=30 per group for statistical validity); impact ratio calculation for each group against the most selected group (groups below 80% threshold — the 4/5ths rule — require documentation and remediation consideration); training data demographic composition; adverse impact analysis for each AEDT output; and full methodology documentation retained for 3 years minimum. If adverse impact is identified, the audit must document remediation steps — alternative feature selection, threshold adjustment, or model retraining — and each subsequent annual audit must update the remediation status.

Per-Decision Documentation for 30-Day Explanation Responses

AB 2930's explanation rights require per-decision records captured at evaluation time — not re-runs of a current model on historical inputs, which may produce different outputs if the model has been updated. Required per-decision records: all candidate data collected and processed by the AEDT for this evaluation, AEDT output (score, rank, pass/fail), principal factors contributing to the output (feature attribution or reason codes specific to this candidate), AEDT version active at evaluation time, evaluation timestamp, and any human review or override outcome. Records must be retained 2 years minimum and retrievable by candidate identifier within 30 days of a written request. Employers using AEDT output as one of several inputs to a human decision must still capture the AEDT output and factors separately — the explanation obligation attaches to the AEDT component even if the final decision is human.