OpenEvidence has confirmed that it raised an additional $250 million at a $15 billion valuation, according to a Reuters report and statements from CEO Daniel Nadler. The round takes the clinical AI company's total funding to roughly $1 billion, and marks a 25% valuation climb from the $12 billion mark set just nine months ago, when it raised an identically sized $250 million tranche in January.
The more consequential news is what the money is for. Nadler says OpenEvidence will use Anthropic's base models to roll out its clinical AI tools for free across nearly 100 countries in Africa and Asia - a geographic expansion into markets where paid clinical reference products have little penetration. The platform is already free to physicians in the United States, where the company says it has become a daily research tool for a large share of practicing doctors.
The business model that pays for all of this is unusual: OpenEvidence charges doctors nothing and sells in-app advertising to pharmaceutical companies and medical device manufacturers instead. Nadler, who previously founded financial analytics firm Kensho and sold it to S&P Global in 2018, founded OpenEvidence in 2021. The company also operates Doctor Dialer, a white-label telemedicine product built on its AI stack.
The incumbents are not standing still. UpToDate, the decades-old clinical reference service owned by Wolters Kluwer, and Doximity - which counts more than 3 million healthcare professionals as users - have both pushed their own AI decision-support features. The pressure is coming from a crowded field more broadly: health AI startups raised more than $4.1 billion in the second quarter of 2026 alone, including at least six rounds above $100 million, and clinical-scribing rival Abridge carries a reported $5.3 billion valuation.
There are open questions worth holding onto. HIStalk's September funding snapshot reports that insiders suggest OpenEvidence may be open to considering a sale - an odd signal for a company that just re-priced upward. And industry commentary has flagged the tension baked into the Anthropic partnership: frontier labs are partnering with vertical clinical applications whose workflows they are simultaneously learning to serve directly, a dependency that cuts in both directions as models improve.
For Anthropic, the deal is distribution into one of the most regulated, hardest-to-enter software domains there is, without taking on clinical compliance itself. For OpenEvidence, it is frontier-model capability without a training bill - the trade that lets a 300-person clinical software company hand out AI tools in a hundred countries for free. Whether advertising revenue can sustain free access at global scale is the assumption the entire structure rests on.
Comments (0)
Log in to join the discussion
Log InNo comments yet