California Expands Its AI Rulebook: Governor Newsom Signs Over Two Dozen Laws Governing AI Oversight and Use

California closed its 2025-2026 legislative session by enacting a number of AI laws, reinforcing its position at the forefront of AI regulation. On September 9, 2026, Governor Gavin Newsom signed SB 813 and AB 1405, creating a first-in-the-nation framework for independent AI verification organizations and a state registry for AI auditors.¹ On September 30, 2026, he signed 13 more AI-related measures addressing AI in the workplace, AI in health care, content provenance and digital replicas, and attorneys’ use of generative AI.²

Nearly 30 AI-related measures were enacted in 2026. Together with an executive order accelerating implementation of a new oversight regime, these laws signal that California will continue to set the pace for AI regulation in the absence of comprehensive federal action.³

Key Takeaways:

• Independent assessment of AI safety arrives: SB 813 and AB 1405 create a voluntary framework for auditing and verifying AI risks, including establishing a licensing regime for auditors and “independent verification organizations” (IVOs) to register with the state. While audits are not mandatory, undergoing an audit under the framework is “relevant to, but not conclusive of,” liability in litigation over AI-caused harm and is gaining broader national traction.
• Employers face new limits on “robo bosses”: SB 947 requires meaningful human review when employers primarily rely on automated decision systems to discipline or terminate employees, and creates notice and data-access rights for employees.
• AI-driven layoffs and surveillance are now regulated: Employers must disclose when AI causes a mass layoff (SB 951), cannot use AI to read workers’ emotions or brain activity (AB 1883), and cannot use AI to monitor workers in bathrooms (AB 1331).
• Lawyers and arbitrators are covered too: SB 574 regulates attorneys’ use of generative AI, including citation verification, confidentiality, and court-disclosure obligations.
• More is coming: Governor Newsom has ordered the state to move faster and to study even tougher rules, such as an emergency “off switch” for the most powerful AI models.

California Establishes Independent Safety Checks for AI

SB 813 establishes a state framework for independent third-party review of AI systems. The law directs California to designate qualified “independent verification organizations” (IVOs)—independent organizations that can test AI systems and models, assess safety and other risks, and evaluate the methodologies underlying those assessments. The IVO ecosystem is still developing, and it remains unclear which types of organizations will qualify for designation in California. Potential participants include nonprofit AI-safety organizations and independent certification or assurance bodies. To qualify, an organization must demonstrate relevant technical expertise and independence from the AI companies it reviews. A related law, AB 1405, creates a state registration regime for AI auditors, which, beginning January 1, 2029, will require that persons conducting covered AI audits in California are registered and comply with requirements relating to independence, recordkeeping, and professional standards.⁴ The precise relationship between SB 813’s IVO designation and AB 1405 registration remains developing and may depend on the type of audit an IVO performs.

Participation in the SB 813 framework is voluntary.⁵ The law does not require AI developers, deployers, or operators to obtain an audit as a condition of doing business in California. However, SB 813 instructs that obtaining a qualifying audit is “relevant to, but not conclusive of,” liability (or lack thereof) in litigation alleging harm caused by an AI model. This language provides an incentive for companies to obtain independent review, particularly for higher-risk systems. More broadly, as the practice becomes more established, the standards developed through California’s IVO framework could provide a reference point for evaluating duties of care under state tort law.

Further, third-party evaluation is gaining national traction as an AI-safety measure. For example, Anthropic recently committed to embedding independent third-party safety evaluators inside the company to verify compliance with safety practices and assess model alignment and training, and called for other frontier AI companies to do the same.⁶ OpenAI has since followed suit and likewise committed to allowing independent safety evaluations.⁷ Similarly, the White House Accord on Super Intelligence called for companies developing and deploying frontier AI models to partner with independent external auditors or evaluators to assess whether safety controls are functioning as intended.⁸ These developments suggest growing momentum toward independent evaluation as an important measure in addressing AI safety concerns.

The framework will be implemented in stages. SB 813 takes effect January 1, 2027 and generally gives the California Government Operations Agency (GovOps) until January 1, 2028 to establish the program. Executive Order N-9-26 accelerates part of that timeline by directing GovOps to publish IVO application requirements and designation criteria by May 1, 2027.⁹ AB 1405’s registration requirement for covered AI auditors takes effect January 1, 2029.

OMM is a longtime legal advisor to Fathom, the nonprofit that spearheaded the independent verification concept at the heart of SB 813. OMM advised Fathom on legal issues regarding the regulatory framework and draft legislation.

California Requires Human Oversight of AI-Driven Discipline and Termination (the “No Robo Bosses Act”)

SB 947 limits how California employers may use automated decision systems (“ADS”) in disciplinary and termination decisions. These systems can include AI- or algorithm-based tools that monitor or evaluate employee performance, productivity, attendance, or other workplace activity and use that information to recommend or inform disciplinary or termination decisions. Employers may not rely solely on an ADS to discipline or terminate an employee. If an employer primarily relies on ADS output, a human reviewer must independently evaluate and corroborate the output using other relevant information, such as supervisory evaluations, personnel records, work product, peer reviews, or witness interviews. If the output cannot be corroborated, or is found to be inaccurate, incomplete, or misleading, it cannot be used as the basis for the decision. The law also requires employers to provide affected employees with a standalone written notice explaining the role of the ADS and human review, and allows employees to request the data and a description of the data about them used by the system during the preceding 12 months.

