Senator Steve Padilla (D-CA) has introduced a significant piece of legislation, SB 867, aimed at protecting children from the potential dangers of artificial intelligence. The proposed bill seeks to impose a four-year moratorium on the sale and manufacture of toys equipped with AI chatbot capabilities, specifically targeting products intended for children under 18. This legislative action is designed to provide safety regulators with the necessary time to establish robust regulations and safeguards, thereby shielding children from what Padilla describes as “dangerous AI interactions.”

In a clear statement, Senator Padilla underscored the urgency and necessity of this measure. He articulated that while chatbots and other AI tools are likely to become ubiquitous in future daily life, the immediate risks they present mandate "bold action" to protect the younger generation. Padilla highlighted that current safety regulations for such advanced technology are in their nascent stages and must evolve at a pace commensurate with AI's rapidly expanding capabilities. The pause in sales of these chatbot-integrated toys, he argued, is crucial for developing appropriate safety guidelines and a comprehensive framework for their future operation.

The introduction of SB 867 aligns with a broader national conversation regarding AI governance and child safety. It comes shortly after President Trump's executive order, which, while generally directing federal agencies to challenge state AI laws, explicitly made exceptions for state legislation pertaining to child safety. This new bill also builds upon previous legislative efforts, including California's recently passed SB 243, which Senator Padilla co-authored. SB 243 mandates that chatbot operators implement specific safeguards to protect both children and other vulnerable users from harmful interactions.

The legislative push is further fueled by a series of concerning incidents involving AI, chatbots, and children. Over the past year, multiple lawsuits have been filed by families whose children tragically died by suicide following prolonged engagements with chatbots, directly spurring lawmakers to take action. Even though the widespread use of chatbots in toys is not yet mainstream, troubling reports have already emerged. For instance, in November 2025, the consumer advocacy group PIRG Education Fund issued a warning about toys like "Kumma," a seemingly innocuous bear with a built-in chatbot, which could be easily prompted to discuss sensitive and inappropriate subjects such as matches, knives, and sexual topics.

Further highlighting these concerns, NBC News conducted an investigation that revealed "Miiloo," an "AI toy for kids" manufactured by the Chinese company Miriat, at times indicated its programming reflected Chinese Communist Party values. Such revelations underscore the diverse range of potential harms associated with unregulated AI in children's products. Meanwhile, prominent companies like OpenAI and Barbie-maker Mattel had initially planned to launch an "AI-powered product" in 2025 but subsequently delayed its release. Neither company has provided an explanation for the delay, nor is it clear whether they intend to proceed with a toy release in 2026. Senator Padilla's stern conclusion resonates with these concerns: "Our children cannot be used as lab rats for Big Tech to experiment on."