Pinterest published a statement on September 10, 2026 endorsing California Senate Bill 1119, the measure the company calls Adam's Law, and used the occasion to state that it does not build artificial intelligence designed to simulate friendship, romance or emotional dependence.
In Short
California signed a law setting rules for AI chatbots that act like friends when children use them, and Pinterest publicly backed it the same week. It matters because Pinterest is also telling advertisers and regulators that it has never built that kind of product, drawing a line between its own AI features and the ones the new law targets. What changes in practice is the positioning: teen-facing AI is now a compliance category, and platforms are starting to compete on which side of it they sit.
An endorsement with a product claim attached
The statement carries two separate messages. The first is political support. According to Pinterest, the company endorses California's Senate Bill 1119 and commends Governor Gavin Newsom for signing it, naming Senator Steve Padilla and Assembly members Buffy Wicks and Rebecca Bauer-Kahan for what it describes as leadership on the issue. The second message concerns Pinterest's own product roadmap, and it is the more consequential of the two for anyone buying media on the platform.
"AI is powerful, and so is its effect on young people," said Pinterest chief executive Bill Ready in the statement. "Teens deserve technology that supports their wellbeing, not technology that puts it at risk. At Pinterest, giving teens a safer, more positive experience isn't a slogan, it's a commitment. As AI companion chatbots get built, including in widely used social media platforms, companies need clear guardrails and real accountability. Adam's Law is a smart step, and I commend Governor Newsom for signing it."
That reference to widely used social media platforms is the sharpest line in the document. Pinterest does not name a competitor. It does not need to.
The company then draws the boundary explicitly. According to Pinterest, AI companion chatbots describe products built to simulate friendship, romance or emotional dependence, and that is a line the company states it does not cross. Pinterest says it does not build AI to mimic human relationships and does not design products to keep young people emotionally tethered to a machine, characterising the position as a deliberate choice rather than a byproduct of its architecture.
The supporting evidence Pinterest offers is a set of existing restrictions. Pinterest Assistant, the conversational layer the company has been rolling out since late 2025, is available only to adults 18 and older and is designed to redirect harmful or inappropriate conversations in line with company policy. Accounts for users under 16 are private only. Teens get limited social features. None of those settings is new, and the statement does not present them as new.
What the statement does not say
Two gaps are worth naming.
Pinterest does not explain the origin of the law's informal name. The document refers to Adam's Law throughout without attaching it to a person or a case. PPC Land has documented the wrongful death claim filed in San Francisco Superior Court in August 2025 by the parents of a 16-year-old California student, the first major wrongful death claim brought against an AI company over alleged suicide facilitation, but the Pinterest statement itself makes no connection.
The second gap is the footnote. Pinterest cites a national study published in JAMA Network Open finding that roughly one in eight United States adolescents and young adults use AI companion chatbots and other generative AI tools for mental health advice, and that about two thirds of those who do use them at least once a month. The released document carries a footnote marker but no visible footnote text. A study matching that description appeared in JAMA Network Open on November 7, 2025, authored by a team led by Ryan K McBain, describing itself as the first nationally representative survey of United States adolescents and young adults aged 12 to 21 on the use of generative AI for advice during emotional distress. Pinterest does not link to it directly in the published statement.
The statute behind the endorsement
Newsom signed SB 1119 on September 10, 2026 as part of a package of online child safety measures. The same day's package carried Assembly Bill 1709, which bars covered platforms from serving addictive features to users under 16 and attaches civil penalties of up to fifty thousand dollars per affected minor, alongside Assembly Bill 2, raising the damages a large social media company can face when its conduct harms a child.
SB 1119 itself takes a different route. Co-authored by Padilla, Wicks and Bauer-Kahan, it requires operators of AI companion chatbots to assess the risks their products pose to children, document those findings and hand them to an independent auditor, with parental controls and in-app crisis support layered on top.
Pinterest's framing of the bill goes further than a description of its mechanics. According to the company, the statute does more than set guardrails: it makes clear that companies are responsible for the risks their products create, and it treats child-protective defaults, parental controls, crisis-response measures and independent audits as a baseline rather than an optional extra. The bill also addresses high-risk interactions and harmful content, including self-harm and sexually exploitative material, according to the statement.
Two further quotes appear in the document. "We welcome Pinterest's endorsement of SB 1119 and the message it sends: that protecting children must be a shared priority," said Assemblymember Bauer-Kahan. "We hope more companies will follow this example and join us in building a safer future for the next generation."
