STAT+: OpenEvidence launches new family of AI models

OpenEvidence, an AI platform already utilized by over 1.12 million licensed U.S. clinicians, has forged a partnership with a nationally leading cancer center to enhance its oncology capabilities. This collaboration aims to integrate a precision oncology knowledge base directly into the platform's workflows, enabling wider use by healthcare providers across the country.

OpenEvidence, founded by Daniel Nadler, Ph.D., is also on a mission to develop "medical superintelligence" through agentic artificial intelligence—a collection of AI agents specializing in various clinical fields. Nadler envisions this system acting as a comprehensive, multispecialty care team that physicians can access on their devices, offering equal access to expert knowledge regardless of geographic or resource limitations.

OpenEvidence has already integrated the National Comprehensive Cancer Network (NCCN) evidence-based oncology guidelines and ASCO Guidelines into its platform. With this new partnership, users gain access to expert-curated interpretations of cancer genomic alterations, patient-specific context, and peer-reviewed evidence. This approach empowers clinicians to make more informed treatment decisions.

The company has also unveiled a new family of medical AI models, including OpenEvidence Darwin, which achieved a perfect score on MedQA, a medical AI benchmark. Darwin is currently being used by institutional partners such as the National Organization for Rare Disorders (NORD). The company's other models, Osler, Sackett, and Snow, are designed to cater to different search speeds and depths, providing accurate clinical information to users.

OpenEvidence is actively expanding its presence in various clinical workflows, facing competition from established players like Wolters Kluwer s UpToDate and Elsevier.