OpenAI has launched ChatGPT for Clinicians, a specialized version of ChatGPT tailored to assist with clinical tasks such as documentation and medical research, enabling clinicians to dedicate more time to patient care. The tool is being offered at no cost to verified physicians, nurse practitioners, physician assistants, and pharmacists in the United States.
The U.S. healthcare system is under significant pressure, with clinicians handling increasing patient loads, growing administrative burdens, and an ever-expanding volume of medical literature. Many have already adopted AI tools for support. A 2026 American Medical Association survey found that physician AI usage has reached an all-time high, with 72% of physicians now using AI in clinical practice-up from 48% the prior year. Millions of clinicians worldwide already use ChatGPT weekly for tasks like care consultation, writing and documentation, and medical research, and clinician usage has more than doubled over the past year.
As AI adoption in clinical settings accelerates, OpenAI emphasizes the growing responsibility to improve model performance and safety for clinical use cases. Earlier in 2026, OpenAI introduced ChatGPT for Healthcare, which allows organizations to deploy ChatGPT to clinicians, administrators, and researchers with enterprise-grade compliance and controls. Clinicians at leading U.S. health systems are already using it to streamline administrative tasks and reclaim time for patient care.
Providing free access to ChatGPT for Clinicians is the next step in OpenAI's mission to ensure AGI benefits all of humanity.
Alongside this launch, OpenAI is also releasing HealthBench Professional, an open benchmark for evaluating AI on real clinician chat tasks across three categories: care consultation, writing and documentation, and medical research. It builds on the broader HealthBench evaluation of health-related conversations.
Designed for and with Clinicians
OpenAI's team collaborated with hundreds of physician advisors to shape and refine ChatGPT for Clinicians, ensuring it supports key clinical use cases. The tool includes:
- Advanced AI models for complex clinical questions: Free access to OpenAI's current frontier models for healthcare scenarios, helping handle questions, research, and documentation more reliably.
- Skills for repeatable clinical workflows: Common workflows can be turned into reusable skills so ChatGPT follows the same steps each time for tasks like referral letters, prior authorization, and patient instructions.
- Trusted clinical search: Real-time, cited answers drawn from millions of reputable, peer-reviewed medical sources to support faster and more confident case reasoning.
- Deep research across medical journals: Medical literature reviews can be delegated to ChatGPT, with the ability to set trusted sources, steer the research, and receive a comprehensive, well-cited report in minutes.
- CME from real clinical questions: Eligible evidence reviews performed in ChatGPT can automatically count toward continuing medical education credits, without requiring separate courses or extra paperwork.
- Optional HIPAA compliance support: While many clinical tasks do not involve protected health information, HIPAA support is available through a Business Associate Agreement (BAA) for eligible accounts.
- Account security and privacy: Conversations are not used to train models, and protections such as multi-factor authentication help keep sensitive work secure.
Ongoing Evaluation and Strengthening of Health Performance and Safety
OpenAI continuously works to improve the safety and accuracy of ChatGPT's responses in health-related scenarios. OpenAI's physician advisors regularly review model outputs, providing feedback on quality, reasoning, trustworthiness, and safety. To date, they have reviewed more than 700,000 model responses reflecting real-world clinician and patient usage patterns, with a new model response reviewed by a physician every few minutes.
OpenAI's models rank as the top-performing systems for real-world healthcare use on third-party evaluations such as Stanford's MedHELM and MedMarks. ChatGPT for Clinicians builds on models like GPT-5.4, which outperform other models on OpenAI's HealthBench.
Before release, physician advisors tested 6,924 conversations across clinical care, documentation, and research in their daily work. Physicians rated 99.6% of responses as safe and accurate. In a subset of 355 examples where three independent physicians specified ground-truth citations, ChatGPT for Clinicians cited those sources more frequently than human physicians. Nevertheless, the tool is designed to support clinicians with information, not replace their judgment or expertise.
HealthBench Professional uses physician-authored conversations and rubrics, multi-stage physician adjudication, and careful data filtering to measure performance and safety in common clinician chats. About a third of the benchmark examples involved physicians deliberately red-teaming the models, and the dataset was curated to include conversations that were 3.5 times more difficult for the models. Results show that GPT-5.4 in the ChatGPT for Clinicians workspace outperforms base GPT-5.4, all other OpenAI and external models, and human physicians.
Global Access
The free version of ChatGPT for Clinicians is currently available to verified U.S. physicians, nurse practitioners, physician assistants, and pharmacists.
OpenAI plans to expand access to additional countries and clinician groups over time. In the coming months, the company will begin working with the Better Evidence Network to pilot access for verified clinicians outside the United States, subject to local regulations.
OpenAI believes that improving human health will be one of the defining impacts of AI, but realizing that potential requires close collaboration among health systems, clinicians, patients, regulators, and technology companies worldwide. Alongside these updates, OpenAI is also releasing a Health Blueprint offering recommendations for the responsible integration of AI in healthcare in the U.S.