Which 2025 AI bills became law in California?
Californians want AI regulation.
This was the big takeaway of our 2025 AI Compass report detailing how Californians really feel about AI. An overwhelming majority specifically favors policies that protect privacy (81%), enforce civil rights (73%), and enact non-discrimination rules (73%). With a more Big Tech-friendly federal administration and Californians trusting state legislators over the federal government to begin with, it’s more important than ever for California legislators to step up to the plate.
So how did they measure up in the 2025 legislative session?
Over 60 bills related to AI were introduced at the start of the California Legislative session, similar to the number of AI bills introduced last year. After going through the California Assembly and/or Senate, 28 made it to the Governor’s desk and, from there, Governor Newsom signed 21 into law. Read on to get an overview of the AI bills that shaped the session, what moved forward and what didn’t, and what became law.
Need a general primer on AI before diving in? Check out our blog What exactly is Artificial Intelligence? And other questions, answered before diving in.
AI & Pricing
Signed Into Law
Authored by Assemblymember Cecilia Aguiar-Curry (D-Winters) and sponsored by TechEquity, the Preventing Algorithmic Price Fixing Act (AB 325) will hold developers and users of pricing algorithms accountable for price fixing. The bill generally prohibits the use or distribution of any pricing algorithm that sets or recommends prices using data from more than one business in a similar market, ensuring that consumers won’t pay more when companies attempt to use digital tools to break the rules. The bill received support from California Attorney General Rob Bonta, and on October 7, Governor Gavin Newsom signed AB 325 into law.
Vetoed
None.
Did Not Pass
TechEquity also sponsored the End AI Rent Hikes Act (SB 52), authored by Senator Sasha Renée Pérez (D–Alhambra), which would have banned algorithms used by landlords to hike rents.
Legislators introduced two other bills to address algorithmic price-fixing and collusion: the California Preventing Algorithmic Collusion Act (SB 295), authored by Senator Melissa Hurtado (D-Bakersfield), and the Preventing Algorithmic Price Fixing Act (SB 384), authored by Senator Aisha Wahab (D-Hayward). Both would have limited the distribution or making of recommendations based on the use of a pricing algorithm.
AB 446 by Assemblymember Christopher Ward (D-Bakersfield) and SB 259 by Senator Wahab (D-Hayward) both would have targeted personalized pricing. AB 446’s focus was on banning retailers from engaging in surveillance pricing, while SB 259 would have prohibited the use of affinity-based pricing, which combines personalized pricing and dynamic pricing.
The California Consumer Privacy Act of 2018 (CCPA) gives consumers the right to limit the use of sensitive personal information collected by businesses. SB 435 by Senator Wahab (D-Hayward) would have removed a provision in the law that states that publicly available personal information is not considered sensitive personal information or personal information, and so not covered by the California Consumer Privacy Act. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
AI & Discrimination
Signed Into Law
None.
Vetoed
None.
Did Not Pass
The Automated Decisions Safety Act (AB 1018), a bill by Assemblymember Rebecca Bauer-Kahan (D-San Ramon) and sponsored by TechEquity and SEIU California, would have provided people with more information, important rights, and greater transparency into the use of Automated Decision Systems for critical areas of their lives—while requiring developers and deployers to take more responsibility for reducing the likelihood of discrimination.
With more work to do to ensure the bill would be signed into law, Assemblymember Bauer-Kahan designated AB 1018 as a 2-year bill. This means the bill will return as part of the legislative session next year, allowing time for continued conversations.
