As artificial intelligence (AI) becomes increasingly present in every aspect of our lives, this technology is also finding its way into toys meant for children. These toys mark a significant technological advancement from what we grew up with, using an Internet connection and a microphone and speaker that allows teddy bears, plushies, robots, and other toys to engage in real-time, dynamic conversations.
 
But just as we’ve seen with AI chatbots more broadly, these responses can sometimes be deeply problematic. Reports by the U.S. Public Interest Research Group Education Fund found that these toys, marketed for children ages 3-12, discussed sexually explicit topics and gave detailed answers on how to find and use potentially dangerous household objects such as matches and knives. An investigation by NBC News produced the same explicit and concerning responses.
 
The developers of most major AI platforms openly recognize that their tools “could produce output that is not appropriate for all ages,” as stated on the help pages for both ChatGPT and Grok, and most, if not all, major AI platforms have minimum age and/or parental consent requirements. Despite this, these same AI models are making their way into toys marketed to children as young as 3.
 
Beyond problematic responses, there are important questions about the effects of AI chatbot use on children’s social development, especially for toys that produce responses specifically designed to build an emotional relationship or prolong interactions with their minor users. These devices also raise important questions about data privacy given their potential to collect significant amounts of information about the minors using them, including their voices, sensitive personal information, and more - all through a device with access to the Internet that feeds it through an AI service.
 
Children deserve to play with toys that we know are safe, and we have a responsibility to make sure they’re not the subjects of real-time testing of a technology that is already causing serious problems. That is why I plan to introduce legislation that would put a moratorium on the sale of children’s toys containing AI chatbots, creating a three-year pause to allow for more research on its effects and give time for industry and government guardrails to catch up.
 
Please join me in co-sponsoring this legislation to make sure children are safe and aren’t exposed to an untested, new technology before appropriate guardrails are put in place. Thank you for your consideration.