AI WORLD HQ
If you have a teen at home, chances are they've already played around with ChatGPT, Gemini, or Claude — for homework, for fun, or just to see what the bot will say. Most of those conversations are harmless. But occasionally something goes wrong: a chatbot gives bad advice, plays a character it shouldn't, or fails to notice when a kid is in real distress.
That's the kind of moment a new California bill is trying to address. OpenAI, the maker of ChatGPT, has publicly come out in support of it. Here's what we actually know, and why it matters to families everywhere — not just in California.
What's being proposed
The bill — known as California SB 1119 — focuses on how AI companies handle conversations with minors (anyone under 18). In plain language, the goal is to make sure that when a teen uses a chatbot, the bot applies stronger safeguards than it would for an adult.
What those safeguards look like in practice is still being worked out. They typically include things like reasonable steps to identify younger users, special handling for conversations involving self-harm or suicide, and tighter limits on certain role-play or companionship features for minors. The exact wording matters — and that's what California lawmakers are still negotiating.
Why OpenAI is publicly backing it
It's unusual for a company to publicly support a law that adds new rules to its own products. OpenAI has framed the bill as making sense: clearer rules mean clearer expectations, both for parents and for the AI industry.
At the same time, OpenAI has been clear it doesn't want the safeguards to be so strict that teens lose access to AI for learning, creativity, or homework help. That balance — protecting kids without locking them out — is the part of the debate that's still very much alive.
Wrap-up
OpenAI's support for SB 1119 is one piece of a much bigger conversation about how AI should handle young users. Nothing is final yet — lawmakers are still working out the exact language.
A simple next step you can take today: ask the teens in your life which AI tools they're actually using, and for what. That five-minute conversation is more useful than any single law.