A bill attempting to regulate the ever-growing industry of companion AI chatbots is now law in California, as of October 13th.

The Governor of California, Gavin Newsom, has approved Senate Bill 243, described as the pioneering “AI chatbot safeguards” by state senator Anthony Padilla. This legislation mandates that developers of companion chatbots integrate new safety measures—specifically, when there’s a likelihood that a user might mistakenly think they are communicating with a human, the chatbot creator must provide a clear notification that confirms it is solely AI, not a human being.

Beginning next year, certain chatbot operators will be required by this law to submit yearly reports to the Office of Suicide Prevention. These reports should detail the measures taken to identify, address, and respond to situations where users exhibit suicidal thoughts. Furthermore, the Office is tasked to make such information publicly available on its website.

“Innovative technologies like chatbots and social media have the power to motivate, inform, and unite people; however, without proper safeguards, they can also manipulate, deceive, and pose threats to our children,” stated Newsom during the signing of the bill, alongside other legislation crafted to enhance online child safety, including new age verification rules for devices. “Our advancement in AI and technology must be balanced with a strong sense of responsibility, ensuring our children are protected at every turn. The safety of our children is non-negotiable.”

This announcement follows the official signing of Senate Bill 53 by Governor Newsom, a significant AI transparency law that sparked debate among AI firms and drew significant public attention over several months in California.