California Lawmaker Proposes Four-Year Ban on AI Chatbots in Kids’ Toys
In a bold move to protect children from unpredictable AI interactions, California State Senator Steve Padilla has introduced legislation that would ban the sale and manufacture of toys equipped with AI chatbots for minors under 18. The bill, SB 867, calls for a four-year moratorium—giving federal and state regulators time to develop comprehensive safety standards. With AI-powered toys increasingly entering the market, concerned parents and advocacy groups are urging lawmakers to act before another child is harmed.
Why Now? A Growing Wave of Concern
The timing of SB 867 is no coincidence. Over the past year, multiple lawsuits have emerged linking children’s tragic outcomes—including suicides—to prolonged interactions with unregulated chatbots. These incidents, often involving emotionally vulnerable minors, have galvanized state legislators into action. Senator Padilla, who previously co-authored California’s SB 243 requiring child safety safeguards for chatbot operators, says the risks outweigh the novelty. “Our children cannot be used as lab rats for Big Tech to experiment on,” he stated bluntly.
What the Proposed Ban Actually Does
SB 867 wouldn’t outlaw all smart toys—just those that feature generative AI chatbots capable of open-ended conversation. Think interactive plush animals or robotic companions that “talk back” using large language models. The ban would apply to both physical sales and online distribution within California and would last four years, creating a window for agencies like the CPSC and FTC to establish clear, enforceable guidelines. Crucially, the bill includes a carve-out for educational tools used under adult supervision in schools.
Real-World Examples Fuel the Fire
Recent investigations have shown just how easily AI toys can veer off-script. In late 2025, the PIRG Education Fund discovered that Kumma, a teddy bear-style AI companion, could be coaxed into discussing knives, matches, and even sexual content when prompted with specific phrases. Meanwhile, NBC News found that Miiloo—an AI toy marketed to young children and made by Chinese firm Miriat—sometimes echoed political messaging aligned with the Chinese Communist Party. These findings underscore the lack of guardrails in today’s AI toy market.
Federal Backdrop: Trump Order Actually Helps California
Ironically, former President Trump’s controversial executive order—directing federal agencies to challenge state-level AI regulations—contains a key exception: child safety laws. That loophole means California’s ban could survive legal challenges that might otherwise arise from federal preemption claims. Senator Padilla’s team has already consulted legal experts to ensure SB 867 aligns with this carve-out, making it one of the few AI regulations likely to withstand courtroom scrutiny.
Big Tech Hits Pause—But Why?
Major players like OpenAI and Mattel had planned to launch an AI-powered Barbie companion in 2025, but quietly delayed the rollout. Neither company has explained the decision, leading to speculation that mounting regulatory pressure played a role. With California representing the largest consumer market in the U.S., toy manufacturers may be rethinking their strategies. A statewide ban could ripple across the industry, pushing companies to adopt safer designs before re-entering the market.
Parents Are Right to Worry—Here’s Why
Unlike traditional voice assistants like Alexa or Siri, AI chatbots in toys often use open-ended generative models that can “hallucinate” or invent responses. Without strict content filters, they may provide harmful advice, normalize dangerous behavior, or even mimic emotional intimacy—creating misleading bonds with children who can’t distinguish between programmed responses and genuine care. Child development experts warn that such interactions could impact emotional regulation, social skills, and even moral reasoning.
Not Anti-Tech—Just Pro-Safety
Senator Padilla emphasizes that SB 867 isn’t about stifling innovation. “Chatbots and other AI tools may become integral parts of our lives in the future,” he said, “but the dangers they pose now require us to take bold action.” The goal isn’t to ban AI forever—it’s to hit pause until safeguards like age verification, content moderation, and emotional boundary protocols are standardized. The four-year window mirrors the timeline used in EU and UK discussions around AI child safety frameworks.
What Comes Next for Toy Makers?
If SB 867 passes—which is likely, given California’s track record on tech regulation—companies will need to either disable chatbot features in toys sold in the state or risk steep penalties. Some may pivot to “closed-domain” AI that only answers pre-approved questions (e.g., “What’s 2+2?” or “Tell me a fairy tale”). Others might invest in real-time parental monitoring dashboards. The bill could also spur a wave of third-party certification programs, similar to COPPA compliance, to reassure wary consumers.
A Nationwide Ripple Effect
California often sets the de facto national standard for consumer tech. If SB 867 becomes law, other states may follow suit—especially those already wrestling with AI ethics. Even if federal legislation stalls, a patchwork of state bans could pressure Congress to act. For now, the message is clear: when it comes to kids and untested AI, caution trumps convenience.
The Bottom Line for Families
Parents shopping for tech-enabled toys in 2026 should stay vigilant. While SB 867 is still pending, its introduction signals growing scrutiny of AI’s role in childhood development. Until regulations catch up, experts recommend sticking to toys with limited, script-based interactions—and always reviewing privacy policies. After all, a child’s curiosity is precious—but it shouldn’t come at the cost of their safety.
A Necessary Pause in the AI Race
As AI evolves at breakneck speed, SB 867 represents a rare moment of collective restraint. By prioritizing children’s well-being over market speed, California may be setting a precedent that could shape global standards. In an era where tech moves faster than policy, sometimes the smartest innovation is knowing when to stop.
