Overview
- The October 29 policy update formalizes limits that bar ChatGPT from providing advice that substitutes for licensed professionals.
- OpenAI confirms the service will still offer general, educational information on health, law and finance without issuing diagnoses, prescriptions or tailored legal counsel.
- The rules also prohibit using the tool to automate high‑risk decisions in sensitive domains without human review, including government services, healthcare, finance, migration, education and national security.
- Initial reports claiming a total removal of medical or legal information were corrected after OpenAI clarified the change is a tightening of existing restrictions, not a blanket ban.
- The update is active in the United States and some European countries, with rollout to Latin America reported as pending.