The company behind ChatGPT didn’t set out to tackle health care. But despite ongoing concerns about its technology’s tendency to hallucinate, OpenAI is already inking deals with health care customers desperate to use it to speed up workflows without burdening their staff.
Best known for its free public chat tool used by millions as a conversational search engine, OpenAI also sells secure, enterprise commercial licenses for ChatGPT and its API to a range of health customers. Early adopters tell STAT they’re aware of generative AI’s risks, and are careful not to let it dispense medical advice directly to patients — instead, they’re harnessing its ability to draw from disparate data sources to generate coherent responses in narrowly-defined settings: summarizing medical records, suggesting treatment plans, or distilling medical jargon into plain language, among others.
OpenAI is hardly the only company selling generative AI technology to health customers; STAT has reported that dozens of health systems across the country are adopting it for purposes ranging from ambient scribing to what’s called “clinical decision support” for practitioners. Companies including Oracle, Palantir, and Microsoft, as well as upstarts like Abridge, are also pushing deeper into generative AI in health care.
This article is exclusive to STAT+ subscribers
Unlock this article — and get additional analysis of the technologies disrupting health care — by subscribing to STAT+.
Already have an account? Log in