California enacts strict AI chatbot and social media rules for kids

California just turned concerns about AI chatbots and youth mental health into detailed statutory obligations. For the AI race, this is a reminder that product design decisions for generative systems are no longer just UX choices, they are regulated features in the biggest U.S. tech jurisdiction. Bans on addictive mechanics like infinite scroll and personalised feeds for teens, alongside age gated chatbot rules, will force major platforms and model providers to build parallel experiences for minors, with constrained recommendation and conversation patterns.

Although these laws do not directly target frontier model training, they shape where commercial value can be safely realized. If it becomes costly or risky to expose young users to open ended chatbots, companies will steer them toward curated, domain specific assistants with narrow, auditable behaviors. That in turn could encourage architectures that separate a powerful core model from tightly controlled child facing shells. Over time, similar design obligations are likely to migrate into other high sensitivity domains such as education and therapy, creating a patchwork of safety centric product standards long before there is consensus on AGI level regulation.