AI Transparency & Editorial Policy

How our content is produced

Much of the content on this site is drafted by an automated pipeline that retrieves and summarizes published research (for example, peer-reviewed studies indexed in PubMed and guidance from recognized medical bodies). Every clinical claim is intended to trace back to a cited source, and our pipeline runs automated checks — including citation and claim verification — before content is published.

Honest review labeling

We do not claim medical review that has not happened. Content that a named, credentialed clinician has personally reviewed is labeled as reviewed, with the reviewer identified. Other automated research content is identified as AI-assisted evidence, not a clinician endorsement. We never attach a fabricated reviewer name or review date.

Our tiered review model

Not all content carries the same risk, so our review depth is tiered:

• Higher-risk guidance (for example, symptoms, treatments, medications, and interactions) requires review by a credentialed clinician before it is presented as guidance.
• Research summaries (such as our COVID-19 monitor pages) are produced by the automated pipeline with claim verification. They are clearly identified as AI-assisted evidence and are not presented as clinician-endorsed.
• Tools and navigational content receive a lighter review appropriate to their lower risk.

If a clinician later reviews a page, we identify that reviewer and date. If reviewed content changes materially, the old review label is removed until the changed content receives a new review.

Sources and dating

Where content is data-driven, we show when the underlying data was last updated so you can judge its freshness. Citations link to their original sources so you can verify claims yourself.

Limitations

AI-assisted content can contain errors, omit context, or become out of date. It is educational information, not medical advice — please read it alongside our medical disclaimer.