Ask the Archive

How This Works — And What It Isn’t

This is a real relevance search, not an AI chatbot. There is no language model writing an answer, no possibility of an invented fact or a hallucinated citation — your question is broken into keywords, matched against real records, and every result is a direct link to something that actually exists. It runs in your browser, and nothing you type is logged or stored.

It searches two quite different things, and keeps them visibly apart. The curated archive is this site’s own researched, sourced, credibility-scored work: case files, theories, key figures, locations, ancient records, craft types, Essential Reading breakdowns and declassified documents. The raw sighting datasets are four independent third-party report archives — NUFORC, Larry Hatch’s UFO*DB, the French government’s GEIPAN, and the U.S. National Archives’ Project Blue Book case files — which this site indexes and links but has not individually investigated or scored. A row in one of those is a report somebody filed, not a finding; every raw result says which archive it came from and links to that archive’s own record.

Ranking rewards pages that match more of your question, not just one word, and gives extra weight to a match in the title over a match buried in the description — so “Who is Bob Lazar?” puts the Bob Lazar key-figure card first, not every case file that happens to mention his name in passing. It also recognizes a handful of common synonyms in this field (alien/extraterrestrial, UFO/UAP, cover-up/deception, and similar), so close wording still finds the right page. The three file-based sighting datasets are searched in your browser and load on your first question rather than on page load, so the page itself stays fast; NUFORC’s 147,000-odd records are far too large to send to a browser, so those are queried through this site’s own search API instead.