OpenAI Releases ChatGPT for Clinicians, Offering Free Access to U.S. Healthcare Professionals

U.S. The tool is designed to support clinical tasks like documentation and medical research. U.S. A 2026 survey by the American Medical Association found that 72% of physicians report using AI in clinical practice.

Physician use of AI in clinical practice increased from 48% in the previous year to 72% in 2026. Millions of clinicians worldwide use ChatGPT to support clinical care every week, with clinician usage more than doubling over the past year. Earlier this year, the company introduced ChatGPT for Healthcare, allowing organizations to deploy ChatGPT to clinicians, administrators and researchers with compliance and controls.

The company also introduced HealthBench Professional on April 22, 2026. HealthBench Professional is an open benchmark for real clinician chat tasks across three use cases: care consult, writing and documentation, and medical research. HealthBench Professional builds on HealthBench’s broader evaluation of health conversations.

The company’s team worked with hundreds of physician advisors to inform and improve capabilities for ChatGPT for Clinicians. ChatGPT for Clinicians includes free access to current frontier models for healthcare use cases. ChatGPT for Clinicians includes skills for repeatable clinical workflows such as referral letters, prior authorization and patient instructions.

It offers trusted clinical search with real-time, cited answers based on evidence from millions of reputable, peer-reviewed medical sources. The tool enables deep research across medical journals to compile comprehensive, well-cited reports in minutes.

ChatGPT for Clinicians includes continuing medical education credits from eligible evidence review in clinical questions without separate courses or extra paperwork.

It offers optional support for HIPAA compliance through a Business Associate Agreement for eligible accounts. Conversations in ChatGPT for Clinicians are not used to train models. The company’s physician advisors have reviewed more than 700,000 model responses.

The company’s models rank as the top-performing systems for real-world healthcare use on Stanford’s MedHELM and MedMarks evaluations. Physician advisors tested 6,924 conversations in ChatGPT for Clinicians before release.

Physicians rated 99.6% of responses in the 6,924 conversations as safe and accurate. On a subset of 355 examples, ChatGPT for Clinicians cited ground-truth sources more often than human physicians. HealthBench Professional uses physician-authored conversations and rubrics with multi-stage physician adjudication.

About a third of HealthBench Professional examples involved physicians red teaming. The company reported results in ChatGPT for Clinicians and across models, with human physicians providing responses as a baseline.

“This version of ChatGPT is as close to an ideal clinical support partner as it gets. It’s like an on-demand consultant I can engage on everything from current guidelines to billing and coding, with the added benefit of broad access to pediatric and pediatric subspecialty literature,” an unnamed clinician stated in the company's announcement.

U.S. healthcare system faces extraordinary strain, with clinicians managing more patients alongside growing administrative demands and expanding medical research, according to the announcement.

U.S. health systems use ChatGPT for Healthcare to handle administrative work like medical research and documentation. The company stated that enabling free access to ChatGPT for Clinicians supports its mission to ensure AGI benefits all of humanity.

The company plans to expand access to additional countries and groups over time, starting with pilots for verified clinicians outside the United States through the Better Evidence Network, as permitted by local regulations. U.S. The company emphasized ongoing collaboration across health systems, clinicians, patients, regulators and technology companies to improve human health through AI.

Physician advisors continuously review model responses and provide feedback on quality, reasoning, trustworthiness and safety. Every few minutes, a new model response is reviewed by a physician. ChatGPT for Clinicians is designed to support clinicians with information, not replace their judgment or expertise.

HealthBench Professional examples were chosen for quality, representativeness and difficulty. The company reported results in ChatGPT for Clinicians and across models, with human physicians providing responses as a baseline.

Account security features include protections like multi-factor authentication to keep sensitive work secure. ChatGPT for Clinicians allows users to turn common workflows into reusable skills for consistent task handling. It enables delegation of medical literature reviews, with users setting trusted sources and steering research as needed.