KYE Authority Risk Signal™ — a signed feed of authority-failure events for boards, insurers & risk platforms.
Every governed action emits an evidence event when authority fails. The KYE Authority Risk Signal™ is a signed feed over those events. It counts seven authority-failure classes — denials, escalations, unauthorised attempts, authority drift, evidence gaps, finality suspensions, and revocations. It is a projection: counts and signed references, never dollar values. KYE Protocol™ emits the signal; your board, insurer, or risk platform prices the exposure. It is not financial-risk quantification or advice.
Boards and insurers need a credible authority-risk input — and the only honest one is the proven event, not a guessed number.
Risk committees, cyber insurers, and risk-quant platforms all want to price the exposure created when AI-agent and delegated authority fails. The credible input is not another vendor risk score — it is the signed, tenant-scoped record of which authority events actually occurred. Four facts converge:
- The consequential moment is the authority failure — and it is already evidenced. A denial, an escalation, an attempt to act outside scope, declared-vs-deployed authority drift, an evidence gap, a finality suspension, a revocation: each already emits an evidence event on the audit chain. The signal is a projection over those events, not a new measurement.
- Counts are defensible; estimates are contestable. A count of authority-failure events in a window is a fact the consumer can independently verify against the referenced audit rows. KYE Protocol™ ships the counts and the references — never a likelihood, a loss, or an exposure range it would have to defend.
- Replay-verifiable from public keys alone. Every feed record is Ed25519-signed; the board, insurer, or platform recomputes the canonical hash and verifies the signature against the published trust keys, with no KYE Protocol™ secret required. The signal is trustworthy without trusting the sender.
- This is a signal, not a model. KYE Protocol™ does not compete with the insurer’s underwriting models or the risk platform’s quantification engine. It supplies the proven authority-failure signal those models lack — and leaves the pricing to them.
Counts you can verify, references you can pull, a signature you can check — the proven authority-failure signal, ready for your model.
- Seven event-class counts per window. Denials, escalations, unauthorised attempts, authority drift, evidence gaps, finality suspensions, revocations — each a tally of governed events derived from a real evidence-event family on the audit chain, never a dollar value.
- Severity mix, classified at the event. The feed classifies the window’s events by operational severity (low / medium / high / critical) — a classification of the events, not a financial-risk score or loss estimate.
- Signed references to the real evidence. Every material event carries a reference back to its audit row, so your team pulls the underlying evidence and verifies the signal independently — the feed asserts no new truth.
- Tenant-isolated by construction. Each feed is scoped to exactly one trust domain; cross-tenant aggregation never happens. Your signal is yours.
- Channels into the Insurance pack and external risk platforms. The same signed feed feeds the KYE Protocol™ Insurance pack and machine-consumable exports for external risk-quant platforms — they price the exposure; KYE Protocol™ proves the events.
A derived, signed projection — computed deterministically over the evidence you already produce.
One deterministic projector reads the audit-chain events for a tenant and a window, counts the seven authority-failure classes, classifies severity, references the most material events, and seals the record with an Ed25519 signature. Nothing is invented; the record is a projection.
- 1 — Authority fails, evidence is emitted. A denial, escalation, unauthorised attempt, drift, evidence gap, finality suspension, or revocation emits its evidence event on the audit chain at the moment it occurs.
- 2 — The projector aggregates the window. For one trust domain and one time window, the projector counts each event class, builds the severity mix, and selects the most material event references — deterministically, with cross-tenant events dropped.
- 3 — The record is signed. The feed record is sealed with an Ed25519 signature over its canonical JSON, carrying the seven counts, the severity mix, the references, and the non-quantification boundary — replay-verifiable from the published trust keys alone.
- 4 — You consume and price. Your board pack, your insurer, or your risk-quant platform ingests the signed feed and prices the exposure on its own model. KYE Protocol™ never prices — it proves the events.
See a worked, signed sample — counts, severity mix, a recomputable SHA-256, a signature check, and a feed export — in the Authority Risk Signal app.
KYE Protocol™ emits the signal; you price the exposure.
This is not financial-risk quantification or advice. KYE Protocol™ proves which authority events occurred — counts and signed references over a tenant and a window. It runs no likelihood×loss model, no exposure-range mathematics, and offers no investment or risk advice. Pricing the exposure is the job of your board, your insurer, or your external risk-quant platform. This is the exact boundary the KYE Model & Risk-Data Governance Pack™ already drew, held end-to-end: KYE Protocol™ supplies the proven signal, and cedes the quantification to the consumer.
Boards, insurers, and risk platforms — talk to us about an Authority Risk Signal feed.
The KYE Authority Risk Signal™ ships in Board, Insurer, and Risk-platform tiers; commercial distribution is value-based, qualification-gated, and disclosed under NDA to qualified applicants.