How It Works
Describe your collection, upload your files, and let AI draft structured metadata — then review every field, with confidence scores, before anything lands in your catalogue. The AI does the first draft; you stay the archivist.
Create an accession and add your files
Start by creating an accession. Give it a title, date range, and any context you have — donor information, archival scope, or condition notes. Then upload your digitised files: photographs, documents, scanned records, audio, or video.
Your observations are injected into the AI prompts, so the more context you provide, the better the results. But even a title and a batch of files is enough to get started.
Every item analysed, transcribed and catalogued
Hit process and the AI gets to work. It reads each item — running OCR on documents, analysing photographs, transcribing audio — and drafts structured metadata: titles, dates, descriptions, subject terms, and ISAD(G)-aligned fields. All AI processing runs in EU data centres.
Before processing starts, the app shows you exactly how many credits the job will use — no surprises. Processing then happens in the background: close the browser and come back later, or watch the progress in real time as each item moves from queued to complete.
You approve, edit, or reject every field
Every AI-generated field carries a confidence score. Low-confidence fields are flagged so you know exactly where to focus your attention. For each suggestion, the AI explains its reasoning — what it saw in the document and why it chose that value.
Edit any field, add your own notes, flag items for colleagues, or mark sensitive content with access restrictions. Nothing is final until you say it is. The AI is a tireless assistant, not a replacement for professional judgement.
Standards-ready outputs for your systems
Export your catalogued collection in the format your systems need: EAD3 XML finding aids, BagIt packages with PREMIS provenance, Dublin Core XML, or CSV files formatted for direct import into AtoM, ArchivesSpace, or Archivematica.
Every export includes provenance for AI-assisted fields — recording model family, timestamp, and review state alongside the archivist who signed off each field. Your exports carry the audit detail downstream systems expect.
Works with: AtoM, ArchivesSpace, Axiell, Modes, Adlib, Mimsy, Preservica, Archivematica, and other repository and CMS platforms. Your system of record stays where it is.
Explore your collection instantly
Once processed, your entire collection is searchable. Find items by keyword, date, subject, or format — all powered by the metadata your AI and archivists created together.
Professional and above
The analysis suite goes further: whole-collection analysis to surface themes and identify gaps, fonds and ISAD(G) drafting, research dossiers, and AI organise for arrangement suggestions. Available on Professional plans and above.
Radical Price Transparency
Archivers.ai is credit-based, with transparent pricing, published when we launch. One credit catalogues one photograph or object, one document page, one minute of video, or five minutes of audio. An Explore message uses one credit; deep AI analyses use five. The app shows you the credit cost before you commit.
Volunteer and reviewer seats are free on every paid plan.
Join the waitlist