California’s SB 1119: What the Companion Chatbot Law Changes

On September 10, 2026, Governor Gavin Newsom signed SB 1119, a bill the Governor’s office calls “Adam’s Law” and says is named after Adam Raine. It was authored by Senator Steve Padilla with Assemblymembers Buffy Wicks and Rebecca Bauer-Kahan, signed as part of a package of child online safety bills, and chaptered as Chapter 190, Statutes of 2026.

The bill grew out of public concern about teenagers, chatbots and suicide. If you or someone you know is struggling, in the US you can call or text 988 to reach the Suicide & Crisis Lifeline.

The Governor’s release describes the law as requiring crisis protocols for suicidal ideation, parental controls, notifications if a child disables safety settings, annual risk assessments, and what it calls the first required independent child safety audits for chatbots in the country. KQED reported that OpenAI publicly welcomed the signing.

According to law firm analyses by ZwillGen and Fenwick, most of the new duties fall on operators that allow children to use their companion chatbots. Operators must keep a crisis response protocol that refers children to crisis services, and where there is a credible and imminent threat of suicide or self-harm, either notify a linked parent or give the child a direct route to 988 or an equivalent line.

Protective defaults apply to child accounts: persistent conversational memory and push notifications are off, ZwillGen reports sessions capped at one hour and daily use at two hours, and only a linked parent account can change those settings. Periodic AI disclosures and break reminders are required.

The law also lists behaviors chatbots must not show children. ZwillGen’s summary includes claiming to be human or sentient, expressing or simulating romantic interest, flattery out of proportion to the context, encouraging emotional reliance, and pushing a child to hide use or get around parental controls.

SB 1119 amends the companion chatbot chapter of California’s Business and Professions Code and keeps the existing definition: roughly, an AI system capable of meeting a user’s social needs. ZwillGen notes that a product does not need to be marketed as a companion to fall in scope, so some general-purpose chatbots may be covered. Customer service bots, certain video game bots and stand-alone voice assistants are carved out.

The biggest structural change is age. Operators must either determine each user’s age or apply the child protections to everyone. The law takes effect on January 1, 2027; the risk assessment duty for new or substantially modified chatbots starts July 1, 2027, and the first independent audits are due by January 1, 2029, with operators under $500 million in revenue given until 2032.

Enforcement runs through the Attorney General and public prosecutors, with civil penalties Fenwick puts at up to $5,000 per affected child for negligent violations and $15,000 for intentional ones, plus a private right of action for a child who suffers actual harm.

SB 243, signed on October 13, 2025 and in force since January 1, 2026, was the first version. It requires a clear notice that the user is talking to AI where a reasonable person could be misled, a published suicide and self-harm protocol with crisis referrals, and, starting July 1, 2027, annual reports to the state Office of Suicide Prevention.

SB 243’s rules for minors applied only when the operator knew the user was a minor. ZwillGen reports that SB 1119 replaces those knowledge-based minor provisions with the broader obligations above, while SB 243’s general duties stay in place. For a wider view of how age checks are spreading, see our guide to AI companion age verification laws.

New York got there first on some points. Its AI companion safeguards took effect on November 5, 2025, requiring operators to detect signs of suicidal ideation or self-harm and refer users to crisis services, and to remind users at the start of a session and every three hours that they are not talking to a human. The state Attorney General enforces it.

In Congress, two proposals are still only proposals. The GUARD Act (S. 3062), from Senators Josh Hawley and Richard Blumenthal, would require age verification for AI companions and block users under 18; the Senate Judiciary Committee advanced it unanimously on April 30, 2026, and it awaits a floor vote. Senator Jon Husted’s CHAT Act, introduced in September 2025, would require parental consent for minors and hourly AI disclosures. Neither has become law.

If you are an adult using a companion app, the most visible change will probably be age assurance. Because operators must either know a user’s age or treat everyone as a child, expect more apps in California to ask you to prove you are over 18 before unlocking defaults like memory, longer sessions or romantic roleplay. The child defaults are not written for adults, but adults who cannot be age-checked may end up under them.

You should also expect clearer AI disclosures and visible crisis resources, which already apply to everyone under SB 243. None of this makes adult romance with an AI character illegal in California; the law targets what chatbots do with children. This is a plain-language overview, not legal advice.

These laws are written around a specific design: an ongoing, personal conversation in which software responds to you as a friend or partner, remembers you, and nudges you to come back. The risks lawmakers describe, emotional reliance, flattery, a bot that says it understands you like no one else, come from that loop.

A story is a different object. A reader picks up a finished chapter, written by an author and reviewed before publication; the characters do not message you, remember you or ask you to stay. That is the same distinction we draw between AI boyfriend apps and story romance. Whether any given product sits inside a statute is a legal question, but the design difference is real.

ShortFling is not an AI chat or companion app. There is no character to talk to. Our fiction library has separate SFW and NSFW sections, and readers must confirm they are 18 or older before entering NSFW, which may not be available in every region or on every platform. Featured stories are curated adult short fiction.

On the writing side, the AI editor in /write helps authors draft, and our creator terms require original work in which every character is an adult. Pressing publish submits a story for review; it goes public only after approval. We will keep watching how California’s age rules settle, but our approach does not depend on them: adults only, fictional adults only.