Reference
Glossary
Stable terminology used across ETHORITY surfaces. This glossary exists to prevent semantic drift across teams, regulators, engineers, and time.
Observed
A recorded factual statement about an AI system or artifact, stored append-only and published without interpretation.
Why this matters:
Trust collapses when facts and interpretation are mixed. Observed statements preserve reality without judgment.
Business relevance:
Procurement, regulators, and engineering teams can reference the same factual substrate without agreeing on policy.
Observed Index
ETHORITY’s public, append-only snapshot of observed entities and related metadata.
Why this matters:
A shared, reproducible source of reality reduces dispute about what exists and when it was recorded.
Business relevance:
Auditors and partners can cite existence without endorsing performance or safety claims.
ETN (ETHORITY Trust Name)
A stable identifier for an entity in the observed layer (e.g., etn:observed:system:<vendor>:<system>).
Why this matters:
Stable identifiers prevent ambiguity across versions, vendors, and lifecycle transitions.
Business relevance:
Contracts, audit trails, and governance artifacts can reference a persistent system identity.
Snapshot
A published, versioned export of observed records intended for citation and independent verification.
Why this matters:
Time-bound exports make historical claims auditable and reproducible.
Business relevance:
Organizations can reference specific historical states during audits or disputes.
Integrity
The property that a published artifact is authentic and unmodified, verifiable via public key cryptography.
Why this matters:
Without integrity, governance decisions cannot be trusted.
Business relevance:
Automated systems can safely ingest only verified artifacts into pipelines.
Verification
The act of validating integrity using a signature and public key. It does not imply safety, quality, or compliance.
Why this matters:
Integrity must be provable independently of trust in ETHORITY itself.
Business relevance:
Regulators and CI systems can confirm authenticity without relying on UI or marketing claims.
Governance Layer
Interpretation of observed facts under explicit rule systems (EU AI Act, ISO, internal policies). Versioned and contestable.
Why this matters:
Rules change. Governance must evolve without rewriting historical facts.
Business relevance:
Boards and compliance teams can understand exposure under multiple frameworks.
TrustCycle
A governance execution that evaluates observed facts under a specific framework version and produces structured artifacts.
Why this matters:
Governance must be reproducible, not anecdotal.
Business relevance:
Enables delta-based reporting across quarters or regulatory changes.
Lifecycle State
A deterministic stage within the Trust Lifecycle (Observed, Declared, Verified, Anchored, Monitored).
Why this matters:
Trust must be structured as a state machine to prevent implied permanence.
Business relevance:
Stakeholders can assess maturity without relying on marketing narratives.
Anchoring
Public timestamp preservation of a verified lifecycle state.
Why this matters:
Historical trust states must remain referenceable even after upgrades or downgrades.
Business relevance:
Critical for litigation defense and regulatory investigations.
Monitoring
Continuous enforcement of lifecycle continuity conditions.
Why this matters:
Verification is time-bound. Systems drift. Monitoring detects change.
Business relevance:
Prevents silent degradation of compliance posture.
Downgrade
An explicit reduction in lifecycle state due to violated conditions or lapsed continuity.
Why this matters:
Trust must degrade deterministically when assumptions fail.
Business relevance:
Ensures transparency when systems change or compliance lapses.