Physicians Choose OpenEvidence Over Every Other AI Chatbot Combined in Independent Stanford-Harvard Study of Clinical AI

When free to consult any resource, physicians turned to OpenEvidence more than ChatGPT, Claude, Gemini, and all other external AI chatbots combined — the most-used AI resource physicians chose on their own in the NOHARM study

MIAMI, July 20, 2026 /PRNewswire/ -- OpenEvidence, America's most widely used medical knowledge platform, today announced a central finding of the independent Stanford-Harvard NOHARM study: when practicing physicians were free to consult any clinical decision support AI tool, they chose OpenEvidence more than every other external AI chatbot combined.

The study, NOHARM (Numerous Options Harm Assessment for Risk in Medicine), conducted by ARISE, a clinical AI research network led by physicians from Stanford and Harvard Medical Schools, evaluated 45 large language models and 4 clinical AI systems, with 12,747 expert annotations across 4,249 potential clinical actions.

Although the NOHARM benchmarking studies are not peer reviewed and should be taken with a grain of salt, the study captured something no benchmark can: the revealed preferences of practicing physicians. In a randomized study of 101 board-certified U.S. physicians, one arm gave physicians complete freedom — any tool of their choosing. Working through real clinical cases, physicians reached for OpenEvidence in 22.3% of their responses: more than ChatGPT, Claude, Gemini, and every other external AI combined (19.8%).

This real world data is a leading indicator of revealed preference and physician trust. Revealed preference is the most demanding metric of a clinical tool. Physicians are incredibly savvy users of medical technology. They train for over a decade to apply evidence to real world cases. A physician knows within a handful of queries whether an AI meets the standard of medicine. Physicians do not return to tools that fail them, and their daily use cannot be bought, bundled, or engineered into a workflow. It can only be earned, one accurate answer at a time. When the choice was entirely theirs, they made it: OpenEvidence.

The same study data records the other end of that spectrum. The number of physicians who reached for Doximity Ask — in any case, in any response — was not released.

Why Physicians Choose OpenEvidence

OpenEvidence has never paid, incentivized, or compensated a single physician to use it. Every one of the hundreds of thousands of verified clinicians who consult OpenEvidence daily arrived the same way the physicians in the NOHARM study did: on their own, because they trust what they find there. Adoption of OpenEvidence is a choice, made independently, hundreds of thousands of times a day, at the point of care.

That trust flows from a simple principle: every OpenEvidence answer is sourced, cited, and grounded in the peer-reviewed medical literature.

EvidenceGrade™: Grading the Evidence, Not the Answers

Last week, OpenEvidence extended that principle with the launch of EvidenceGrade™, which grades and visualizes — in real time, on every answer — the quality of the published evidence cited in and used to inform that answer. Built on the GRADE framework, the methodology behind Cochrane, the World Health Organization, and most major clinical guidelines, EvidenceGrade gives every clinician an at-a-glance view of how much weight the underlying evidence can bear.

OpenEvidence believes the trustworthiness of a medical answer is determined by the strength of the evidence beneath it — not by the number of reviewers retained to approve it after the fact. Answer-checking happens behind closed doors and asks physicians to take the result on faith. Evidence grading happens in the open, on every answer, where every clinician can see it and judge it for themselves. That is the standard evidence-based medicine has demanded for thirty years, and it is the standard clinical AI must meet.

"The most important benchmark in clinical AI isn't administered by researchers — it's administered by physicians, hundreds of thousands of times a day, every time they decide what to consult before making a decision that affects a patient," said Daniel Nadler, founder and CEO of OpenEvidence. "Physicians choose OpenEvidence for the same reason they cite the literature: because trust in medicine is earned through evidence, in the open, one answer at a time."

To learn more, visit www.openevidence.com.

About OpenEvidence

OpenEvidence is the most widely-used clinical decision support platform in the United States, and the most widely used medical search engine among U.S. clinicians. OpenEvidence is trusted by hundreds of thousands of verified healthcare professionals to make high-stakes clinical decisions at the point of care that are sourced, cited, and grounded in peer-reviewed medical literature. Independent studies show that more American doctors use OpenEvidence than use all other AI platforms combined. Founded with the mission to organize and expand global medical knowledge, OpenEvidence is actively used daily, on average, by the majority of physicians in the United States.

SOURCE OpenEvidence