The Potential and Future of Medical AI: What is Open Evidence?

I usually write articles for people who use generative AI to write or study, but
this time, I am changing my target audience slightly.

This article is intended for:
👨🏻‍⚕️ Healthcare professionals
🩺 People who encounter medical and healthcare information on a daily basis.

1. The sense of discomfort often felt when researching medical information with AI

Generative AI like ChatGPT is useful for
summarizing disease overviews and catching up on fields outside your expertise.

On the other hand, in a medical context, you often encounter the following situations:

・The explanation is easy to understand
・It is convincing as a general theory
・However, no evidence is provided

“Which paper is this information based on?”
“How is this positioned in the guidelines?”

This kind of verification becomes necessary,
and in the end, you return to searching for literature yourself.

There must be many people whose time and concentration are taken up by this back-and-forth.

What solves such problems is
👉“Open Evidence”.

2. What is Open Evidence?

Open Evidence is
👉an AI search tool specialized in the medical field.

It was founded in 2022 by entrepreneur Daniel Nadler.

This service was born out of the Mayo Clinic's “Platform Accelerate” program,
and researchers associated with Harvard University and MIT also participated in its development.

Unlike general-purpose generative AI,
it is characterized by being designed from the start for use by doctors and healthcare professionals✍

While general generative AI handles “broad knowledge,”
Open Evidence constructs its answers based on primary and secondary information such as:

・Medical papers
・Systematic reviews
・Clinical guidelines.

According to reports,

its usage has spread rapidly since its founding,
and as of 2025,
it is used by over 40% of US physicians👨🏻‍⚕️Moreover,

anyone can use it for free🚀

3. Features I feel when actually using it

Personally, the points I find helpful while using it are as follows👇

・Citations are clearly indicated in the answers
・You can check related papers on the spot
・Conclusions and prerequisites are organized separately

Therefore,
“Read only the conclusion”
“Pick up only the paper titles”
“Proceed to the original text”
You can control these options yourself.

Rather than a substitute for PubMed searches,
I have the impression that it is a “tool to prepare the entrance” before entering into literature research.

4. How it is used in the field

Open Evidence is
starting to be used in actual medical settings.

For example,

・When organizing differential diagnoses
・When checking evidence for treatment options
・When checking if your own thinking is not significantly off from the guidelines

It is not an “answer” to decide on diagnosis or treatment,
but it seems to be used in many cases as a way to verify the materials leading up to a judgment.In medical education settings,

it is also used as an aid to create a flow of:

・First, grasp the overall picture
・Next, read the original paper.

I myself also


often use it as a “warm-up before reading papers directly.”

5. How to use it differently from ChatGPT

Just because Open Evidence exists,
it does not mean ChatGPT becomes unnecessary.

👉The roles are different

ChatGPT is for:
・Grasping the overall picture
・Explanations in plain language
・An entrance to fields outside your expertise

Open Evidence is for:
・Verification of evidence
・Backing up information based on papers
・Grasping clinical positioning

By using both in parallel,
the time spent getting lost from researching too much has decreased.

6. For those who will use it from now on

If you are going to use Open Evidence from now on,
there are points you should keep in mind.

・Treat the answers as reference information
・Read through the citations once
・Think about whether it applies to your own case background

Open Evidence is
not a tool that acts as a proxy for judgment.

It is an entity that presents materials for thinking
in an organized form🧠

7. The future of Open Evidence and medical AI

The amount of information handled in the medical field will continue to increase in the future.

What is required in that context is
rather than “reading everything,”
⚡️“reaching the necessary information quickly.”Medical-specialized AI like Open Evidence

should take on the role of organizing the process leading to judgment,
rather than the quality of the judgment itself.
Before deciding whether to use it,

just knowing that such options exist
might change how you face information a little bit.