In short
- OpenAI launched ChatGPT for Clinicians, a unique model designed to support clinical tasks such as documentation and clinical research.
- The platform is free for US-certified physicians, nurses, physician assistants, and pharmacists.
- OpenAI said the chatbot outperformed human doctors in some medical tasks, according to its tests.
OpenAI on Wednesday unveiled a free, exclusive version of the ChatGPT for doctorsnurses, physician assistants, and pharmacists, expanding the company’s commitment to a healthcare sector struggling with staff shortages, workloads, and burnout.
The product, called ChatGPT for Clinicians, is designed to handle documentation, clinical research and care dialogue—time-consuming tasks that eat up the hours doctors can spend with patients. Access is currently limited to certified physicians in the United States, with plans to expand globally.
The announcement comes as the adoption of AI in medicine continues to grow. According to the American Medical Association’s 2026 survey cited by OpenAI, 72% of physicians now use AI in healthcare, up from 48% a year ago. The company says that clinical use of its platform has doubled in the past year, with millions relying on ChatGPT every week.
Some of the uses of these tools are a medical research service that captures millions of peer-reviewed subjects, an in-depth search method for reviewing medical records, user-friendly templates that can be used for applications such as referral letters and pre-authorization requests, and the ability to earn money for continuing medical education by researching clinical questions on the platform.
Discussions will not be used to teach OpenAI models, and HIPAA compliance support is available through a Business Associate agreement for eligible accounts.
Along with the launch, OpenAI released HealthBench Professional, a new benchmark designed to evaluate the performance of AI in healthcare in three areas: care communication, documentation, and clinical research.
The company reported that GPT-5.4, running in the ChatGPT for Clinicians workspace, scored 59.0 on the benchmark—higher than human doctors, who scored 43.7 even with unlimited time and internet, and higher than competing models from Anthropic, Google, and xAI.
These results, however, come with an important caveat: OpenAI generated all the data generated by the benchmark used for testing.
To develop the tool, the company says it worked with hundreds of medical consultants and reviewed more than 700,000 responses. In testing, doctors rated 99.6 percent of responses as safe and accurate in nearly 7,000 conversations.
OpenAI has been careful to present the tool as a tool rather than a change in clinical practice – regulators and skeptics alike are likely to keep an eye on it as the product becomes more widespread.
Healthcare represents the fastest growing market for AI tools. Further to OpenAI data showing that clinical use of ChatGPT has doubled in the past year, McKinsey data shows 50% of healthcare leaders say their organizations will use artificial intelligence, up from 47% in Q4 2024 and 25% in Q4 2023. BCG research, meanwhile, represents 60% of consumers already use AI for your health.
Daily Debrief A letter
Start each day with top stories right here, including originals, podcasts, videos and more.