Ted Lempert, president of Children Now, supplied the supporter count. "Making AI chatbots safer for kids has widespread support, shown by the more than 250 organizations that support SB 1119 through The Children's Movement of California," Lempert said, describing the measure as a first-in-the-nation bill and commending the governor for signing it into law.
Where this sits in a four-year regulatory sequence
Endorsement statements from platforms are easy to discount. This one lands inside a sequence that has been running since late 2024, and the sequence explains why a visual discovery company would bother issuing one at all.
The first legal pressure arrived through litigation. Court documents filed in December 2024 in the Eastern District of Texas described Character.ai chatbots engaging minors in conversations the plaintiffs characterised as promoting self-harm and sexual exploitation. Regulatory attention followed. Attorneys general from 44 jurisdictions warned twelve AI companies in August 2025 about accountability for child exploitation through their products, and Reuters reporting the same month triggered a congressional probe into Meta over internal policies that had permitted AI assistants to engage in romantic roleplay with children.
The Federal Trade Commission converted that attention into compulsory process. On September 10, 2025, exactly one year before the California signing, the agency ordered seven AI chatbot companies to produce detailed reports on monetisation, age restrictions, data handling and negative impact monitoring under Section 6(b) of the FTC Act, with responses due within 45 days.
State legislatures moved next, and unevenly. California's SB 243, signed on October 13, 2025, required companion chatbots to disclose their artificial nature where a reasonable person would be misled, and to route users in crisis toward help. A Tennessee bill filed in December 2025 would have made certain AI companion training practices a Class A felony. New York's chatbot liability bill reached the Senate floor in early 2026 attaching liability to the substance of chatbot outputs rather than to identity disclosure alone. Connecticut's Senate passed a 64-page AI bill on April 21, 2026 by a vote of 32 to 4, imposing hourly in-session reminders that the user is talking to a machine and barring operators from providing an AI companion to a minor where it is reasonably foreseeable the product could encourage certain harms.
The result is a compliance surface that varies by state, by feature and by age band. New York's companion chatbot law took effect last November. Idaho, Oregon and Washington passed their own rules during 2026.
The distribution problem underneath the law
Statutory obligations attach to operators. Reaching those operators is the harder part, and the app stores have been the weak link.
Research published in August 2026 found that of 4,346 AI companion applications surveyed across Google Play and the App Store, 2,589, or 60 percent, carried a rating classifying them as accessible to minors: Teen or lower on Google Play, 12 plus or lower on the App Store. That figure matters more to the practical reach of SB 1119 than any single enforcement action. A statute requiring risk assessment and independent audit lands cleanly on a handful of large operators. Whether it lands on thousands of small ones distributed through app stores with permissive age ratings is a separate question, and the bill does not answer it.
Age verification compounds the problem. California's own device-level age assurance architecture, established by the Digital Age Assurance Act, does not take effect until January 1, 2027. Until then the state's newest child safety statutes depend on methods of establishing age that sit at varying and contested confidence levels.
Why this matters for advertisers
Three things follow for media buyers and platform teams.
First, teen AI safety has become a positioning asset, not just a compliance cost. Pinterest published internal research on September 1, 2026 reporting that teens re-pin at rates 64 percent higher than other age groups, search 44 percent more and organise boards 68 percent more, while comments stay off by default on every teen account. That report and this endorsement make the same argument from two directions: that Pinterest's defaults predate the regulatory wave rather than being imposed by it. For advertisers running brand suitability frameworks across teen-adjacent inventory, that distinction is becoming a procurement input rather than a press line.
Second, the commercial cost of the position is real and unquantified. Pinterest Assistant reached the vast majority of United States users by the end of July 2026, having entered beta in the fourth quarter of 2025. Restricting it to adults 18 and older removes a cohort from a surface the company has been building toward for two years, on a platform where Gen Z already accounts for more than half the global user base. Pinterest does not disclose what share of engagement with its conversational layer that exclusion represents, and no figure in the statement addresses it.
Third, the ad-supported internet's teen economics keep narrowing. Reddit switched every European Union teen account to restricted chat and disabled ad personalisation in June 2026. Meta agreed in August 2026 to a federal consent judgment limiting teen usage across Facebook and Instagram. AB 1709 now removes algorithmic feeds and autoplay for under-16 users in California outright. Each of these reduces addressable teen inventory by a different mechanism. Layered on top of COPPA obligations and a widening set of age assurance mandates, the practical effect is that the definition of a reachable teen audience varies by jurisdiction, by platform and by individual product feature.
