Licensing
The immigration corpus your product is missing.
If you’re building legal AI, practice-management software, or research tooling that touches immigration law, BIA Edge licenses the data layer so you don’t spend a year rebuilding it: a cleaned, enriched, citator-tracked corpus that updates itself every morning.
What’s in the corpus
- Hundreds of thousands of unpublished BIA and AAO decisions — acquired, OCR’d, deduplicated, and enriched with outcome and favorability tags. Available nowhere else at this scale.
- The published record — BIA and Attorney General precedent, circuit and Supreme Court immigration decisions, curated district court families, BALCA, and OCAHO.
- Statutes and regulations — the INA, Title 8, the relevant CFR titles, and state criminal codes ingested from official sources for crimmigration work.
- Agency guidance — USCIS Policy Manual, policy memos, the FAM, practice manuals, country-conditions reporting, and visa bulletin history.
The enrichment graph
Raw documents are a commodity; the structured layer on top is what makes agent answers trustworthy. Licensed access can include:
- ImmCite — validity and treatment signals for precedential authorities, maintained against current official directives.
- Holdings and headnotes — structured, citable summaries extracted at corpus scale: 215,000+ holdings and 220,000+ headnotes.
- Citation graph — cross-references linking cases, statutes, and regulations in both directions.
- Pinpoint passages — source-text spans with stable IDs, built for retrieval-augmented citation.
- Doctrine graph — reviewed relationships among precedents: limiting authority, related holdings, doctrine families.
How licensing works
Start the conversation
Tell us what you’re building and which slices of the corpus matter to you. Pilot evaluations against the live MCP endpoint are straightforward to arrange.
BIA Edge was built by a practicing immigration attorney. The public library at biaedge.com stays free; licensing covers programmatic and bulk access to the data layer.