Research literacy, for more of us.

Our mission is to help people understand how research earns its conclusions—and where the evidence leaves room for doubt.
A better relationship with research starts with being able to question it.
A headline, a journal name, or a confident AI answer is not enough. MedLit gives curious readers, students, and researchers a structured place to examine a paper’s methods, source passages, missing information, and limitations.

What we’re building toward

Evidence you can inspect

Each assessable judgment should lead back to the text that supports it. A matched quotation helps you check an interpretation; it does not prove the interpretation is correct.

Uncertainty you can see

Missing methods, abstract-only coverage, and unsupported claims should remain visible. “Unknown” is a useful answer, not a product failure to disguise.

A reader who stays in control

Choose your own AI provider, review its costs and privacy settings, and decide what to save or share. An account is optional. Your provider keys stay out of cloud library sync.

Learning that lasts beyond one paper

Understand the questions behind an appraisal, annotate what matters to you, and return to a saved report without paying for another AI response.

What we do not promise

MedLit cannot certify that a paper is true, unbiased, reproducible, or appropriate for a patient. It does not replace peer review, systematic-review teams, domain expertise, or medical care. Our appraisal confidence score describes source coverage and assessability—not scientific truth.
Read the methodology

Other work from the founder

Anya Ozmen also builds tools for cognitive practice, creative work, and financial education.
Synaptik: research-informed cognitive practiceGameer: games and creative toolsRedvest: financial educationRedko: product and AI studio
Explore an example
Bring your own paper