For carrier underwriting & fraud teams
Property-protection evidence — signed at the chip, audit-grade by default.
Controlled-environment agriculture, perishable cold-chain, and high-value equipment all carry property-protection policies that pay out on claims-side guesswork. We turn the same loss events into continuous signed evidence streams with cryptographic chain of custody — a property-protection telemetry layer your underwriting and fraud teams can audit on their own infrastructure.
What this replaces.
Today's claims workflow is reactive — a phone call, an adjuster site visit, photos, paper. Evidence is human-curated after the fact, often after the loss has compounded. With signed telemetry running in the policy period, the evidence layer changes shape.
Phone call → adjuster visit → photos → paper claim form → carrier review → days or weeks to decision. Evidence is curated after the loss event by the party paying out.
Continuous signed telemetry → event consensus at the chip → cryptographic proof bundle → underwriter / carrier reads the same canonical JSON the operator does. Decision-grade evidence on day one.
Three concrete use cases.
Each ships against an existing platform surface. The proof chain is the same chain operations sees on the live console — it doesn't get assembled at claim time; it's already there.
Greenhouse flood / freeze parametric triggers
Sensor consensus on a wet or sub-freezing event fires a signed proof bundle the moment the threshold is crossed. Parametric triggers pay out from telemetry, not from a clipboard.
Working prototype: detects < 1 cm of standing water and fires an alarm within ~20 minutes. Below the threshold that damages greenhouse benches; above the noise floor of normal irrigation runoff. Self-powered: biological battery, no power line / PoE / cellular contract.
Cold-chain perishable lots
Temperature excursions on perishable inventory get a signed event the moment they breach. Pay-fast decision evidence for lots that stayed in spec; denial-of-claim evidence for lots that didn't.
Continuous-monitoring discount band
Operators who keep monitoring active throughout the policy period qualify for a continuous-monitoring discount — the evidence is the audit-grade monthly report we already produce. Underwriters get verifiable proof the discount is earned, not asserted.
Audit on your own infrastructure.
Your fraud team runs the BioSwarm Attestor verifier on their own infrastructure. Three commands. We never see your keys; you never trust ours. The verifier re-fetches live data, re-computes the KPIs from raw node readings, walks the signature chain, and produces its own Ed25519 + ML-DSA-65 (FIPS 204) co-signed attestation envelope. ~1.8k LoC Python, 120+ unit tests, RFC 6962 Merkle bundle for 24-hour evidence rollups.
If our telemetry ever diverges from the recomputation, the verifier emits a mismatch verdict. That's the property-protection asymmetry — we publish; you audit; nothing crosses except canonical JSON.
What carriers get.
All real surfaces, live today. Replace each link with the corresponding view in your own underwriter dashboard.
/admin/contact-queue.csv demonstrates the pattern — every queue, log, and event exports the same way.
GET /proof/{zone} · HMAC + Ed25519 + ML-DSA-65 co-signature · partner pulls + archives, never round-trips through us.
Pilot scope.
Pilot a single zone for 60 days. Watch one greenhouse, cold room, or equipment unit. At the end of the pilot:
- A stack of signed monthly reports your underwriting team has already seen.
- Two carrier-side test cases — at least one real event recorded end-to-end through the proof chain.
- A verifier audit run by your fraud team (we don't have to be present).
- A claims-side ROI delta against a no-telemetry baseline operator in the same region.
After 60 days you decide: kill, scope, or expand. The platform stays running either way; the operator is already paying for the underlying monitoring.