OpenEvidence is pitching an entrepreneurial bet that goes beyond a single medical chatbot: build “medical super-intelligence” by assembling agentic AI subspecialists that can collaborate on a case the way specialist teams do inside major hospitals. Co-Founder Daniel Nadler said the company is already laying groundwork with a multimodal, multicloud approach and argued that the next step in healthcare AI is moving from one general system to many domain-trained agents that can “have a discussion” about treatment decisions.
The pitch is backed by distribution and usage numbers that most early-stage health AI companies can’t claim. OpenEvidence says the product is used daily by more than 40% of U.S. physicians across more than 10,000 hospitals and medical centers, and Nadler said it supported about 18 million clinical consultations in December 2025. The company also frames differentiation around content access and trust, pointing to licensing partnerships with major medical organizations and journals, plus an oncology push through a licensing agreement that makes NCCN guidelines available through its platform.
Financing is following the roadmap. The three-year-old startup reached a $6 billion valuation after a $200 million Series C round in October 2025 and has raised nearly $500 million since 2022, which Nadler tied to plans to train specialist models for clinical subspecialties and then connect them through an “ensemble” architecture. He also positioned that approach as a way to reach smaller and rural practices that often sit outside big enterprise deployments, arguing the product can scale through the same grassroots adoption pattern that got it traction with clinicians in the first place.
