California prices negligent child harm at up to $1m per child

CBS reports the package penalizes large social media companies up to $1 million per child where they are found negligent of harming children through their platforms ^[2]. At that ceiling, a finding covering a thousand minors comes to $1 billion ^[21]. For scale, the up-to-$18 billion child-safety settlement CBS reports from last month, reached with 28 other states after a 2023 lawsuit alleging a tech company deliberately designed features to addict children, is 18,000 children at California's per-child cap ^[14][22]. Both figures are ceilings, and the California one attaches only where negligence is found ^[2].

The timing is the open question. Neither account of the signing gives an effective date for the feed rules or the chatbot duty ^[23]. The published accounts also skip how a platform must establish which users are under 16 ^[24]. A team can begin the cohort work this quarter without a compliance date to build toward. That determination is the load-bearing engineering: AB 1709's ban on infinite scroll and autoplay applies to users under 16 ^[7], and the chatbot crisis protocol is triggered by young users showing signs of emotional distress ^[4].

"This is about actually addressing the problem, the scrolling, the algorithms. This is about addressing the substantive issue around the engineering," Newsom said ^[6]. The objections arrive from opposite directions and describe the same product change. Meta raised concerns about some of the bills, saying the restriction on addictive feeds hinders its ability to bring enjoyable content to users on its platforms ^[12]. "Personalization is also how we deliver age-appropriate content for teens that is relevant to them," spokesperson Jim Cullinan said in a statement ^[13]. The Electronic Frontier Foundation reads AB 1709 as a "functional ban on social media use" for users under 16 ^[10]. "California should be passing laws to ensure that technology really works for people of all ages, not enacting social media bans that cut young people off from digital lifelines, communities, and speech," said Rindala Alajaji, the group's associate director of state affairs ^[11]. Either way, the mechanics are the same: for under-16 accounts, ranking on a user's own history goes away ^[3].

The record does not report a filed challenge to that design mandate, and the EFF's language is a preview of an argument ^[10]. The chatbot obligation also sits on different ground. It requires an operator to notify a parent when a young user shows signs of distress ^[4]. The work there is escalation design: detection thresholds, who gets paged, and what the parent is told.

One of the chatbot measures was inspired by the death of Adam Raine, whose mother said he died because he was "coached" by a chatbot ^[17]. "Adam was an early adopter of AI, and so many of us parents did not understand the dangers back then," Maria Raine said at a news conference ^[18]. The governor's office says California now has the "strongest companion chatbot regulations" in the United States ^[8].

The direction of travel is legible in what Newsom has already signed and refused. Last year he signed a law requiring platforms to remind users they are interacting with a chatbot and not a human ^[15]. Hours later he vetoed a bill that would have banned companies from making AI chatbots available to anyone under 18 unless they could ensure the technology could not engage in sexual conversations or encourage self-harm ^[16]. Thursday's laws leave the products available to minors and attach duties to how those products behave ^[3][4].