OpenEvidence alternative: ownership and workflow compared

Clinicians searching for an OpenEvidence alternative usually have two questions. Who is behind OpenEvidence, and how should another product compare? The first question needs dated public evidence. The second needs a workflow test, not a marketing claim.

This article is published by Primary AI, so the comparison is not independent. It uses public sources, names the limits of those sources, and gives clinicians a test they can run themselves.

Who owns OpenEvidence?

OpenEvidence is a privately held company founded by Daniel Nadler and Zachary Ziegler. Its official About page names the founders and lists institutional supporters that include Sequoia, Kleiner Perkins, GV, Nvidia, Thrive, Andreessen Horowitz, Coatue, and Mayo Clinic, among others.[1]

The company announced a large funding round in July 2025 and named GV and Kleiner Perkins as co-leads, with other investors participating.[2] A Forbes profile from the same month reported an estimate of Daniel Nadler's ownership at that time.[3] That report is dated and came before later financing. It should not be read as a current cap table.

The careful answer is simple: the founders and investors own a private company, but the exact current ownership split is not available on the public company pages cited here. A private financing announcement states who invested. It does not usually show every shareholder, option pool, dilution term, or current percentage.

What makes a useful alternative?

An alternative should be compared on the job it performs. For clinical evidence tools, use criteria that affect the answer in practice.

Evidence and citation behaviour

Ask whether important claims have sources that open. Then check whether the source supports the exact claim. A product can cite a real paper and still overstate the result or apply it to the wrong population. Evidence certainty also matters. GRADE gives clinicians a structured way to think about confidence in an effect estimate.[6]

Guideline geography

A clinician in Canada may need national society guidance, Health Canada information, provincial coverage rules, and local pathways. A clinician in the United States may need FDA information, US society guidance, and local formulary rules. The product should show which geography shaped the answer instead of silently combining different recommendations.

Workflow breadth

Some clinicians want a focused medical search product. Others want one clinical assistant that connects evidence support with notes, referrals, forms, or requisitions. Neither design is automatically better. The right choice depends on whether the user wants a narrow evidence workflow or a broader assistant.

Privacy boundary

Read the current terms, privacy policy, and security material for each product. Confirm whether identifiable health information is permitted, which organizations process data, and what your workplace allows. Do not infer a privacy promise from a medical design or a clinician verification screen.

Business model

Ask how the service is funded and how that could affect the interface. Relevant questions include subscription fees, advertising, sponsored placement, institutional contracts, and limits on free use. A business model does not by itself prove bias, but clinicians should be able to recognize commercial content and separate it from evidence.

Where Primary AI differs

Primary AI describes itself as a physician founded clinical AI company for licensed clinicians.[4] Its product combines cited clinical decision support with workflow tools. The answer engine is designed to use Canadian and US sources. The Scribe product supports documentation and connected clinical work.

Primary AI also publishes its security and data boundaries.[5] Those statements should be read directly before use because product terms can change. As with any clinical AI tool, users remain responsible for removing identifiers when required, checking sources, and applying professional judgment.

This does not make Primary AI the correct choice for every OpenEvidence user. A clinician who wants one specific search experience may prefer a different interface. A team that needs Canadian context and connected workflow tools may give more weight to Primary AI.

A fair comparison test

Run the same set of questions in each product. Do not start with an obscure question designed to make one system fail. Use normal work.

• Ask one common question with stable guideline support.
• Ask one question where Canadian and US guidance may differ.
• Ask one question with limited or conflicting evidence.
• Open the sources for every claim that could change care.
• Record how often the source supports the wording.
• Record whether the product states uncertainty.
• Note how much time verification takes.
• Review privacy and commercial disclosures separately.

The evidence gap question is especially useful. A safe answer should become more cautious when reliable support is weak. It should not compensate with a longer response.

Bottom line

Public information shows that OpenEvidence is a private, venture backed company founded by Daniel Nadler and Zachary Ziegler.[1][2] Public reports can describe ownership at a point in time, but they do not provide a live cap table.[3]

Primary AI is an alternative for licensed clinicians who want cited clinical answers, Canadian and US context, and broader workflow support.[4][5] The most reliable comparison is direct. Use the same questions, inspect the sources, check the privacy boundary, and choose the tool that fits the work you actually do.