A bill by Senator Steve Padilla (D-San Diego), AB 420, would have required AI developers to provide a tool for users to detect AI-generated content and generally regulate a developer or a deployer of a high-risk automated decision system. It would have also prohibited a state agency from awarding a contract for a high-risk automated decision system unless the person to whom the contract is awarded has certified that the high-risk automated decision system does not violate any civil rights laws. [This bill also appears in the AI & Safety section]
SB 503, authored by Senator Akilah Weber Pierson (D-San Diego), would have required that health care providers make reasonable efforts to identify when patient care decision support tools use input variables that reflect protected characteristics, as well as mitigate any potential risk of discrimination that could result from the use of those tools. The bill was placed in the inactive file, so it may return next year. [This bill also appears in the AI & Health section]
AI & Education
Signed Into Law
Authored by Senator Sabrina Cervantes (D-Corona), SB 241 is not an AI bill within itself, but does stipulate that the amendment that will be made to a section of the Education Code in relation to community college will not prohibit community college staff from using AI tools to assist in the operations of a community college or in providing services to community college students. [This bill also appears in the AI & Labor section]
Vetoed
None.
Did Not Pass
AI feeds off of information. To protect students, AB 1159 by Assemblymember Dawn Junea Addis (D-Morro Bay) would have prohibited an AI operator from using information, including persistent unique identifiers, created or gathered by the operator’s site, service, or application to train a generative AI system or service or develop an AI system unless it is in the furtherance of the relevant type of school purpose and for the use and benefit of the school. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
AB 1053, authored by Assemblymember Rick Chavez Zbur (D-Los Angeles), would have created councils of parents, teachers, and district IT to make sure AI used in the classroom is appropriate, meets standards, and more.
A bill by Assemblymember Marc Berman (D-Palo Alto), AB 887, would have established the California Computer Science Demonstration Grant Program to support high schools in creating, maintaining, and/or expanding computer science education, which includes AI. The bill was referred to the suspense file—a way to hold a piece of legislation to consider the fiscal impacts of implementation.
Two other bills would have also tried to set up California students of all ages to learn about AI. AB 279 by Assemblymember Darshana Ramesh Patel (D-San Diego) would have urged the state superintendent to regularly convene a panel of experts in consultation to develop recommendations for revisions to school library standards on a range of educational topics, including “trending technologies, media literacy, artificial intelligence, and social media.”
Authored by Assemblymember Roger Niello (R-Fair Oaks), SCR 82 would have encouraged the President of the University of California, the Chancellor of the California State University, and the Chancellor of the California Community Colleges to create a workgroup of faculty, staff, and administrators to review the use of AI in higher education, and create a report and make public the strategies and best practices for artificial intelligence usage agreed upon by the workgroup.
AI & Kids
Signed Into Law
SB 857, authored by Jesse Arreguín (D-Berkeley), is a general public safety bill that includes provisions on digitally altered or artificial-intelligence-generated obscene matter depicting a person under 18 years of age personally engaging in or personally simulating sexual conduct.
Gavin Newsom signed SB 857 into law, along with Assemblymember Buffy Wicks (D-Oakland)’s AB 1043, which will require platforms to collect age identification and provide it to app developers based on service need.
This year, there has been increasing concern around AI psychosis and the impact of AI chatbots on children. Authored by Senator Padilla (D-San Diego), SB 243 will attempt to address some of these concerns by requiring chatbots to disclose that they are not human and obliging platforms to intervene and redirect toward real crisis support in situations of self-harm and limit certain forms of sexualized interaction with minors. [This bill also appears in the AI & Safety section]
Vetoed
Despite child advocates arguing that the Leading Ethical AI Development (LEAD) for Kids Act (AB 1064) by Assemblymember Bauer-Kahan (D-San Ramon) would better protect kids in comparison to SB 243, Governor Newsom chose to veto this bill. AB 1064 would have required risk-level assessments by developers to classify AI systems based on their potential harm to children, prohibited certain harmful AI applications for children (ex, emotionally manipulative AI chatbots), and strengthened privacy protections by requiring affirmative, written parental consent before a child’s personal information can be used to train an AI model.
Did Not Pass
Existing law requires a social media platform to provide a reporting mechanism for child sexual abuse material. AB 1137, authored by Assemblymember Maggy Krell (D-Sacramento), would have deleted the requirement that the reporting user be depicted in the material, would require that the depicted individual in the reported image be depicted as a minor, and would additionally require the reporting mechanism to be clear and conspicuous.
