Google helps draft AI chatbot safety laws while seeking product exemptions

State legislators across the country are accelerating efforts to impose regulatory oversight on artificial‑intelligence chatbots, citing concerns over misinformation, privacy breaches and potential harms to vulnerable users. Bills introduced in California, Texas and New York propose mandatory safety testing, transparency disclosures and the establishment of state‑run oversight boards to certify that conversational agents meet defined risk‑mitigation standards before they can be offered to the public. The proposals have drawn bipartisan support, with lawmakers arguing that existing federal guidance does not adequately address the rapid deployment of generative AI tools in consumer applications.

At the same time, major technology firms are shaping the emerging regulatory framework by submitting self‑crafted safety guidelines that seek to carve out exemptions for their flagship products. Companies such as OpenAI, Google and Microsoft have presented model‑specific compliance documents that argue certain large‑scale language models already incorporate “robust internal safeguards” and therefore should not be subject to the same external testing requirements as smaller, third‑party chatbots. Critics say these exemptions could create an uneven playing field and undermine the intended consumer protections, prompting consumer‑advocacy groups to call for stricter, uniform standards that apply to all AI conversational agents regardless of the developer.