California lawmakers have sent Gov. Gavin Newsom a bill that would impose sweeping safety requirements on AI chatbots used by minors, setting up a crucial test for a governor who has often sided with the technology industry. The bill, Senate Bill 1119, was passed by the legislature on Monday and now waits for Newsom’s signature. The final stretch was marked by a report that OpenAI CEO Sam Altman tried to speak with Newsom last week in an effort to influence the legislation.

A new regulatory floor for chatbots

The measure is designed to address growing concern about children interacting with AI companions and assistants. It would require AI companies to determine how old a user is before allowing chatbot access. It would also require companies to give parents tools to monitor and limit how much time their children spend with services such as ChatGPT. Independent audits would be required to show whether the companies are complying with the law’s rules, and providers would need to prepare for potential enforcement actions.

Lawmakers wrote the bill with a series of privacy protections likely to shape the national debate over AI and child safety. The bill prohibits AI companies from serving advertisements to minors if those ads rely on personal information obtained through the child’s conversation with a chatbot. It would also make it unlawful to sell data collected from children’s chats. Services would still be allowed to present “age-appropriate contextual advertising” to users under 18, but the legislation says such ads must be clearly labeled as advertisements.

The bill was introduced by state Sen. Steve Padilla and co-authored by Assemblymembers Buffy Wicks and Rebecca Bauer-Kahan. It was amended on Friday, with changes that narrowed the circumstances in which families could sue AI companies. That revision was significant because private rights of action can be much more threatening to tech firms than enforcement by state agencies alone. It is unclear exactly what prompted the Friday amendments and what role, if any, Altman played in shaping them.

OpenAI’s lobbying and response

According to a report, Altman attempted to get time with Newsom before the bill reached its final vote. The report did not say whether Altman supported or opposed the bill in that conversation, and it did not describe any policy concessions he may have sought. A person familiar with the matter denied that the governor and Altman spoke directly last week, but OpenAI did not dispute that outreach was attempted. An OpenAI spokesperson declined to comment on the record about Altman’s contact with Newsom’s office.

OpenAI’s public posture toward SB 1119 has been largely supportive. The company published a statement shortly before the legislature passed the bill in which it urged Newsom to sign it. The statement said SB 1119 builds on youth safety measures that OpenAI has supported “through our products, global policy principles, advocacy in California, and work on the Parents & Kids Safe AI Act.”

OpenAI also pointed to its recent launch of a dedicated version of ChatGPT for teenagers. The company rolled out “ChatGPT for Teens” on Aug. 18, adding safeguards aimed at users between the ages of 13 and 17. OpenAI said the product is designed to help users under 18 “learn, think critically, and build while promoting healthy use.” The safeguards are not optional, according to the company. If OpenAI’s system estimates that a user is under 18, or if the user states that they are 13 to 17, that user is automatically placed into the teen version with protections already enabled.

Those protections, OpenAI said, are part of the baseline experience rather than settings users can turn off. The company’s statement did not explicitly say that the bill should be changed, and it described the legislation as an important step toward protecting young users. Still, the report of Altman’s outreach underscored how much is at stake for companies that build generative AI products popular with teenagers.

Newsom’s complicated relationship with Big Tech

Newsom has historically been a friend to Silicon Valley. During his time as mayor of San Francisco and later as California’s governor, he built close relationships with technology executives and has repeatedly vetoed legislation he believed would be too restrictive for innovation. His record has frustrated labor groups and consumer advocates who argue that the state has been slow to regulate data collection, algorithmic decision-making and AI safety.

The political landscape has shifted, however. Recent months have brought intensive public scrutiny to AI chatbots following reports of young people who died by suicide after being encouraged by AI systems. Those cases have intensified pressure on lawmakers to act quickly, and they have made it harder for tech companies to dismiss safety rules as unnecessary. A backlash against data centers and the energy demands of AI has also weakened the assumption that technology growth should be accepted unconditionally.

Newsom is term-limited and will leave the governor’s office on Jan. 7, 2027. He is widely expected to run for president in 2028, and his recent public activity has been viewed through that lens. He has worked in recent years to appeal to right-leaning audiences, including by hosting a podcast on which he appeared with figures such as Charlie Kirk and Steve Bannon. In those conversations, Newsom has at times sounded more aligned with the far right than with progressives who demand stricter controls on tech.

What happens next in California

The bill cannot become law unless Newsom signs it. He has not said publicly whether he supports SB 1119, and his office has not released details about conversations with stakeholders. OpenAI told reporters that Newsom spoke with a wide variety of interested parties before the bill passed, which suggests the governor’s team was navigating competing pressures from industry and safety advocates.

The next governor of California will inherit whatever policy framework is in place. Democrat Xavier Becerra and Republican Steve Hilton are the leading candidates in the November election. Becerra has proposed modest guardrails for AI and has received substantial campaign donations from Silicon Valley. Hilton has argued for less regulation across the board. He is married to Rachel Whetstone, a longtime tech executive who has worked for Uber, Netflix, Facebook and Google.

Hilton’s campaign has emphasized economic freedom and has criticized what he describes as overreach by Sacramento. Becerra, by contrast, has said the state needs safeguards against the worst harms of AI, though his proposals are less aggressive than the versions of SB 1119 that were introduced earlier in the legislative session.

Recent polling shows Becerra with a commanding lead. The most recent survey from the University of California Berkeley Institute of Government Studies put Becerra ahead of Hilton among likely voters by 55 percent to 37 percent. A Becerra victory would likely continue California’s current approach of cautious regulation. A Hilton victory could mean a swift move in the opposite direction, with less enforcement and a more hands-off attitude toward AI companies.

For now, the immediate question is what Newsom will do with SB 1119. Because the legislative session has ended, the bill could be signed, vetoed, or allowed to become law without his signature. If he vetoes it, supporters will need to start over with a new measure. If he signs it, California would become the first state with a comprehensive kids’ AI chatbot safety framework, and that could set a nationwide standard for how ChatGPT and similar products treat minors.

Source:Gizmodo News