SB 947 also prohibits employers from using ADS tools to infer an employee’s protected status under California’s Fair Employment and Housing Act or to predict that an employee will exercise a legal right and take adverse action on that basis. The law applies broadly to computational systems based on machine learning, statistical modeling, data analytics, or AI that generate scores, classifications, or recommendations used to assist or replace human judgment.

SB 947 takes effect July 1, 2027.¹⁰

Technology-Displacement Disclosures in California WARN Act Notices

SB 951 requires additional disclosures when a mass layoff, relocation, or termination covered by the California WARN Act is caused “in whole or in substantial part” by AI or other automated technology. In addition to the information already required in a California WARN Act notice, such as the timing of a mass layoff, the affected worksite, and the job titles and number of affected employees, SB 951 requires that the notice state at the top that “[t]his notice is for a technology displacement,” and identify the number, classification or occupation, and work location of affected positions, the job functions being automated, and the category or type of technology responsible.

Limits on AI-Powered Workplace Surveillance

AB 1883 prohibits employers from using an AI-powered “workplace surveillance tool” to recognize, infer, or predict an employee’s emotional state, or to collect “neural data,” information collected through measuring nervous-system activity.¹¹ “Workplace surveillance tool” is defined broadly to include systems that collect employee data, communications, biometrics, or behavioral information other than by direct human observation, such as video or audio monitoring, time-tracking, geolocation, and software that analyzes emails, chats, or calls.¹² For example, a workplace surveillance tool could include software that analyze workplace chats and uses those signals to classify an employee’s mood, stress level, engagement or other emotional state, or sensor-equipped headphones that measure electrical signals from the brain or nervous system to assess an employee’s attention, fatigue, or other neurological activity. Companion bill AB 1331 bars employers from monitoring or surveilling employees in workplace bathrooms and allows employees to leave surveillance tools on their person behind when entering a bathroom.¹³

AB 1883 applies to private employers of all sizes and covers employees, with an exception for tools used to ensure safety, the scope of which is currently unclear. Both AB 1883 and AB 1331 are expected to take effect January 1, 2027.

Rules for Attorneys’ and Arbitrators’ Use of Generative AI

SB 574 makes California the first state to regulate attorneys’ use of generative AI by statute. It states attorneys may not delegate the practice of law to generative AI, may not enter confidential, personally identifying, or nonpublic information into a generative AI system unless access to the system is restricted to the attorney and authorized individuals bound by confidentiality, and must take reasonable steps to verify AI output and correct erroneous or hallucinated content.¹⁴ Court filings may not contain citations the responsible attorney has not personally verified, and attorneys must disclose generative AI use in documents submitted to the court. The statute does not prescribe the form or placement of that disclosure, specify when it must be made, or expressly identify the degree or type of generative AI use that triggers the disclosure requirement. It separately provides that attorneys should consider whether to disclose generative AI use when it is used to create content provided to the public, but does not provide further criteria for assessing the circumstances or manner in which such disclosure should be made. Arbitrators may not delegate decision-making to generative AI and must disclose to the parties if they rely on AI-generated information.¹⁵

SB 574 applies to attorneys practicing in California, arbitrators and judicial officers, and alternative dispute resolution providers. The statute does not expressly address whether its attorney-specific requirements apply to practice in federal courts or administrative agency proceedings, or to attorneys who are not licensed in California but are admitted to practice before a federal court located in California.

Other AI Laws in the Signing Package

The table below summarizes additional AI-related laws recently enacted by California.¹⁶

Conclusion

California’s 2026 package reinforces several themes we have tracked across the states: a focus on restricting use of AI in employment and workplace decisions, transparency and provenance requirements for AI-generated content, and growing reliance on third-party verification.

Companies operating in California should begin compliance planning now for the January 1 and July 1, 2027 effective dates. Among other things, companies should:

• Review how automated decision systems are currently used in employment-related decisions, and consider whether existing HR processes provide for meaningful human review and appropriate employee notice when ADS tools are involved.
• Review the use of any AI tools for workplace monitoring in light of AB 1883 and forthcoming guidance.
• Verify citations in court filings, including citations generated or suggested by AI tools, and consider whether disclosures regarding generative AI uses are warranted.

For more information, please reach out to one of the attorneys listed in this alert or your regular O’Melveny contact.

¹ Office of Governor Gavin Newsom, Press Release, “Governor Newsom signs first-in-the-nation AI safeguards to protect Californians, calls on the federal government to do its part” (Sept. 9, 2026), https://www.gov.ca.gov/2026/09/09/governor-newsom-signs-first-in-the-nation-ai-safeguards-to-protect-californians-calls-on-the-federal-government-to-do-its-part/.

