OpenEvidence, a three-year-old healthcare AI startup often described as the “ChatGPT for doctors,” is raising $250 million in fresh capital at a $12 billion valuation, doubling its valuation in just two months. The surge comes as the company reaches $150 million in annualized advertising revenue, nearly three times growth since August, underscoring strong momentum in physician adoption and monetization.

OpenEvidence has rapidly emerged as a dominant clinical AI platform in the US, claiming usage by 45% of American physicians, far outpacing ChatGPT’s 16% and Abridge’s 5% penetration among doctors. With nearly one million physicians in the US, the platform now answers around 20 million medical questions every month, averaging 20 queries per physician user. This scale highlights the platform’s deep integration into clinical workflows, where accuracy and trust are paramount.

Unlike general-purpose AI chatbots, OpenEvidence operates on an ad-supported medical AI model, selling targeted advertising to pharmaceutical companies similar to Google’s search advertising playbook. Currently, the startup monetizes only 10% of its available ad inventory, suggesting the potential to exceed $1 billion in annualized revenue once fully scaled. This model allows OpenEvidence to maintain gross margins above 90%, avoiding the high variable costs associated with API-based AI pricing while capitalizing on physician attention as a premium channel for pharma marketing.

A key differentiator for OpenEvidence is its licensed content moat. The platform sources authoritative medical information from leading journals, including the New England Journal of Medicine, offering clinicians legally sound, evidence-backed answers. This gives OpenEvidence a crucial advantage in regulated healthcare environments where accuracy, liability, and trust are non-negotiable areas where general AI models face limitations.

OpenEvidence’s rise places it among a select group of vertical AI companies achieving $10B+ valuations and $100M+ revenue. Its trajectory reinforces a broader industry lesson: domain-specific AI, trained on licensed expert content and paired with vertical-native monetization models, can outperform broad AI platforms in regulated industries. The company’s success signals that in healthcare AI, going deep rather than broad may be the winning strategy.