OpenAI Opens a Health Hub in ChatGPT. Experts Urge Caution.

OpenAI has launched a dedicated health section inside ChatGPT that lets users upload medical records such as lab results and medication lists, and connect Apple Health so the chatbot can draw on fitness and health data.

"With your permission, ChatGPT can now draw on relevant information you've connected in Health to help you compare a new result with prior tests, summarize changes since your last appointment, or explore how sleep, activity, and workouts relate to your routine," the company said in a news release Thursday.

OpenAI says 300 million people already bring health-related questions to ChatGPT each week. To Tanzeem Choudhury, chief of health innovation at Cornell Tech, that number says as much about American medicine as it does about the chatbot.

"Our healthcare is broken," Choudhury told CBS News. "I think that's why we are seeing so many people asking for advice."

Choudhury, who has worked at the intersection of artificial intelligence and health for more than 15 years, said AI can be a democratizing force for people who face barriers to care, but added that "we just need to think about the risks and have better protections in place."

Those risks include hallucinations — confident but inaccurate or misleading answers. CBS News reported this week on a Florida man who said ChatGPT's medical advice nearly cost him his life when he suffered a pulmonary embolism. An OpenAI spokesperson told CBS News that "ChatGPT is not a doctor and should never be used as a substitute for medical care, diagnosis or treatment." Months earlier, according to CBS News, a California teenager died after the chatbot told him a combination of substances was safe to consume. OpenAI said at the time that "this is a heartbreaking situation, and our thoughts are with the family," and said that version of ChatGPT is no longer publicly available.

OpenAI said physicians helped build and test realistic health scenarios before release, and that the health feature adds encryption protections and will confirm with users before acting through a connected plugin. Records shared with ChatGPT will not be used to train its models or target ads, the company said. It also cautioned that ChatGPT "does not replace the care and judgment of qualified medical professionals."

Choudhury said it is unclear whether such privacy commitments are permanent, noting that AI companies have quietly revised terms of service before and that the Federal Trade Commission has warned consumers about sharing data. For now, she said, the burden falls on users to know when to escalate a concern — a judgment that may be harder for elderly or cognitively impaired people.

Vishal Misra, a Columbia University computer science professor and vice dean of computing and artificial intelligence, said he has used ChatGPT to track his own health data and sees real value. "It can find connections that maybe a really competent physician might miss," he said. But he warned: "The same risks that exist sharing health data with anyone transfer here. You are sharing something very personal."

His rule is simple. "Whatever it says, take it to an actual human physician," he said. "Do not act. The only action that you should take is to consult your physician."

CBS News has asked OpenAI whether ChatGPT is equipped to recognize when a user should be told to see a doctor.