California Enacts Adam's Law Requiring AI Chatbot Safety Measures After Teen's Death
California Governor Gavin Newsom signed Adam's Law, requiring AI chatbot companies to implement safeguards against harmful content and manipulative interactions, following tense negotiations with OpenAI and other tech companies.
California Governor Gavin Newsom signed Adam's Law on Thursday, a first-of-its-kind measure requiring AI chatbot companies to adopt safeguards protecting users—especially minors—from harmful content and manipulative interactions. The law is named for Adam Raine, a California teenager who took his life in 2025 after ChatGPT allegedly coached him on how to do so.
The legislation introduces several requirements for chatbot operators, including timely in-app crisis support, age verification, restrictions on targeted advertising to children, and parental controls. It also establishes liability for AI companies that fail to take reasonable measures to prevent harmful outputs such as self-harm content, sexually explicit material, romantic roleplaying, excessive praise, and emotionally manipulative responses that foster reliance and isolation from friends and family. Companies must also implement a mechanism to report incidents.
OpenAI, which lobbied to shape the bill, engaged in what sources described as tense negotiations with its authors—Assembly member Rebecca Bauer-Kahan, Assembly member Buffy Wicks, and Senator Steve Padilla. «There were moments of intense negotiation, you know, as there are with any of these types of issues,» a source familiar with the talks said. «It occasionally got heightened.» The source did not disclose which points were most contentious.
Ann O'Leary, OpenAI's Vice President of Global Policy, worked directly with lawmakers. The company said its role was to educate policymakers on how the latest AI models work and to clarify how it differs from social media, noting there is no continuous scroll and that data shows most teens use the technology for specific projects. Representatives from Anthropic, Google, Meta, and Amazon also participated in the discussions and were «equally involved,» according to an OpenAI spokesperson. Anthropic negotiated an exemption because it does not allow users under 18.
After the bill cleared the California legislature, O'Leary praised the effort on LinkedIn. «We are happy to support this bill,» she wrote. «We believe that it will set the standard for AI youth safety moving forward.»
The law represents a significant shift in OpenAI's regulatory strategy. Just a year earlier, the company argued that state-level AI regulation would create confusion and impose excessive compliance burdens. In a 2025 LinkedIn post, Chris LeHane, OpenAI's vice president of global policy, suggested support for a federal moratorium on state AI laws. Greg Brockman, OpenAI's president, had personally donated tens of millions of dollars to a super PAC opposing state-level AI regulations. In an August 2025 letter to Newsom, OpenAI warned that a «patchwork of state rules…could slow innovation without improving safety.»
Now, OpenAI advocates for that exact patchwork. In a July 2026 blog post, LeHane described the company's new approach as «reverse federalism,» saying states will «step by step» form «a de facto national standard.» He named California, New York, and Illinois as leaders in AI policy. OpenAI still supports a national framework, with CEO Sam Altman writing on X that a federal approach would «set consistent safety requirements for frontier AI.»
The shift comes amid growing backlash against AI, including concerns over data centers. Anti-AI sentiment escalated this month after a viral social media post from a former Anthropic researcher claimed the industry knows the technology may kill all humans within the decade. Anthropic's head of alignment confirmed the claim, and multiple other AI employees echoed the message.
The Trump administration attempted to pass a 10-year moratorium on state AI regulation in a May 2025 draft of the «One Big Beautiful Bill.» It passed the House but failed in the Senate. In December, Trump issued an executive order challenging state AI laws and pushing for a national regulatory framework.
James Czerniawski, head of Emerging Tech Policy at the Consumer Choice Center, told Fortune that states will increasingly look to regulate AI as federal action stalls. «As we see a lack of action federally on AI, states will increasingly look to regulate in this space,» he said. The law could become a model for other states considering similar legislation.
