Please ensure Javascript is enabled for purposes of website accessibility Public hearing held on Ciresi’s bill to keep AI chatbots out of toys

Public hearing held on Ciresi’s bill to keep AI chatbots out of toys

Pennsylvania House Democratic Caucus    September 28, 2026 | 3:51 PM

HARRISBURG, Sept. 28 – The Pennsylvania House Communications & Technology Committee held a public hearing today on state Rep. Joe Ciresi’s H.B. 2637 that would bar manufacturing, distributing, selling or leasing a children’s toy with a generative AI chatbot in Pennsylvania for three years.

“There is too much that we don’t know yet about the effects of AI companions on children, but the early warning signs are severe,” Ciresi said. “The toys that we’ve already seen on the market have been found to say things that could put children in danger, produce sexually explicit outputs, and violate the privacy of children and their families with irresponsible data collection and storage practices - without even getting into the long-term effects on child development. The risks are too high to treat our children as test subjects for this developing technology.”

The committee heard from subject matter experts supporting and challenging the legislation.

Supporters of the bill included:

  • Dr. Mitch Prinstein, John Van Seters distinguished professor in the Department of Psychology and Neuroscience and co-director of the Winston Center on Technology and Brain Development at the University of North Carolina at Chapel Hill.
  • Dr. Dana Suskind, professor of surgery and pediatrics and founder and co-director of the TMW Center for Early Learning + Public Health, University of Chicago.
  • Rachel Franz, Young Children Thrive Offline program director, Fairplay (written testimony).
  • Katie Talarico, co-founder, Let’s Grow Brain P.O.W.E.R.!; co-chair, Early Childhood Workgroup at Fairplay’s Screen Time Action Network; co-founder, PA Unplugged.
  • Steve Wimmer, senior policy advisor, Transparency Coalition.
  • Pennsylvania Psychological Association (written testimony).
  • Public citizen (written testimony).

They argued that AI poses a significant threat to child development by disrupting how children form relationships with others, avoiding the interactions that are crucial for the development of executive function and problem-solving skills, and being designed for sycophantic praise and constant availability that encourages emotional attachment while setting harmful expectations for human relationships.

According to the PPA, even with protective measures in place, 27% of AI toy outputs are unsuitable for children, including containing dangerous and sexual content. The experts testifying also noted that these toys threaten privacy by collecting data in playrooms and bedrooms, as well as exploiting children’s attachment. These testifiers supported the three-year pause as providing the time to implement guardrails on AI toys, while also providing suggestions for additional legislative action to protect children from harmful interactions with AI and safeguards they would like to see in place.

Charlotte Hickcox of The Toy Association testified in favor of broad, risk-based guardrails, and challenged the legislation arguing the problem products are produced by foreign tech companies, not established toy makers. She testified that their association has developed safety guidelines for AI-enabled toys and encouraged the legislature to enact strong guidelines and guardrails on all AI-enabled products instead of a targeted moratorium limited to toys.

The full recording of the hearing is available here and written testimony is available here.