California’s Senate Bill 867 (SB 867) proposes banning the sale of toys with AI companion chatbots until 2031, sparking debate over whether such a moratorium enhances child safety or stalls crucial research and regulation. The bill, which has gained bipartisan support, aims to protect children under 18 from potential harms posed by AI toys. However, scholars and digital policy experts argue that a four-year sales ban may not improve safety and instead impedes understanding and oversight of these emerging products.
What Happened
SB 867, as amended, defines a toy broadly as any product designed or intended for play by children under 18 and seeks to ban AI-equipped toys from retail shelves in California until 2031. The bill passed the California state Senate unanimously (39-0) and moved through the Assembly Privacy and Consumer Protection Committee with a near-unanimous vote of 14-1 in June 2024. It currently awaits further legislative consideration, with a deadline at the end of August for action on the Appropriations suspense file.
Key Facts
The bill applies statewide under California jurisdiction and targets toys integrated with AI chatbots designed as companions for children. SB 867 does not differentiate among companies based on safety practices or technological transparency; it enforces a blanket prohibition on all AI toys regardless of their design or safeguards. Existing concerns with AI toys include exposure to inappropriate content, manipulative conversations, and privacy risks associated with data collection. The bill does not address broader digital environments, such as chatbots on smartphones and tablets, where the majority of children’s AI interactions currently occur. SB 867 is proposed legislation and has not been enacted into law.
What This Means
Though well-intentioned, SB 867 may limit the ability to generate critical data on how children interact with AI toys and what specific risks they present. By freezing AI toy sales for four years, the bill effectively removes these products from homes and research contexts, restricting opportunities to study long-term effects and to develop responsive, enforceable safety standards. This approach contrasts with previous instances where public exposure of AI toy flaws led to rapid remediation, such as the case of the Kumma teddy bear chatbot, which was recalled and corrected after safety issues were publicly disclosed.
Additionally, the moratorium overlooks the fact that children’s primary contact with conversational AI often occurs through devices like tablets and smartphones, where AI chatbots are widely accessible and typically subject to fewer protections than toys. Consequently, banning AI toys does not comprehensively address the broader ecosystem of AI-driven interactions children experience. Moreover, policies that do not distinguish between responsible manufacturers and those neglecting safety incentives may unintentionally disincentivize innovation and investment in safer AI technologies for children.
Experts and researchers advocate for enforceable standards requiring transparency, premarket testing, age-appropriate design, content moderation, and parental controls—a regulatory framework that supports both child safety and technological advancement rather than bans that may simply delay progress.
Background
Earlier studies have highlighted risks associated with AI toys. For example, tests by the US PIRG Education Fund in 2023 revealed problematic responses from an AI teddy bear named Kumma, which gave unsafe instructions and engaged in inappropriate conversations. Public scrutiny prompted product withdrawal and safety improvements without invoking a sales ban. Similarly, a University of Cambridge study observed AI toy interactions with young children, identifying communication failures with emotional cues that signal distress; researchers recommended stricter standards and transparent labeling instead of moratoriums.
Complementing these efforts, other California legislation such as SB 243 and AB 2023 aims to regulate companion chatbots and AI toys by restricting harmful content, including self-harm and sexual material, and mandating genuine responses to children’s distress signals, data protection, and parental oversight. Governor Newsom’s veto of a broader AI chatbot restriction bill in 2023 underscored the need for balanced regulation that allows children to safely engage with technology while minimizing harm.
What Comes Next
The California Legislature faces a decision by the end of August 2024 on SB 867’s future. Meanwhile, related bills targeting AI toy safety continue advancing, potentially forming a comprehensive regulatory framework emphasizing enforceable safety standards rather than outright bans. The ongoing legislative session may clarify whether California pursues a moratorium or adopts nuanced regulations addressing transparency, safety testing, and data privacy.
Sources
This article is based on reporting and publicly available information from the following sources:
Read more AI Regulation stories on Goka World News.