Before the California Legislative term concluded, we reported on the proposed bill, AB 2930, which aimed to limit bias in artificial intelligence (AI) and automated decision-making tools.

Recently, the San Francisco Business Times reported that Assemblymember Rebecca Bauer-Kahan held the bill back from a vote, writing,"[g]etting the policy right is priority one” and “I look forward to putting forth a stronger bill next year that protects Californians from biased ADTs in all necessary spaces with strong enforcement.”

Despite AB 2930 being held back, not all AI-related bills were on the Legislative chopping block. Here are a few AI-related bills companies should take note of before they take effect soon.

AB 3030

Concerning healthcare providers, AB 3030 regulates the use of generative AI when sending written or verbal communications to patients about their clinical information. The bill requires a healthcare provider that uses generative AI to generate communications, including a (1) disclaimer that notifies the patient of the AI-generated communication and (2) clear instructions on how the patient may contact a human healthcare provider. 

However, there are exceptions to this rule. When a “communication” has been read and reviewed by a human licensed healthcare provider, the communication is exempt from AB 3030’s requirements. In addition, AB 3030’s focus on “patient clinical information”, meaning, “information relating to the health status of a patient” and not “administrative matters, including, but not limited to, appointment scheduling, billing, or other clerical or business matters” emphasizes the Legislature’s desire to ensure that personal health diagnoses and doctor-patient communications are not supplanted or replaced by AI-generated medical instructions. 

SB 1120

For healthcare insurance providers, SB 1120 regulates how AI, algorithms, or other software can be used to decide on a health plan's prior authorization and management. Currently, existing laws do not directly address how AI is regulated when used to make these decisions. SB 1120 clarifies this gray area and will enact specific requirements and prohibitions on how AI is used by insurers and healthcare providers when deciding on healthcare service plans. First, SB 1120 prohibits decisions from being solely on a group dataset. The utilization review process will need to be based on the following factors, including (1) the patient’s medical or clinical history, (2) individual clinical circumstances presented by the requesting provider, and (3) other relevant clinical information in the medical or clinical record. Next, a licensed physician or licensed healthcare professional must make a medical necessity determination to evaluate the specific clinical issues involved in the healthcare service requested by the provider. Moreover, SB 1120 requires the AI, algorithm, or software systems in use to be open to inspection for audit or compliance reviews based on state and federal law, as well as, periodic reviews and revisions to maximize accuracy and reliability in its usage. As a result, healthcare insurers and providers should stay alert about how they utilize and deploy AI systems when crafting healthcare plans and evaluating healthcare service requests from providers.

AB 2602 and AB 1836

In the entertainment and media industry, AB 2602 requires a provision in a contract between an individual and any other person for performance or professional services, only as it relates to a new performance, to be unenforceable if the digital replica of the individual’s voice or likeness is used in lieu of the individual’s actual performance. A “digital replica” is defined as “a computer-generated, highly realistic electronic representation that is readily identifiable as the voice or visual likeness of an individual that is embodied in a sound recording, image, audiovisual work, or transmission in which the actual individual either did not actually perform or appear, or the actual individual did perform or appear, but the fundamental character of the performance or appearance has been materially altered, except as prescribed.” 

Similarly, AB 1836 amended existing law regarding the unauthorized use of digital replicas of a deceased personality’s name, voice, signature, photograph, or likeness in any manner. Starting on January 1, 2025, any person who produces, distributes, or makes a digital replica of a deceased personality’s voice or likeness in an audiovisual or sound recording without specified prior consent, is liable to the injured party for an amount of at least $10,000 or the actual damages suffered. The rights under AB 1836 are property rights that are freely transferrable by means of contract, trust, or any other testamentary instrument.

When Governor Newsom signed AB 2602 and AB 1836 into law, SAG-AFTA’s National Executive Director & Chief Negotiator, Duncan Crabtree-Ireland, applauded the Governor’s decision, stating, “AB 1836 and AB 2602 represent much-needed legislation prioritizing the rights of individuals in the A.I. age. No one should live in fear of becoming someone else’s unpaid digital puppet. Gov. Newsom has led the way in protecting people - and families - from A.I. replication without real consent.”

This brief overview of just a few AI-related bills demonstrates the steps the California Legislature is taking to stay ahead of AI regulation and protect individuals who have been or will be impacted by the use of AI systems. However, the Legislature toes a blurry line between protecting the public, while not stifling AI development and its beneficial uses. As Assemblymember Bauer-Kahan explained: “getting the policy right is priority one”.