Preprints explained: arXiv, bioRxiv and reading research early

If you want to read the newest research in a field, you'll keep running into preprints. They're how a lot of science is shared first — fast, free and open. But reading them well means understanding exactly what you're getting, and what you're not.

What a preprint is

A preprint is a complete research paper that the authors post publicly before — or alongside — submitting it to a journal for peer review. It's the real study, written up in full, but it hasn't yet been checked by independent experts. Think of it as the author's own draft of record: fast to appear, free to read, but not yet vetted.

The main preprint servers

• arXiv — physics, mathematics, computer science and related fields; the original preprint server.
• bioRxiv — biology and life sciences.
• medRxiv — medicine and health (read these with extra care, as findings can affect clinical decisions).
• SSRN — social sciences, economics, law and the humanities.
• ChemRxiv, Research Square and others cover chemistry and broad multidisciplinary work.

The trade-off, in plain terms

• Upside — you read results months or years earlier, for free, with no paywall.
• Upside — you see work that might never fit a traditional journal, including negative results.
• Downside — no peer review yet, so errors, overstated claims or flaws may not have been caught.
• Downside — the findings can change between the preprint and the final published version.

How to cite a preprint

Cite it as a preprint, not as a journal article — make the status explicit so readers know it wasn't peer-reviewed. Include the authors, year, title, the server name (e.g. arXiv), the identifier or DOI, and the word “preprint.” If the work is later published in a journal, cite that final version instead whenever you can.