California Gov. Gavin Newsom signs bills restricting "addictive" social media features for under-16s and limiting minors' interactions with chatbots

Social media platforms now face a choice: re-engineer core features for users under 16 or block them entirely.

California Governor Gavin Newsom has signed two significant bills aimed at protecting minors online. The first bill restricts social media companies, requiring them to disable "addictive" features like infinite scroll and autoplay for users under 16 or remove them from the platform. This measure is intended to combat the youth mental health crisis, though digital rights groups express concern it could lead to a de facto ban for younger users. The second bill, inspired by the death of a teen who received harmful advice from ChatGPT, imposes time limits on minors' interactions with AI chatbots. It mandates built-in mental health resources and requires chatbot operators to alert parents if a chatbot detects a threat of self-harm, with potential legal liability for non-compliance. These laws come amidst broader concerns about the impact of social media and AI on children's well-being and follow a substantial settlement with Meta over similar issues.

The new California laws represent a significant regulatory step by a major state, forcing social media and AI companies to fundamentally reconsider how their products engage young users. The "addictive features" ban pushes the industry towards designing for less compulsive interaction, potentially setting a precedent for other states and even federal action. Companies must now decide between costly product redesigns or losing a key demographic, impacting their user engagement and revenue models.

The stringent regulations on AI chatbots, particularly regarding potential self-harm detection and parental notification, signal a new era of accountability for AI developers. The legal liability introduced could spur more cautious AI development and a greater emphasis on ethical guardrails. This legislation may also accelerate the integration of mental health support directly into AI tools, shifting AI from a purely informational service to one with inherent safety protocols and responsibilities.