California Building aims to protect children from AI chatbots

SACROMENTO, CA — California lawmakers are pushing for new laws aimed at protecting children from potentially dangerous artificial intelligence chatbots.

Senate Bill 243 requires a chatbot platform to install safety features, and explicitly discloses when users are interacting with AI rather than real people.

The bill was partly inspired by the death of 14-year-old Sewell Setzer, a Florida boy who committed suicide several months after an emotionally intense, sexually explicit conversation, according to a lawsuit filed by his mother.

“What I saw was the dim light of his light,” said Setzer’s mother, Megan Garcia, who traveled from Orlando to Sacramento in favor of the law. “He was a beautiful, bright, athletic young man, and became someone who wanted to isolate himself in his bedroom and do what he wanted to do.”

Garcia is currently suing Charition.ai, a chatbot company that denies liability for his son’s death. But her defense energised California lawmakers to act on what they say is a growth in threat to vulnerable users.

“Technological innovation is extremely important, but our kids cannot use it as guinea pigs to test the safety of new products in real time,” said Sen. Padilla (D-San Diego). “The interest is too high to keep vulnerable users accessing this technology without having the right guardrails in place.”

One of the provisions of SB 243: protocols for detecting and responding to suicidal thoughts and essential labeling, and clarifying that users are involved in AI rather than humans.

“Teens experiencing anxiety, depression or loneliness have formed attachment to AI bots that cannot truly respond to their needs,” said Sen. D-La Mesa Weber. “What’s worse, sometimes (these bots) reinforce harmful thoughts and patterns.”

For Garcia, the bill is to save lives. “It could prevent other families from experiencing my experience,” she said.

“Turn point”: lawmakers push newspapers to stabilize fuel supply