A tamper-evident audit layer for the whole transaction — identity, offer, escrow, funds, title, deed — across both US and EU law in one kernel. Court-reproducible proof that a specific party made a specific decision at a specific time. It does not warrant clean title; it proves what happened.
Buyer, seller, agents, title, escrow, lender, recorder — each in a siloed system. The closing package is a PDF stack on a file server. When wire instructions are spoofed, a deed is stolen, or a closing is litigated, there is no canonical signed event ledger to replay. The same gap exists in every jurisdiction.
US real-estate wire fraud losses in 2025 — up 59% YoY (FBI IC3). Business Email Compromise overall hit $3.04B; the vector is spoofed wire instructions between parties with no shared signed record.
A signed escrow chain with verified parties makes a rerouted wire detectable before funds move.
of industry participants witnessed title fraud or deed theft in the past year (NAR 2025). Reversing a fraudulent transfer costs $50–150K in legal fees; 16 US states have no deed-fraud law.
No cryptographic binding ties the grantor's identity to the signed closing record.
Average US residential close — 150–200+ discrete tasks across 6+ independent parties. The EU adds 3–4M closings/yr under 27 distinct land-law regimes.
Reconstructing a disputed closing means subpoenaing six parties for six different versions.
Each party records their own slice in their own system. Nothing binds the escrow release to the offer it settles, or the deed to the identity that signed it. Anima has each party sign their own event at the moment it happens — and chains each capsule to the digest of the one upstream.
→ The deed-transfer capsule will not hash unless both a title-verification digest and a funds-confirmation digest are present. The closing condition is enforced in code, not workflow.
E-signatures carry handwritten weight when attribution is reliable. No specific technical standard is required — EIP-712 signed digests with party-identity chaining satisfy it.
49 states + DC have permanent remote-notarization statutes; the 2025 SECURE Notarization Act adds interstate recognition. RON records need anchoring into the closing chain.
QES has handwritten-signature effect across all 27 states. EUDI Wallets issue by end-2026; relying parties must accept them in 2027. Our capsule encodes all four regimes.
HM Land Registry digital-by-default; Estonia's e-Land Register; Sweden's blockchain pilot. Mandatory digital submission with cross-border eIDAS recognition.
Non-financed transfers to entities/trusts must be reported (vacated 2025, revision anticipated). The duty falls on settlement, title, and escrow agents — our RegulatorAuditView buyers.
No law mandates the technical form — but every one rewards a record that opposing counsel or a regulator can self-verify offline in minutes, not weeks of discovery.
Practice-management software for title and escrow. They produce a closing file that lives in their database — a workflow extract, not independently verifiable outside the platform.
E-signature, RON, eVault. They attest that a signature happened — but the audit log is proprietary and platform-bound. They don't chain identity → offer → escrow → funds → deed.
AI title underwriting and blockchain land-registry pilots. Faster risk pricing, or a new on-chain registry needing a government mandate — not a signed, party-held chain over existing rails.
Every capsule is an EIP-712 typed struct, secp256k1-signed, with input schema, output schema, unit test, and a Go↔Rust/WASM parity-emit test in CI. EU-jurisdiction capsules marked with a purple edge. Source: internal/realestate/signing/ · range 0x9000…0010–00F0.
No incumbent produces this: a chain where the title agent signs the title capsule, the bank signs the funds capsule, the recorder signs the deed — each referencing the digest above it, each re-verifiable offline.
The deed_transfer capsule structurally refuses to hash unless both a title-verification digest and a funds-confirmation digest are non-zero. The closing condition — title clear, money settled — is a cryptographic invariant, not a checklist. The same gate exists in both the US and EU deed structs.
→ One kernel, two legal systems: distinct US and EU structs for title and deed, a four-regime signature capsule, and an anchor to any existing trust authority. No new blockchain required.
Closings per year — ~5–6M in the US ($2–3T value) plus ~3–4M across EU member states. Every one needs a signed identity → deed chain that no system produces today.
US title-insurance premiums written (ALTA, 2023) — the industry that directly absorbs the fraud and title risk Anima attests to. EU notarial and registry fees add a parallel pool under eIDAS.
Anima is not a title oracle. It does not warrant that title is clean or that no fraud occurred — it attests that a specific party made a specific decision, with these inputs, at this time, and lets anyone replay it. That distinction is the legal moat.
Each link signed by the party who was there. Re-verifiable offline, by anyone, in any jurisdiction.
For the first time, a closing can prove itself.