Krux
April 24, 2026
OpenAI's Doctor-Focused ChatGPT Beats Human Physicians on Clinical Benchmark
Published: April 24, 2026 at 12:20 AM
Updated: April 24, 2026 at 12:20 AM
100-word summary
OpenAI launched ChatGPT for Clinicians, a free tool for verified U.S. doctors, nurse practitioners, physician assistants, and pharmacists. It handles documentation, medical research, and care consults. The surprise? On a new 525-task benchmark built from real physician conversations, the specialized model outperformed both standard ChatGPT and human doctors. The gap was biggest in writing tasks: the clinician version scored 64.1 versus 32.1 for the next-best option. OpenAI tested 700,000 responses with physician reviewers and found 99.6% safe and accurate. The tool also cited sources more often than human doctors did. Free access means no budget meetings, just verification and login.
What happened
OpenAI launched ChatGPT for Clinicians, a free tool for verified U.S. doctors, nurse practitioners, physician assistants, and pharmacists. It handles documentation, medical research, and care consults. The surprise? On a new 525-task benchmark built from real physician conversations, the specialized model outperformed both standard ChatGPT and human doctors. The gap was biggest in writing tasks: the clinician version scored 64.1 versus 32.1 for the next-best option. OpenAI tested 700,000 responses with physician reviewers and found 99.6% safe and accurate. The tool also cited sources more often than human doctors did.
Why it matters
Free access means no budget meetings, just verification and login.