OpenAI is backing a California bill that would turn some youth-AI safeguards from voluntary company policies into legal requirements.
The company is urging Gov. Gavin Newsom to sign Senate Bill 1119, which would require operators of covered AI companion chatbots to determine users’ ages or apply child protections by default. Providers would also face safety assessments, published child-safety policies and independent audits.
“In the absence of federal action, California has an opportunity to set a strong standard for youth AI safety,” Ann O’Leary, OpenAI’s vice president of global policy, wrote in the company’s announcement.
The bill would require AI operators to determine a user’s age or apply child protections by default. Providers would also have to assess safety risks before releasing new or substantially changed companion chatbots, publish child safety policies and undergo independent audits.
What changes for young users
SB 1119 would require stronger protections against harmful interactions, including encouragement of self-harm or physical violence, simulated romantic interest, claims that a chatbot is human or sentient, and attempts to encourage emotional dependence.
For children, the bill would also establish default restrictions around usage and certain product features. These include limits on continuous and daily use, disabled persistent conversational memory and push notifications by default, and restrictions on custom system prompts.
Parents would gain tools to manage or disable access, including the ability to shut off access for children under 16. The legislation would also require crisis-response procedures and connections to crisis services when serious risks emerge.
More must-read AI coverage
- SS&C Intralinks DealCentre AI vs. Datasite: Which platform is built for the future of dealmaking?
- SS&C Intralinks FundCentre AI vs. Juniper Square: Which platform better supports modern private markets fund managers?
- Why Data, Not Models, Determines AI Success
- The Rise of the AI-Native Factory: How Physical AI Is Transforming Manufacturing
OpenAI wants AI treated differently from social media
OpenAI argues that teenagers use AI tools differently from social platforms and that regulation should preserve beneficial uses such as learning, research and productivity.
The company says nearly nine in 10 teens use ChatGPT for learning, information, skill-building or productivity during a given week. OpenAI argues that regulation should therefore protect those uses rather than simply restricting access.
“SB 1119 also appropriately recognizes that AI is not social media,” OpenAI said.
That distinction matters as California considers separate legislation targeting social media features designed to keep young users engaged, including infinite feeds and autoplay.
The most important change may be that SB 1119 could move some youth-safety practices from company policy into legal requirements.
OpenAI already says ChatGPT for Teens automatically applies stronger protections to users identified as 13 to 17, while its broader safety work includes parental controls, age prediction and under-18 principles.
If SB 1119 becomes law, companies would face formal obligations, audits and potential penalties rather than relying solely on voluntary safeguards. That could raise the cost of building and operating AI products for minors, particularly for smaller companies, but it could also create a clearer baseline for safety across the industry.
The challenge will be keeping those rules useful as AI products change faster than legislation typically does.
What happens next
Most of SB 1119’s requirements would take effect July 1, 2027, while independent child-safety audits would begin later under the bill’s implementation schedule.
OpenAI has acknowledged that the rules will need continued study and adjustment as AI systems evolve. Newsom has until Sept. 30 to sign or veto the legislation.
If he signs it, the significance could extend well beyond California. As with other major technology regulations, companies may decide it is easier to apply one youth-safety standard broadly than maintain different protections for users in different states.
Other news: Meta agreed to pay up to $18 billion and introduce new teen protections on Facebook and Instagram to settle claims that its platforms were designed to encourage addictive use among young people.