Assemblymember Josh Lowenthal (D-Long Beach) reintroduced his bill to hold social media companies financially accountable for harm caused to children and teens, AB 2. The bill passed the Assembly but didn’t make it through the Senate.
AB 970 by Assemblymember Tina McKinnor (D-Hawthorne) would have authorized the County of Los Angeles to establish a 2-year pilot program to test a new model for the mandatory reporting of child abuse or neglect, including the development and deployment of an internet-based decision-support tool.
AI & Health
Signed Into Law
To tackle AI-generated medical misinformation, Assemblymember Mia Bonta (D-Alameda) authored AB 489, which was signed into law by Governor Newsom. AB 489 will prohibit AI or GenAI technology from using certain terms, letters, or phrases that indicate or imply that the advice, care, reports, or assessments being provided through AI or GenAI is being provided by a natural person with the appropriate health care license or certificate.
The Governor also signed into law AB 56 by Assemblymember Bauer-Kahan (D-San Ramon), which will require social media platforms to have a black box warning label consistent with a recent call to action by the U.S. Surgeon General’s Office about addictive feeds.
Vetoed
Two other bills in this category made it to the Governor’s desk but were ultimately vetoed. AB 512, authored by Assemblymember Harabedian (D–Sierra Madre), would have put guardrails on AI approvals/denials of health insurance coverage and shortened the timeline for prior authorization requests. It would have also put guardrails on approvals/denials made by artificial intelligence.
Authored by Assemblymember Liz Ortega (D-Hayward), AB 682 would have mandated more thorough healthcare claims reports that would include the number of claims denied or partially denied that at any point were processed, adjudicated, or reviewed with AI or other predictive algorithms.
Did Not Pass
A proposal on AI and mental health, authored by Senator Padilla (D-San Diego), SB 579 would have required the secretary of the Government Operations Agency to appoint a mental health and AI working group that would evaluate certain issues to determine the role of artificial intelligence in mental health settings.
Authored by Senator Monique Limón (D-Santa Barbara), the Insurance Consumer Privacy Protection Act (SB 354) would have established new standards for the collection, processing, retaining, or sharing of consumers’ personal information by insurance licensees and their third-party service providers and given consumers more rights to explanations around insurance decisions. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
SB 503, authored by Senator Weber Pierson (D-San Diego), would have required that health care providers identify when patient care decision support tools use input variables that reflect protected characteristics and mitigate any potential risk of discrimination that could result from the use of those tools. The bill was placed in the inactive file, so it may return next year. [This bill also appears in the AI & Discrimination section]
AI & Labor
Signed Into Law
Authored by Senator Cervantes (D-Corona), SB 241 is not an AI bill within itself, but does stipulate that the amendment that will be made to a section of the Education Code, which ensures instructors are faculty members and not AI, will not prohibit community college staff from using AI tools to assist in the operations of a community college or in providing services to community college students. [This bill also appears in the AI & Education section]
Vetoed
The No Robo Bosses Act (SB 7) by Senator Jerry McNerney (D-Stockton) was vetoed by Governor Newsom. The bill would have required that employers provide pre- and post-use notification to workers when an Automated Decision System is being used on them. It would also have established a process for workers to appeal a decision made by an employer utilizing an ADS. TechEquity supported this bill as part of our 2025 priority legislative slate.
In 2023, members of the Screen Actors Guild and the American Federation of Television and Radio Artists (SAG-AFTRA) went on strike and won detailed provisions governing the creation and use of both “Digital Replicas” of individual performers and entirely AI-generated “Synthetic Performers.” To further protect performers, Senator Angelique Ashby (D-Sacramento) introduced SB 11, which would have defined various terms related to AI and digital replication and created additional regulations and limits around the use of digital replicas. The bill, though, was ultimately vetoed by Governor Newsom.
