Gov. Gavin Newsom Signs New AI Laws Affecting California Employers

On Sept. 30, California Gov. Gavin Newsom signed into law a package of bills regulating the use of artificial intelligence (AI) in the workplace, including SB 947, SB 951 and AB 1883. These laws will affect how employers make automated employment decisions, use AI for workplace surveillance and report mass layoffs caused by AI.

All three bills rely on the same broad definition of AI: “an engineered or machine-based system that varies in its level of autonomy and that can, for explicit or implicit objectives, infer from the input it receives how to generate outputs that can influence physical or virtual environments.” It is the same definition incorporated into SB 53, California’s Transparency in Frontier Artificial Intelligence Act, which created a regulatory framework to standardize safety disclosure for developers of “frontier models.”

SB 947

SB 947, referred to as the No Robo Bosses Act of 2026, limits employers’ use of automated decision systems (ADSs) to make disciplinary or termination decisions. It is a narrower version of an earlier bill, SB 7, that would have expanded employer obligations more dramatically had it not been vetoed by the governor last year. In its new form, the law is the first in the nation to require human validation before employers can take certain adverse actions recommended by an ADS.

Scope

SB 947 defines an ADS as “any computational process derived from machine learning, statistical modeling, data analytics, or artificial intelligence that issues simplified output, including a score, classification, or recommendation, that is used to assist or replace human discretionary decisionmaking and materially impacts natural persons.” Cal. Labor Code § 1520(c). The definition of a covered ADS is therefore broader than just AI and extends to many other computational processes.[i] SB 947 applies to all employers in California with very limited exceptions.

Employer Obligations

Beginning July 1, 2027, employers are entirely prohibited from “solely” relying on an ADS to make termination or disciplinary decisions.[ii] Id. § 1522(b)(1). Employers who “primarily” rely on ADS output for disciplinary or termination decisions must have a human “corroborate” the decision using data collected or used by the ADS output or other relevant “corroborating or supporting information,” which may include performance evaluations, personnel files, work product, peer reviews, witness interviews or online customer reviews. Id. § 1522(b)(2). An employer may not rely on the ADS output to discipline or terminate an employee if a human cannot corroborate the ADS output, or determines that the output is “inaccurate, incomplete, or misleading.” Id. § 1522(c). The law does not define what qualifies as a “disciplinary decision,” what it means to “primarily” rely on ADS output, or explain when an output might be deemed “incomplete” or “misleading.”

Affected employees may also request — and in such cases, the employer must provide in an anonymized manner — a description of the data that was used to make the disciplinary or termination decision. Id. §§ 1522(d), 1523(a)–(b). Additionally, if an employer “primarily” relies on ADS to make a disciplinary or termination decision, the employer must provide the affected employee a standalone written post-use notice in plain language that states (1) that the employer primarily relied on an ADS output, (2) a human reviewer corroborated the output and (3) that retaliation is prohibited. Id. § 1522(a). The post-use notice must also provide contact information for a human who can explain the decision and receive the employee’s request for a description of their data. Id. Employers must provide this notice when informing the employee of the decision that primarily relied on an ADS output, regardless of whether the employee requests this information.

Enforcement

Violations can incur a civil penalty of $500 per violation and may be enforced by the labor commissioner and public prosecutors.[iii] Id. § 1526.1(a)–(e). Additionally, the law does not include an express private right of action for employees.

SB 951

Gov. Newsom also signed SB 951 into law, which requires that employers make additional disclosures when the adoption of AI results in events that trigger disclosures under the California Worker Adjustment and Retraining Act (Cal/WARN Act).

Scope

The law will apply to employers that have employed 75 people or more within the preceding 12 months, which is consistent with the Cal/WARN Act’s previous requirements. Id. § 1400.5(b).