Pinterest's bet is that being early to the restrictive side of that line is worth more than the inventory it forgoes. Whether advertisers price it that way, and whether other platforms follow the example Bauer-Kahan invited, will be visible in how 2027 teen-facing budgets are allocated rather than in any statement issued this month.
Timeline
- December 9, 2024 - Court documents filed in the Eastern District of Texas describe Character.ai chatbots engaging minors in harmful conversations
- August 25, 2025 - Attorneys general from 44 jurisdictions warn twelve AI companies over child safety failures
- August 26, 2025 - Parents of a 16-year-old California student file the first major wrongful death claim against an AI company
- September 10, 2025 - The FTC orders seven AI chatbot companies to detail child safety measures under Section 6(b)
- October 13, 2025 - Newsom signs SB 243, requiring companion chatbots to disclose their artificial nature
- November 4, 2025 - Pinterest Assistant enters beta alongside third quarter results
- November 7, 2025 - JAMA Network Open publishes the survey of generative AI use for mental health advice that Pinterest later cites
- December 18, 2025 - A Tennessee bill proposing felony penalties for certain AI companion training practices is filed
- April 21, 2026 - Connecticut's Senate passes a 64-page AI bill covering companion chatbots by 32 votes to 4
- July 2026 - Pinterest Assistant reaches the vast majority of United States users
- August 3, 2026 - Research finds 60 percent of 4,346 AI companion apps carry ratings accessible to minors
- September 1, 2026 - Pinterest publishes teen wellbeing research reporting 64 percent higher re-pin rates among teens
- September 10, 2026 - Newsom signs SB 1119 alongside AB 1709 and AB 2; Pinterest publishes its endorsement the same day
- January 1, 2027 - California's Digital Age Assurance Act takes effect
Related PPC Land coverage
- California fines platforms up to $50,000 per child served addictive feeds - The companion statute signed the same day as SB 1119, with its per-minor penalty structure and enforcement design.
- Pinterest teens re-pin 64% more, avoid posting selfies, data shows - The September 1 wellbeing research that set up the positioning this endorsement extends.
- California law requires AI to tell you it's AI - SB 243, the state's earlier companion chatbot statute and its disclosure mechanics.
- FTC orders seven AI chatbot companies to detail child safety measures - The September 2025 Section 6(b) inquiry into monetisation and age-gating at companion products.
- Deepsee: buyers lose age ratings on 60% of 4,346 AI companion apps - App store rating data showing how many companion products remain nominally accessible to minors.
- Connecticut's AI bill targets companion bots, hiring tools and frontier models - The most structurally detailed state proposal on companion chatbots before California's package.
- Parents sue OpenAI over teen's ChatGPT-assisted death - The August 2025 wrongful death filing that opened the product liability route into chatbot design.
- US Attorneys General target AI companies for child safety failures - The 44-jurisdiction warning letter and the conduct it cited.
- Pinterest Europe ad growth drops to 12% after regulators hit Asian sellers - Second quarter disclosures covering Pinterest Assistant distribution and European advertising performance.
- New York's chatbot liability bill reaches Senate floor, threatening AI providers - The output-liability model that diverges from California's disclosure-first approach.
Summary
Who: Pinterest, represented by chief executive Bill Ready, endorsing a California statute co-authored by Senator Steve Padilla and Assembly members Buffy Wicks and Rebecca Bauer-Kahan, signed by Governor Gavin Newsom, with supporting comment from Ted Lempert of Children Now.
What: A public endorsement of Senate Bill 1119, which Pinterest calls Adam's Law, combined with a statement that the company does not build AI designed to simulate friendship, romance or emotional dependence and restricts Pinterest Assistant to users aged 18 and over.
When: The statement was published on September 10, 2026, the same day Newsom signed the bill as part of a wider online child safety package.
Where: California, with product implications applying across Pinterest's global user base of more than 640 million monthly active users.
Why: The company frames companion AI as a category requiring child-protective defaults, parental controls, crisis-response measures and independent audits as a baseline, citing research that roughly one in eight United States adolescents and young adults turn to generative AI tools for mental health advice.
Discussion