Did Not Pass
The public is increasingly realizing that AI is not all that artificial. Generative AI, in particular, involves training algorithms on existing data and media, which is then used to generate “new” content. There have been several instances in which this was done without the consent of those who created the media used in the training process, which some have argued violates copyright laws.
Assemblymember Bauer-Kahan (D-San Ramon) put forth AB 412, a bill that would have addressed the concerns of writers, actors, musicians, and other artists by requiring developers of generative AI models to document and publish information on some categories of copyrighted material used during the training process. The bill was pulled as it was moving through the California Senate committees. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
A bill by Assemblymember Sade Elhawary (D-Los Angeles), AB 1331, would have updated and expanded existing workplace privacy laws to address new surveillance technology. In an attempt to get ahead of technological advancements, it would have also prohibited employers from requiring workers to implant or embed tracking devices in their bodies. The bill was ordered to the inactive file.
SB 366, by Senator Lola Smallwood-Cuevas (D-Los Angeles), was placed on the suspense file. If passed, it would have required the Department of General Services to contract with the University of California, Los Angeles Labor Center to conduct a study evaluating the impact of AI on worker well-being, job quality, job types, different populations, and state revenues.
AB 1221, authored by Assemblymember Isaac Bryan (D-Los Angeles), would have prohibited the use of invasive and potentially discriminatory surveillance in the workplace and would have established much-needed worker data protections.
AI & Data, Surveillance, and Privacy
Signed Into Law
This year, the California Privacy Protection Agency (CPPA) finalized rules around technologies like Automated Decision Systems. Meanwhile, Assemblymember Lowenthal (D-Long Beach) introduced the California Opt Me Out Act (AB 566), which requires browsers to include a setting that enables a consumer to send an opt-out preference signal, giving Californians a simple way to exercise their digital privacy rights.
SB 361, authored by Senator Josh Becker (D-Menlo Park), will require data brokers to provide additional information to the government, including whether the data broker collects consumers’ names, dates of birth, or login or account information, citizenship data, union membership status, sexual orientation status, gender identity and gender expression data, and biometric data. The bill will also require a data broker to provide information regarding whether, in the past year, the data broker shared or sold consumers’ data to a foreign government.
Authored by Assemblymember Bauer-Kahan (D-San Ramon), AB 45 is also an attempt to protect consumer data, particularly the face of the practice of geofencing, a data harvesting tactic that creates a virtual boundary around a physical area, which triggers an action when a device enters or exits it. In order to protect people’s healthcare privacy, especially at sensitive locations like family planning centers, AB 45 will prohibit geofencing, or selling or sharing personal information with a third party to geofence, a place that provides in-person health care services in California.
Assemblymember Jacqui Irwin (D-Thousand Oaks) put forth AB 979, which Governor Newsom signed. AB 979 will require the California Cybersecurity Integration Center to develop a California AI Cybersecurity Collaboration Playbook to facilitate information sharing across the cyber and AI communities and to strengthen collective cyber defenses against emerging threats. [This bill also appears in the AI & Government section]
A bill by Senator Arreguín (D-Berkeley), SB 524, mandates that any police report written using AI contain disclaimers that AI was used and that the first draft must be maintained. It also requires officers to read and verify the contents of the report and sign off, and bans AI vendors from sharing the information a police agency provided to the AI. [This bill also appears in the AI & Government section]
Vetoed
Authored by Senator Cervantes (D-Corona), SB 274 would have prevented entities with access to the government’s license plate tracking system from selling data collected through license plate recognition. Governor Newsom vetoed the bill.
Did Not Pass
Assemblymember Ward (D-San Diego) introduced 3 different bills under this theme.
- AB 1355 would have prohibited some businesses from collecting or processing the location information of an individual unless doing so is necessary to provide goods or services requested by that individual.
- AB 1337 would have tightened restrictions on how agencies can use and disclose personal information and prohibited agencies from using personal information for purposes other than those for which it was collected, except as required by law.