Employer Obligations

Effective Jan. 1, 2027, this measure expands the Cal/WARN Act by requiring that employers give notice of “mass layoff, relocation, or termination caused in whole or in substantial part” by AI or similar technologies. Cal. Labor Code § 14001(d). The law preserves the Cal/WARN Act’s existing requirements, and adds new requirements for employer mass layoff, relocation or termination orders that are “caused in whole or in substantial part by an AI system or other automated technology replacing or automating employment positions,” without defining what level of AI involvement satisfies the “in substantial part” threshold. Employers must also include in the notice: (1) the number, occupation and work location of layoffs due to AI automations; (2) the job functions previously performed by workers that will be automated; (3) the category or type of AI that caused the layoffs; and (4) a required statement — “This notice is for a technology displacement” — at the top of the notice. Id. § 1401(d)(1)(A)–(D).

The Employment Development Department must publish on its website a summary of notices received pursuant to the law, as well as a quarterly statewide summary of reported instances where AI or other automated technologies replace or automate employment decisions resulting in mass layoffs, relocation or termination. Id. § 1401(d)(2).

Enforcement

SB 951 does not change the Cal/WARN Act’s existing enforcement framework.

AB 1883

Finally, Gov. Newsom signed AB 1883, a law that limits an employer’s ability to use AI to monitor employees in certain contexts.

Scope

AB 1883 applies to all employers with employees in California, with very limited exceptions, and that use workplace surveillance tools to recognize, or make inferences or predictions about, employee emotions or collect neural data. The statute on its face is limited to “employees” as defined by the law and omits a reference to independent contractors included in an earlier version of the bill. The statute is silent as to whether the employer’s obligations extend to applicants for employment.

The law defines a workplace surveillance tool as “any system, application, instrument, or device that collects or facilitates the collection of employee data, activities, communications, actions, biometrics, or behaviors by means other than direct observation by a person, including, but not limited to, video or audio surveillance, continuous incremental time-tracking tools, geolocation, electromagnetic tracking, photoelectronic tracking, or that utilizes a photo-optical system or other means.” Cal. Labor Code § 1580(g). The law defines neural data as “information that is generated by measuring the activity of an employee’s central or peripheral nervous system, and that is not inferred from nonneural information.” § 1580(e).

In their commentary, the legislature provided two examples of the types of workplace surveillance tools that would be regulated:

• Emotional recognition technology, which uses biometric data such as a person's facial expression, heart rate, skin conductance or tone of voice to infer a person's emotional state, such as emotion recognition tools that allow employers to monitor call center employees to ensure the caller provided a happy and receptive tone when speaking to customers.
• Neural recognition technology, which works by capturing the brain activity of a person's central and peripheral nervous system, which developers claim can determine when a worker is concentrating, tired or stressed.

Although not addressed expressly in the statute, the definition of workplace surveillance tools could be read broadly to include tools that analyze written communications to perform a sentiment analysis to identify and classify the emotional tone (such as positive, negative or neutral) expressed in text data.

Employer Obligations

Starting Jan. 1, 2027, employers are prohibited from using AI workplace surveillance tools to collect neural data of employees or predict employees’ emotional states. Id. § 1581(a)(1)–(2). The prohibition applies to the use of workplace surveillance tools generally, and there is no qualification that the tools be used or relied upon to make employment decisions. Importantly, however, employers are still permitted to use AI workplace surveillance tools “to ensure safety.” Id. § 1581(b).

Employers that are found to have violated this law can incur a civil penalty of $500 per violation. Id. § 1582(d)(1)–(2). Additionally, the labor commissioner and public prosecutors are granted enforcement powers. Id. § 1582(a)–(b). Civil actions are also permitted with remedies, including injunctions and punitive damages. Id. § 1582(c). The current version of the law omits an express private right of action for employees that appeared in earlier iterations but permits public prosecutors to bring a civil action in state court. Id. § 1582(e).

Conclusion

Due to the rapid development of the law and related regulations, California employers that are integrating AI tools should be mindful of the changing legal landscape and adapt their practices accordingly.

Paul Hastings actively monitors updates on this topic, and we are available to assist with emerging compliance challenges.