Nearly nine in ten teens who use ChatGPT turn to it for learning, information, skill-building or productivity in a given week. The figure is OpenAI's and opens its position on California's SB 1119.
Fact
The facts: on 31 August 2026 OpenAI announced its support for California Senate Bill 1119 and urged Governor Newsom to sign it. The post is by Ann O'Leary, VP of Global Policy. According to OpenAI, the bill requires determining a user's age, identifying and addressing safety risks before a product is made available to young people, independent audits, protection from harmful content including self-harm and sexually exploitative content, parental tools, connecting young people with crisis support when serious risks arise, limiting targeted advertising and protecting young people's personal information. For users aged 13 to 17 the protections should apply automatically. The company also points to its own ChatGPT for Teens: if the system estimates a user is under 18, or the user states an age between 13 and 17, they are placed into it automatically. OpenAI notes that the bill preserves access to educational features, including responsible use of ChatGPT's memory.
The requirements are good and I cannot see what is missing from them. But when you read a company's position on a law, look at where it gets most specific.
Here it is in one paragraph: the bill rightly recognises that AI is not social media, and preserves responsible use of features like ChatGPT's memory. Memory carries information over from your earlier conversations. OpenAI defends it as a tool for safety and learning. It may well be, but it is also part of the product the company sells.
When a company backs a law, look for the paragraph where it praises it most specifically.
Whether Newsom will sign, we do not know. We do know that, in OpenAI's own words, the bill builds on measures it already supports in its products. That is not a bad thing. It is just good to see it.
Official primary sources
OpenAI - OpenAI supports California's bill to advance youth AI safety, 31.08.2026The visual is generated code-art, no third-party images.