- Lastly, AB 322 would have prohibited the use of geolocation data outside of a specific business or service purpose.
TechEquity signed a letter opposing SB 690 by Senator Anna Caballero (D-Merced), which would which would amend the California Invasion of Privacy Act (CIPA) to provide a broad loophole, allowing businesses to collect sensitive personal data such as immigration status or healthcare conditions, as long as they can justify it with “commercial business purpose,” and would place no restrictions on sharing that information with data brokers, immigration officials, or anyone else.
Authored by Senator Limón (D-Santa Barbara), the Insurance Consumer Privacy Protection Act (SB 354) would have established new standards for the collection, processing, retaining, or sharing of consumers’ personal information by insurance licensees and their third-party service providers and given consumers more rights to explanations around insurance decisions. [This bill also appears in the AI & Healthcare section]
SB 468 by Senator Becker (D-Menlo Park) would have required businesses that deploy high-risk AI systems processing personal information to establish and maintain a comprehensive security program to protect consumers.
Assemblymember Bauer-Kahan (D-San Ramon) put forth AB 302, which would have required the automatic deletion of personal information on state and local elected officials and judges from data broker registries. [This bill also appears in the AI & Government section]
The California Consumer Privacy Act of 2018 (CCPA) gives consumers the right to limit the use of sensitive personal information collected by businesses. SB 435 by Senator Wahab (D-Hayward) would have removed a provision in the law that states that publicly available personal information is not considered sensitive personal information or personal information, and so not covered by the California Consumer Privacy Act. [This bill also appears in the AI & Pricing section]
SB 44, authored by Senator Tom Umberg (D-Santa Ana), would have required, under the CCPA, a covered business to use neural data only for the purpose for which the neural data was collected and would require a covered business to delete neural data when the purpose for which the neural data was collected is accomplished.
To protect students, AB 1159 by Assemblymember Addis (D-Morro Bay) would have prohibited an AI operator from using information, including persistent unique identifiers, created or gathered by the operator’s site, service, or application to train a generative AI system or service or develop an AI system unless it is in the furtherance of the relevant type of school purpose and for the use and benefit of the school. [This bill also appears in the AI & Education section]
Lastly, for this section, AB 1300 by Assemblymember Jessica Caloza (D-Los Angeles) would have limited the sharing of data with immigration enforcement to the fullest extent of the law.
AI & Disinformation
Signed Into Law
The California AI Transparency Act (AB 853) by Assemblymember Wicks (D-Oakland) was signed by the Governor and will ensure online platforms make it clear when content is AI-generated, altered, or authentic.
The Governor also signed AB 723, authored by Assemblymember Gail Pellerin (D-Santa Cruz), which will require a real estate broker or salesperson who includes a digitally altered image, as defined, in an advertisement or other promotional material for the sale of real property to include a disclosure.
Signed into law by the Governor, AB 621 by Assemblymember Bauer-Kahan (D-San Ramon), will strengthen civil enforcement options against “nudify” websites that use AI to generate fake nude photos of real people without their consent.
Vetoed
None.
Did Not Pass
Authored by Assemblymember Diane Dixon (R-Newport Beach), AB 392 would have also targeted AI deepfakes, requiring websites to take precautions around them as well as non-AI revenge pornography.
AB 410 by Assemblymember Lori Wilson (D-Fresno) would have required someone using a bot to disclose that the bot is a bot and not a human to whoever they’re communicating with.
Assemblymember Bauer-Kahan (D-San Ramon) put forth AB 412, which would have required developers of generative AI models to document and publish information on some categories of copyrighted material used during the training process. [This bill also appears in the AI & Labor section.]