² Office of Governor Gavin Newsom, Press Release, “California’s nation-leading AI framework just got stronger, Governor Newsom signs more first-in-the-nation worker protections and more” (Sept. 30, 2026), https://www.gov.ca.gov/2026/09/30/californias-nation-leading-ai-framework-just-got-stronger-governor-newsom-signs-more-first-in-the-nation-worker-protections-and-more/.

³ Cal. Exec. Order No. N-9-26 (Sept. 18, 2026), https://www.gov.ca.gov/wp-content/uploads/2026/09/FINAL-N-9-26-AI-EO-9.18.26-SIGNED.pdf.

⁴ A.B. 1405, 2025–2026 Reg. Sess., ch. 178, 2026 Cal. Stat., https://legiscan.com/CA/text/AB1405/id/3458353.

⁵ S.B. 813, 2025–2026 Reg. Sess., ch. 179, 2026 Cal. Stat., https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB813.

⁶ Dario Amodei, “We Must Pace the Frontier” (Sept. 12, 2026), https://darioamodei.com/post/we-must-pace-the-frontier.

⁷ Anthony Ha, “Anthropic CEO Outlines Plan to Slow AI Development,” TechCrunch (Sept. 12, 2026), https://techcrunch.com/2026/09/12/anthropic-ceo-outlines-plan-to-pace-the-frontier/.

⁸ Donald J. Trump (@realDonaldTrump), Truth Social (Sept. 29, 2026), https://truthsocial.com/@realDonaldTrump/posts/117356435739432952.

⁹ Cal. Exec. Order No. N-9-26 (Sept. 18, 2026), https://www.gov.ca.gov/wp-content/uploads/2026/09/FINAL-N-9-26-AI-EO-9.18.26-SIGNED.pdf.

¹⁰ S.B. 947, 2025–2026 Reg. Sess., ch. 859, 2026 Cal. Stat., https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB947&version=20250SB94789CHP.

¹¹ A.B. 1883, 2025–2026 Reg. Sess., ch. 853, 2026 Cal. Stat., https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=202520260AB1883.

¹² A.B. 1883, 2025–2026 Reg. Sess., ch. 853, 2026 Cal. Stat., https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=202520260AB1883.

¹³ A.B. 1331, 2025–2026 Reg. Sess., ch. 851, 2026 Cal. Stat., https://legiscan.com/CA/text/AB1331/id/3461171.

¹⁴ S.B. 574, 2025–2026 Reg. Sess., ch. 858, 2026 Cal. Stat., https://legiscan.com/CA/text/SB574/id/3461240.

¹⁵ S.B. 574, 2025–2026 Reg. Sess., ch. 858, 2026 Cal. Stat., https://legiscan.com/CA/text/SB574/id/3461240.

¹⁶ Office of Governor Gavin Newsom, Press Release, “California’s nation-leading AI framework just got stronger, Governor Newsom signs more first-in-the-nation worker protections and more” (Sept. 30, 2026), https://www.gov.ca.gov/2026/09/30/californias-nation-leading-ai-framework-just-got-stronger-governor-newsom-signs-more-first-in-the-nation-worker-protections-and-more/.

¹⁷ S.B. 1119, 2025–2026 Reg. Sess., https://legiscan.com/CA/text/SB1119/id/3405020.

This memorandum is a summary for general information and discussion only and may be considered an advertisement for certain purposes. It is not a full analysis of the matters presented, may not be relied upon as legal advice, and does not purport to represent the views of our clients or the Firm. Jonathan P. Schneller , an O’Melveny partner licensed to practice law in California; Reema Shah, an O’Melveny partner licensed to practice law in New York; Eric Amdursky, an O’Melveny partner licensed to practice law in California; Adam Karr, an O’Melveny partner licensed to practice law in California and Utah; Adam P. KohSweeney, an O’Melveny partner licensed to practice law in California and New York; Mark Liang, an O’Melveny partner licensed to practice law in California; James K. Rothstein, an O’Melveny partner licensed to practice law in California; Daniel R. Suvor, an O’Melveny partner licensed to practice law in California; Damali A. Taylor, an O’Melveny partner licensed to practice law in California and New York; Kelly Wood, an O’Melveny partner licensed to practice law in California; Sergei Zaslavsky, an O’Melveny partner licensed to practice law in Maryland and the District of Columbia; and Rosie Kaur, an O’Melveny associate licensed to practice law in New York, contributed to the content of this newsletter. The views expressed in this newsletter are the views of the authors except as otherwise noted.

© 2026 O’Melveny & Myers LLP. All Rights Reserved. Portions of this communication may contain attorney advertising. Prior results do not guarantee a similar outcome. Please direct all inquiries regarding New York’s Rules of Professional Conduct to O’Melveny & Myers LLP, 1301 Avenue of the Americas, Suite 1700, New York, NY, 10019, T: +1 212 326 2000.