How Google Is Drafting AI Chatbot Bills Across the U.S.

Summary

An NPR investigation examines how Google and other technology companies influenced state legislation intended to make AI chatbots safer, particularly for children. The reporting begins with Cynthia Montoya, whose 13-year-old daughter Juliana used Character.AI chatbots before dying by suicide in 2023; phone records showed sexual content and distressing conversations, and Montoya later became an advocate for stronger safeguards. NPR found that at least 75 lawsuits have been filed against AI developers over alleged chatbot harms, many involving children, while the federal government has enacted no general laws governing AI systems or their developers. In 2026, chatbot bills were introduced in at least 10 states with identical or near-identical provisions, including exemptions for developer tools, chatbots embedded in other software, search engines, business products, voice assistants, and services without user accounts. Critics say such language could exclude widely used systems such as ChatGPT, Claude, Gemini, Meta AI, Grok, and Alexa, while also giving companies liability protections or allowing them to decide what safety measures are technically feasible. NPR reports that Google lobbyists supported bills in several states and that lawmakers in Colorado and Hawaii said Google provided or helped shape legislative language; Google said it supports effective legislation and works with lawmakers and community advocates to promote safe AI. Colorado lawmakers defended their law as a first step shaped by practical enforcement concerns, while advocates argued that its exemptions could leave children with fewer protections. Hawaii removed proposed exemptions before enacting its law, and Arizona’s governor vetoed a bill she said protected technology companies more than children. Chatbot laws were nevertheless enacted in several states, including Colorado, Idaho, Georgia, Iowa, Washington, Nebraska, and Oregon.