How We Work

Everyman Science exists to explain real research to people who are curious about it but do not work in it. This page sets out plainly how the articles on this site are put together, because you are entitled to know that before you decide how much to trust them.

Where the facts come from

Every article on this site is built from human-produced reporting and human-produced research: press releases and mission updates from NASA and ESA, papers and news from Nature, university and laboratory announcements, and established science publications. Those sources are named and linked at the foot of every article, and where an article is about a specific study we link the paper itself, so you can go and read the primary source rather than taking our word for it.

We do not use AI to generate facts, findings, quotes or figures. Anything specific in an article — a number, a date, an instrument, a name, a result — comes from the cited sources. Where the sources do not establish something, we say so rather than filling the gap.

Where AI is used

We use AI tools to structure, draft, format and optimise articles from that sourced material, and occasionally to produce an illustration when no suitably licensed photograph of a subject exists. Where a cover image is an illustration rather than a photograph, the credit line says so. Photographs carry their original photographer and licence.

One article is published per day. Each one is written to a specific format — a reported feature, a close reading of a single new paper, an explainer, or a piece built around a data table — and each one carries a section we call the deeper cut: a short excursus pitched deliberately above the level of the rest of the article, for readers who have a scientific background and find pure popular science unsatisfying. If you skip it, you lose nothing from the article. If you want it, it is there.

Corrections

An automated desk gets things wrong, and when it does we would rather hear about it than not. If you find an error — a misread finding, a wrong figure, a misattributed image, a study characterised in a way its authors would not recognise — write to [email protected]. We read every message and correct what needs correcting.

— The editors, Everyman Science