A bill by Assemblymember Pellerin (D-Santa Cruz), AB 502, would have prohibited someone from distributing media that portrays candidates and officials as doing or saying something that they did not do or falsely undermines confidence in the outcome of an election. [This bill also appears in the AI & Government section]
AI & Government
Signed Into Law
A bill by Senator Arreguín (D-Berkeley), SB 524, mandates that any police report written using AI contain disclaimers that AI was used and that the first draft must be maintained. It also requires officers to read and verify the contents of the report and sign off, and bans AI vendors from sharing the information a police agency provided to the AI. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
Assemblymember Jacqui Irwin (D-Thousand Oaks) put forth AB 979, which Governor Newsom signed. AB 979 will require the California Cybersecurity Integration Center to develop a California AI Cybersecurity Collaboration Playbook to facilitate information sharing across the cyber and AI communities and to strengthen collective cyber defenses against emerging threats. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
Vetoed
None.
Did Not Pass
Assemblymember Bauer-Kahan (D-San Ramon) put forth AB 1405 and AB 302. AB 1405 would have required the Government Operations Agency to establish a mechanism on the agency’s website, allowing AI auditors to enroll with the agency and allowing natural persons to report misconduct by an enrolled AI auditor. AB 302 would have required the automatic deletion of personal information on state and local elected officials and judges from data broker registries. [This bill also appears in the AI & Data, Surveillance, and Privacy section]
A bill by Assemblymember Pellerin (D-Santa Cruz), AB 502, would have prohibited someone from distributing media that portrays candidates and officials as doing or saying something that they did not do or falsely undermines confidence in the outcome of an election. [This bill also appears in the AI & Disinformation section]
SB 69, authored by Senator McNerney (D-Stockton), would have required the Attorney General to establish and maintain a specified program to build internal expertise in artificial intelligence, including its applications, risks, regulatory implications, and civil rights impacts.
A bill by Assemblymember Stephanie Nguyen (D-Elk Grove), AB 1242, would have required that the use of AI or other technology-enabled translation services be reviewed and corrected by a person before being distributed, ensuring accurate translation. AB 1242 also establishes other requirements for language access not related to AI.
AI & Environment
Signed Into Law
With data centers popping up across the country to meet the needs of AI, the public is increasingly worried about the impacts of AI on electricity bills and the environment. SB 57 by Senator Padilla (D-San Diego) will require the Public Utilities Commission to study the impact of data centers on costs for other customers and report their findings to the legislature.
Vetoed
Another bill attempting to address this issue, AB 93, was vetoed by the Governor. Authored by Assemblymember Diane Papan (D-San Mateo), this bill would have required data center operators to report their estimated water usage.
Did Not Pass
AB 222, authored by Assemblymember Bauer-Kahan (D-San Ramon), would have created transparency around energy usage by developers of AI models and data centers and energy efficiency standards for data centers, as well as prevented the cost of constructing new data centers from being passed onto ratepayers.
To incentivize data centers to decrease their carbon emissions, Senator Padilla (D-San Diego) authored SB 58, which would have provided tax credits to data centers utilizing more carbon-free energy, not using diesel fuel, and utilizing recycled water cooling. It would also require the facility to invest in and use a skilled and trained workforce.
Assemblymember Papan (D-San Mateo) took a different approach to incentives with AB 1095, providing a subsidy for data centers through financing and renewable energy credits for building waste heat recapture projects if the data center operators reuse the heat generated.
SB 620 by Senator Henry Stern (D-Malibu) would have required the State Energy Resources Conservation and Development Commission (Energy Commission) to create the Data Access Governance Committee and would create the Customer Energy Utility Data Bill of Rights, with specified protections and customer rights.
AI & Safety
Signed Into Law
After Governor Newsom vetoed SB 1047 last year despite wide support, Senator Scott Wiener (D-San Francisco) returned with the Transparency in Frontier Artificial Intelligence Act (SB 53), which Newsom signed into law. SB 53 requires advanced AI developers to publish governance frameworks and transparency reports, establishes mechanisms for reporting critical safety incidents, extends whistleblower protections, and calls for the development of a public computing cluster.
