The two kinds of thing on this site, and which ones you should verify before relying on them.
Preserved sources (primary). Original documents, posts, and transcripts, rendered from the collection’s mirrors and never altered. These are faithful to the record.
Machine-generated layers (derived). Summaries, extracted titles, “authority” and affiliation records, timelines, and the librarian’s answers are produced by software, including AI models, from the primary sources. They are navigation aids, not authoritative statements — and, especially where they describe specific people, dates, positions, or quotations, they can be wrong. Always verify against the primary source before relying on or quoting them, and tell us if we got something wrong — email the curator. Corrections from the people who were there are the most valuable thing we receive.
This is a plain-language notice for a research prototype, not legal advice. See also our privacy notice, content & copyright statement, preserved-vs-machine-generated note, and terms of use.