Editor's Note: This story has been updated since its initial publication on Sept. 3 to reflect the MSK partnership.
OpenEvidence is leveraging a partnership with Memorial Sloan Kettering Cancer Center to bolster its oncology capabilities, adding trusted genomic interpretation expertise to the AI platform already used by millions of clinicians.
Through the partnership, the company will integrate OncoKB, MSK's precision oncology knowledge base, directly into its workflows for broader use by providers outside of MSK. OpenEvidence previewed the partnership exclusively with Fierce Healthcare on Sept. 3, initially referring to the partner as a "nationally leading cancer center." Last week, OpenEvidence revealed the partner as MSK.
The partnership brings precision oncology decision support to clinicians nationwide and deepens OpenEvidence's push into cancer care as the company already directly integrates the foundational frameworks of cancer treatment into its medical search engine. The company has a licensing agreement with the National Comprehensive Cancer Network (NCCN), which integrates its evidence-based oncology guidelines into its platform. The company also integrates ASCO Guidelines, figures and flowcharts into its model.
Through its latest partnership with MSK, OpenEvidence users gain access to expert-curated interpretations of cancer genomic alterations alongside patient-specific context and peer-reviewed evidence, helping clinicians make more informed treatment decisions, according to OpenEvidence executives.
MSK will also integrate OpenEvidence within its Epic workflow.
MSK's OncoKB provides evidence-based assertions on whether genetic alterations detected in cancer are clinically actionable biomarkers that may guide subsequent treatment decisions. Through this partnership, OncoKB's curated, cancer type-specific annotations that include the oncogenic effect, level of evidence and associated therapies for each alteration will be surfaced within OpenEvidence, within Epic, alongside patient context, so that clinicians asking about a specific alteration can see MSK's expert assessment of its clinical actionability alongside answers grounded in the peer-reviewed literature.
"OncoKB is the gold standard for making sense of a tumor's genetics, built and maintained by the experts at MSK who have defined this field. Bringing it into OpenEvidence, within the EHR, with patient context, means a physician anywhere can pair the primary literature with MSK-curated interpretation of exactly which alterations matter and what to do about them, turning a complex genomic report into a clear, evidence-based path forward," Travis Zack, M.D., OpenEvidence's chief medical officer, said in a statement.
"AI has the potential to fundamentally change how we translate an increasingly complex and rapidly expanding body of knowledge into better decisions for patients," said Anaeze Offodile, M.D., chief strategy officer at MSK, in a statement. "Realizing that potential requires more than powerful technology—it requires thoughtful integration into how care is delivered. Our work with OpenEvidence reflects MSK’s broader approach to AI: embedding trusted tools into clinical practice in ways that augment the expertise of our clinicians, reduce friction, and ultimately enable more informed, evidence-based care."
The partnership also builds on OpenEvidence's vision to develop “medical superintelligence” based on agentic artificial intelligence, or a system of AI agents that can be subspecialists in clinical areas, as OpenEvidence founder Daniel Nadler, Ph.D. detailed back in January at the 2026 J.P. Morgan Healthcare Conference.
OpenEvidence has set its sights on building medical specialist AI models, starting with oncology.
Just as neural networks digitally recreated the architecture of intelligence, "medical superintelligence" will digitally recreate groups of experts and specialists, "composing many subspecialty expert models to build a single, more powerful intelligent system acting as a full multispecialty care team that every physician can hold in their hand," Nadler told Fierce Healthcare.
The company is now integrating MSK's precision oncology knowledge base directly into its specialized oncology sub-agent—an AI agent that has already officially digitized the NCCN treatment algorithms, Nadler noted. He said the company has partnered with several nationally leading cancer centers and those partnerships would be announced "in the coming weeks."
"In the near future, we will be rolling out the next OpenEvidence specialist agents in genetics, cardiology and neurology. Our vision is that, in aggregate, this dynamic ensemble of medically specialized agents will not only transcend the limitations of any individual human expert but will also result in a seismic-scale equalization of healthcare quality in America. This will allow every rural or under-resourced county in the country—whose physicians historically had scarce access to sub-specialist knowledge—to carry in their white coat pocket a world-class multispecialty care team or tumor board, on-call 24/7—for free," Nadler said.
Nadler told Forbes last week that OpenEvidence also plans to use the MSK data, combined with the NCCN data, to develop its own pipeline of potential therapies.
The company recently reached a $15 billion valuation after a previously unreported $250 million funding round led by Byers Capital and Andreessen Horowitz in early September, Forbes also reported. Nadler confirmed the funding round to Fierce Healthcare.
OpenEvidence developed an AI-powered medical search engine and a generative AI chatbot exclusively for doctors, which summarizes and simplifies evidence-based medical information. As of September, there were 1.12 million medical licensed-verified U.S. clinicians, including physicians, nurses, nurse practitioners and physician assistants, using OpenEvidence.
August is tracking to over 40 million NPI-verified queries from U.S. clinicians in the last 30 days, Nadler told Fierce Healthcare.
The company also released a new family of medical AI models, including OpenEvidence Darwin, what the company calls its most advanced medical AI model, which is now in research preview.
According to OpenEvidence, Darwin is the first AI model in history to achieve a perfect score on MedQA, a fully-independent medical AI benchmark. The company also touted Darwin's performance on other leading benchmarks of medical AI: MedXpertQA (72.8%), HealthBench Professional (82.7%) and NOHARM (87.2%), ahead of the next-best models (Claude Fable 5 and Gemini 3.7).
Darwin is available by application only, currently being used by institutional partners such as the National Organization for Rare Disorders (NORD), research collaborators and accredited AI researchers at academic institutions, the company said. "As Darwin's safeguards are validated with these partners, its capabilities will flow, model by model, into Osler, Sackett, and Snow — the frontier arriving at the point of care as fast as it can arrive safely," the company said.
The company named the models after the founders of modern medicine. OpenEvidence said it built Osler (named for William Osler), what it calls its fastest model that provides answers in about five seconds, for the pace of point of care. Osler is the successor to the model that currently powers OpenEvidence answers, and will continue to be the default model on the platform, the company said.
OpenEvidence Sackett (named for David Sackett) is a deeper search model for questions that turn on the weight of the evidence, according to the company, and Snow (named for John Snow) is OpenEvidence's deepest production model that runs a full investigation of world medical literature before providing a report.
"Every model in the family is held to the same standard of clinical accuracy. What varies is time: how long a model thinks, and how deep it searches," the company said in a blog post.
The company said the Osler, Sackett and Snow models are rolling out to all OpenEvidence users today—free to verified clinicians.
In the past four years, OpenEvidence has expanded from clinical search into other clinical workflows, which puts it into more direct competition with players like Wolters Kluwer's UpToDate and Elsevier as well as Abridge and Doximity.
In August, the company rolled out its Visits feature, a clinical AI assistant that transcribes patient visits. The company made an AI-integrated doctor dialer feature more widely available, directly taking on Doximity's core business. In March, it released an AI-powered medical coding feature that provides automatic Current Procedural Terminology (CPT) code suggestions and evaluation and management (E/M) level recommendations. In May, OpenEvidence rolled out a voice AI feature that gives physicians a hands-free way to ask questions and get evidence-based answers. The company said Voice Mode is a native speech-to-speech medical AI interface, calling it the first multimodal medical AI offering for clinical decision support.