With increasing concern around AI psychosis and the impact of AI chatbots on children, Senator Padilla (D-San Diego) authored SB 243. SB 243 will make companion AI chatbot platforms safer to use by requiring chatbots to disclose that they are not human and obliging platforms to intervene and redirect toward real crisis support in situations of self-harm and limit certain forms of sexualized interaction with minors. [This bill also appears in the AI & Kids section]
SB 480, authored by Senator Bob Archuleta (D-Pico Rivera), also got the Governor’s signature. SB 480 will authorize an autonomous vehicle to be equipped with marker lamps to alert drivers that the vehicle is being operated by an automated driving system (ADS).
Vetoed
None.
Did Not Pass
Two other bills targeting autonomous vehicles did not fare as well. SB 511, authored by Senator Pérez (D–Alhambra), would have brought autonomous vehicles or Level 2 (partial driving automation) vehicles under the DMV purview, establishing a special training and permitting requirement for all drivers of Level 2 vehicles under DMV authority.
Assemblymember Aguiar-Curry (D-Winters) also introduced a bill around autonomous vehicles, AB 33, which would have prohibited the delivery of commercial goods through an autonomous vehicle without a human safety operator on any highway within the State of California. The bill was ordered to the inactive file, so it may return next year.
Authored by Assemblymember Petrie-Norris (D-Irvine), AB 270 would have established a pilot project to assess whether a firefighting helicopter equipped with autonomous aerial suppression technology can be transitioned into operational use in the State of California.
Senator McNerney (D–Stockton) authored two bills under this category. SB 833 would have required human oversight of AI systems and the Department of Technology to develop specialized training in AI safety protocols and risk management techniques for oversight personnel. SB 813 would have protected AI developers from liability for certain harms caused by their AI systems if a private multistakeholder regulatory organization certified the safety of their AI models and AI applications.
AB 420 by Senator Padilla (D-San Diego) would have required AI developers to provide a tool for users to detect AI-generated content and generally regulate a developer or a deployer of a high-risk automated decision system. It would have also prohibited a state agency from awarding a contract for a high-risk automated decision system unless the person to whom the contract is awarded has certified that the high-risk automated decision system does not violate any civil rights laws. [This bill also appears in the AI & Discrimination section]
AI & Law
Signed Into Law
Earlier this year, the California State Bar got in hot water after it turned out that they used AI to write some of the multiple-choice questions in the February 2025 State Bar exam. Signed by Governor Newsom, SB 47 by Senator Umberg (D-Santa Ana), will set into motion an audit of the events that led to this. Senator Umberg also authored SB 253, which will require the Committee of Bar Examiners to give notice if AI is used in creating questions for or grading the bar examination.
AB 316, authored by Assemblymember Krell (D-Sacramento), prohibits a defendant who developed, modified, or used AI from asserting a defense that the AI autonomously caused the harm to the plaintiff, negating their own responsibility. The bill was signed into law by Governor Newsom.
Vetoed
None.
Did Not Pass
None.
Lessons for next year’s session
If you’ve made it to the end of this round-up, congratulations! It was a doozy! AI is an expansive topic that touches on many areas of the law and our lives, and there are sure to be more bills on the topic in the 2026 legislative session.
Of the more than 60 bills that entered the fray at the start of this session, only 28 made it to the Governor’s desk, 21 of which he signed and 7 of which he vetoed. Many of those bills were deeply amended, cut back, and watered down through aggressive and persistent industry lobbying.
If you want to be a part of building a people-first approach to tech and AI, one of the most important factors in determining our legislative slate is how everyday people are affected by the tech economy. So if you have a story about the tech economy, we want to hear it.
By sharing your experience working under surveillance technology or as a renter who has been denied housing by an algorithm, you can help us develop bills and then help lawmakers understand why they should support these bills to close the gaps and protect everyday people.
Sign up for our newsletter to stay up-to-date on the bills we’re running and other work we’re doing to make tech work for everyone.