A startup widely known as “ChatGPT for doctors” raised a new round of funding that values the company at $12 billion.
Miami, Florida-based OpenEvidence closed a $250 million financing, led by Thrive Capital and DST, the company told CNBC. The startup first raised outside capital in February, when it raised $75 million from Sequoia at a $1 billion valuation, before its value jumped to $6 billion in October.
In less than a year, OpenEvidence has raised $700 million from investors including from google risk arm, NVIDIAKleiner Perkins, David Sacks' Craft Ventures, and Mayo Clinic.
The company was founded in 2022 by Daniel Nadler, who previously built Kensho Technologies, an artificial intelligence company that was acquired by Standard & Poor's for about $700 million in 2018, and by Zachary Ziegler, a Harvard AI doctoral student. Nadler's new company provides a chatbot for doctors, whose artificial intelligence models have been trained with data and information from leading scientific journals, Nadler said in an interview.
“'ChatGPT for physicians' is a useful abbreviation, but what we really do is help physicians make high-risk clinical decisions at the point of care,” Nadler said. “He is not trained in the open Internet or social media, which can introduce low-quality medical information.”
Nadler said OpenEvidence is the most widely used AI platform by doctors in the US, with more than 40% of doctors using the tool. He pointed to the enormous opportunity in healthcare, which represents nearly 20% of the US gross domestic product with $5 trillion in annual spending.
“Health care is the largest segment of the real economy,” Nadler said. “People realize there could be a lot of winners in this space.”
Those names could include OpenAI and Anthropic.
OpenAI launched “ChatGPT Health” earlier this month, while Anthropic has “Claude Healthcare.” Both are extensions of their popular consumer chatbots that are Health Insurance Portability and Accountability Act, or HIPAA, compliant.
While competition is emerging, Nadler said his company's moat is its focus on physicians, data quality and first-mover advantage.
“We've already collected hundreds of millions of real-world clinical queries from verified doctors; that feedback loop is incredibly difficult to replicate,” Nadler said. “Even if someone copied the manual today, they would still be far behind because it's not just about associations, it's about real-world usage data.”
Trusting advertising
OpenEvidence said it surpassed $100 million in annualized revenue last year, driven primarily by organic growth. Nadler said 95% of new users hear about OpenEvidence from another doctor.
“Most health care in America is not provided in billion-dollar hospitals in New York or San Francisco,” Nadler said. “It's happening in small practices that don't have IT departments or budgets for expensive software.”
OpenEvidence was one of the first AI startups to rely on advertising for revenue, which Nadler said allows for faster adoption and broader use compared to a paid subscription model. Businesses can pay for promotions through video ads in the OpenEvidence app.
The artificial intelligence industry has begun to prepare for advertising-based revenue. Last week, OpenAI said it was testing an ad-supported version of ChatGPT. Nadler said he is trying to be more disciplined than some companies that “openly plan to burn billions or tens of billions in the coming years.”
“That's a big, very risky bet,” Nadler said, adding that OpenEvidence is trying to balance growth with eventual profitability. “We don't run this like a private equity holding company, but we also don't plan to burn billions of dollars over the next year.”
AI startups are picking up this year where they left off in 2025.
In the third quarter of last year, there were six AI funding rounds worth more than $1 billion, according to CB Insights. Anthropic is in talks to raise an additional $10 billion starting in January, while Elon Musk's xAI announced a $20 billion round this month.
Nadler said that with big tech companies aggressively pursuing acquisitions in the space, he felt the pressure, but remained determined to build OpenEvidence as an independent company.
“I've gone down the acquisition route before,” he said. “It can be fantastic, but this time I want to build something that will combine over many years.”
As for an IPO, he said SpaceX, OpenAI and Anthropic, all of which are rumored to be potential candidates by 2026, should go public first.
“There is an order in nature,” Nadler said. “The grassroots model companies go public first. Then the application layer follows. That's how the Internet developed, and that's how this cycle will develop as well.”
LOOK: OpenAI Investor Letter
