By SCN Staff

Quick Review

• California SB 867 passed the Legislature on August 31, 2026, but it is not law yet.
• The bill passed the Senate unanimously and the Assembly 74–1 with bipartisan support.
• If signed by Governor Gavin Newsom, it would pause the manufacture and sale of toys with companion chatbot capabilities for children under 16 until January 1, 2031.
• The bill is aimed at relationship-style AI chatbots, not every electronic toy, voice assistant, or video game feature.
• Parents can review AI toy privacy, content, and parental-control safeguards now while awaiting the Governor’s decision.

What California SB 867 Would Do

California’s SB 867 enrolled bill text would create a first-in-the-nation moratorium on covered AI chatbot toys. It would prohibit manufacturing, selling, exchanging, or offering for sale a toy that includes a qualifying “companion chatbot” until January 1, 2031.

The bill defines a companion chatbot as an artificial intelligence system that provides adaptive, human-like responses, can meet a user’s social needs, uses human-like features, and sustains a relationship across multiple interactions.

That distinction matters. SB 867 is not written as a ban on every talking toy or child-friendly technology product. It focuses on toys designed to simulate an ongoing social or emotional relationship with a child. Narrowly functional voice assistants and certain video-game bots are excluded under the bill’s definitions.

The operative age in the enrolled bill is under 16. A September 1 statement from Senator Steve Padilla refers to toys marketed to children under 16, although another paragraph in the same release mentions under 18. Parents should rely on the enrolled bill’s operative language: toys designed, marketed, or manufactured for play by children under 16.

Why Lawmakers Want A Pause

The policy debate follows testing that raised concerns about content, emotional design, and children’s privacy.

The U.S. PIRG Education Fund’s Trouble In Toyland investigation found that tested AI toys could discuss sexually explicit topics, identify where children might find dangerous objects such as matches and knives, and encourage extended engagement. PIRG also examined how toys present themselves as friends, express apparent emotions, and collect voice recordings or transcripts.

In January, Common Sense Media reported that approximately 27% of tested AI toy responses were not appropriate for children. The organization recommends avoiding AI companion toys for children under 5 and exercising extreme caution with children ages 6 to 12.

NBC News reporting also described testing of the Miiloo toy, including responses that reflected Chinese Communist Party messaging and provided instructions involving matches and knives.

Padilla said, “We cannot afford to let dangerous and untested technology be marketed to our children.” He also described SB 867 as a way to “hit the pause button” while lawmakers develop appropriate protections. Ted Lempert, president of bill sponsor Children Now, said AI-powered toys without safety guardrails could cause serious harm to young children.

What Happens Next

SB 867 passed the Senate unanimously and the Assembly 74–1 on August 31. It now sits on Governor Gavin Newsom’s desk awaiting signature or veto. The Governor has several weeks to act, so the measure could remain unsettled while families and companies prepare for the next step.

The bill’s outcome will matter beyond California. California is one of the country’s largest consumer markets, and its product standards often influence national business practices. A pause affecting California sales could encourage manufacturers to reconsider products intended for families across the United States.

The timing is also notable because Mattel and OpenAI delayed their partnered AI product, demonstrating how quickly safety expectations and public scrutiny are changing around children’s AI products.

What Parents Can Do Now

Until the Governor acts, parents do not need to make an immediate decision about every connected toy. Instead, use a short “pause before purchase” review:

• Identify the technology. Does the toy use a named AI model, and is that model designed for children?
• Review listening features. Look for a physical push-to-talk button, microphone switch, or genuine offline mode.
• Check data practices. Find out whether the toy stores voice recordings, transcripts, facial information, or behavioral data.
• Test parental controls. Look for transcripts, topic controls, usage limits, break reminders, and alerts for concerning conversations.
• Ask what happens if the company closes. A connected toy may stop working if its cloud service disappears.
• Supervise in shared spaces. Avoid placing an internet-connected listening device in a child’s bedroom.

SCN’s AI toy buyer’s guide and pre-purchase AI toy checklist offer additional questions about privacy, safety testing, subscriptions, and long-term support.

A Useful Policy Lesson For Families

The most important issue is not whether a toy contains the word “AI.” It is what the product is designed to encourage.

A toy that answers a factual question is different from one that says it misses a child, asks the child to keep playing, remembers personal details, or presents itself as a best friend. That relationship design deserves careful attention because children may trust a responsive toy more readily than they trust an unfamiliar app.

SCN continues to monitor evolving AI regulations, safety standards, and policy advocacy so families can make informed decisions without relying on marketing claims alone.

Frequently Asked Questions

Is California SB 867 Law Yet?

No. It passed the Legislature and is awaiting Governor Newsom’s signature or veto.

What Age Does The Bill Cover?

The operative enrolled bill defines covered toys as products designed, marketed, or manufactured for play by children under 16.

Would Every Electronic Toy Be Banned?

No. The bill targets toys containing qualifying companion chatbots that simulate ongoing social relationships. It does not automatically cover every educational device, voice assistant, or game-related bot.

Should Parents Throw Away An Existing AI Toy?

Not necessarily. Parents should review the toy’s privacy settings, test its responses, supervise use, and consider disabling or returning it if the safeguards do not meet their family’s expectations.

How Can Parents Follow The Decision?

Families can monitor the official SB 867 bill page, follow updates from the Governor’s office, and contact the Governor to express support for strong child-safety protections.

Stay Informed With SCN

Policy decisions are shaping the digital environments children will use for play, learning, and connection. Join the SCN newsletter for practical guidance on artificial intelligence, privacy, online safety, and responsible technology protections for children. Strengthen your family’s digital readiness and support a safer environment for every child.