Why this page exists
The homepage provides the buyer path. This appendix preserves the deeper technical record.
This appendix preserves the deeper OntoGuard proof, semantic governance, ontology, evidence-retrieval, BM25 candidate-generation, and implementation-facing record needed by auditors, technical buyers, security reviewers, and strategic acquirers.
ArchitecturePlatform, how it works, state-transition governance, and ontology layer.Proof surfaceInvariants, packet logic, runtime seam, route proof, and controlled-boundary evidence.Technical depthBM25/evidence retrieval, semantic traces, governance paths, and packet surfaces.Markets + visualsRegulatory readiness, buyer value, workflow examples, roadmap, and diagrams.
BM25 and semantic evidence retrieval — public-safe implementation boundary
BM25 is used as a lexical candidate-retrieval surface within a scoped governance run. It helps identify potentially relevant evidence, clauses, and enterprise terms; it does not itself authorize release or prove legal applicability. Candidate relevance is evaluated alongside semantic scope, evidence sufficiency, authority, uncertainty, consequence, and route state.
The public site intentionally omits protected scoring formulas, private corpus configuration, routing internals, and proprietary arbitration logic. Missing or weak retrieval evidence must remain visible as a gap and may drive ESCALATE rather than fabricated certainty.
Deep Technical Index
Find the OntoGuard concept you need
Semantic Control Plane
3-layer Semantic Governance Stack
Proposed State Transition Builder
Execution Boundary Trace
No-Bind Receipt
Route Completeness
Decision Projection
Coherent Governance Loading
Decision-Relevant Uncertainty
Threshold Provenance
Decision Admissibility
Semantic Constraint Governance
Human Review Closure
Strict-Six Audit Credential
Batch DAP
Proof Harness
L3 Training Signal Export
ROI KPI Framework
Procurement / Security Readiness
Product Ladder
Partner / Licensing Path
Proof Harness
PDF / JSON exportBuyer-readable and machine-readable proof.
Schema-valid packetStructured artifact contract for diligence.
Confidence tier + trust scoreExplanatory signals, not release authority.
Cold / Heat / Mercury triadCoverage, anomaly pressure, and trace fidelity posture.
Hallucination taxonomyRisk classes and unsafe action language categories.
Proof harness statusWhich proof checks passed, warned, or remained pending.
Benchmark + policy statusBuyer-safe readiness and policy posture.
Evidence / citation gapsExplicit gaps instead of invented authority.
Abstention / human-review routeSAFE_TEMPLATE or HUMAN_REVIEW when release is not admissible.
Proof Harness: Export Valid Evidence or an Explicit Failure State
OntoGuard packages buyer-safe decision artifacts even when evidence is sparse, uncertainty is material, or the answer is withheld. The public proof surface emphasizes graceful degradation and explicit routing rather than silent blanks.
Platform detailsControl-plane licensing and technical package details.
Semantic Control Plane for Runtime Authorization of Proposed State Transitions
OntoGuard is the Ontology AI and Semantic Layer for enterprise AI: a runtime cognitive control plane, Reasoning Layer, and Cognition Layer for state-transition governance. Modern AI systems no longer just answer prompts; they call tools, write memory, update knowledge graphs, trigger workflows, hand work to other agents, and generate training signals that change future behavior.
From LLM outputs and training signals to tool calls, memory updates, ontology changes, policy mappings, and multi-agent handoffs (via integration), OntoGuard turns every proposed state change into a governed, auditable decision with evidence, routing, risk, uncertainty, arbitration transparency, audit hashes, and improvement signals.
OntoGuard governs the commit boundary. It asks whether a proposed AI state transition is authorized, traceable, evidence-backed, and safe to release before the change touches a customer, workflow, regulator, record, or downstream system.
The result is a Governed State Transition Record — buyer-safe proof showing decision, scope, evidence, and improvement signal.
The Decision Authorization Layer is the high-stakes release-control capability inside the broader Semantic Control Plane.
Regulatory compliance is not the ceiling. It is the first high-value use case for a broader enterprise control plane.
Proof invariantsRoute, packet, and governance invariants for technical diligence.
Explicit Governance Outcomes. Evidence-Preserving Export. No Silent Blank Artifacts.
Regulated buyers need governance that is resilient under ambiguity. OntoGuard is designed so uncertainty does not erase the audit trail.
No Silent Artifact Failure
OntoGuard may block, escalate, withhold, or refuse an operation. The export layer should still produce a valid packet or an explicit failure-state record rather than silently omitting evidence.
Always-Export Artifacts
Every governed run is expected to produce a buyer-readable PDF and complete JSON packet, even when the answer is withheld.
Evidence-Never-Blank
If evidence is sparse, OntoGuard records fallback reasons, provisional review anchors, uncertainty, and next human step instead of silently emitting an empty proof trail.
What the Runtime Control Plane Gives You
- Decision API: ALLOW, BLOCK, or ESCALATE with release status, routed_to, business effect, and reason codes.
- Governed State Transition Record: before_state, proposed_action, evidence_state, decision, audit credential, and improvement signal.
- Enterprise ontology grounding: object, relationship, rule, clause, and evidence mapping instead of opaque prompt-only scoring.
- Semantic Governance Triad: Cold, Heat, and Mercury signals for coverage, anomaly pressure, and trace fidelity.
- Buyer-safe telemetry: clean decision-relevant proof separated from internal repair flags, schema fences, and hardening markers.
- Business impact: pilot KPIs that map to review minutes, escalation rate, cycle time, false approvals, and incident prevention.
How it worksLayered semantic governance architecture and operating model.
How It Works — The 3-Layer Semantic Governance Stack
OntoGuard is not just ontology lookup and not just scoring. It uses ontology throughout the governance path to connect an LLM output to policy, evidence, scope, trace, human review, and learning feedback.
Executive Summary
OntoGuard uses a three-layer system to turn raw AI outputs into governed, auditable decisions. Layer 1 grounds outputs to real enterprise objects and rules. Layer 2 checks accuracy, risk, uncertainty, and compliance. Layer 3 turns every approved or corrected decision into a learning signal that improves future retrieval, policy, and agent behavior.
Result: fewer incidents, faster audits, clearer human review, and safer autonomy without changing model weights.
Full Technical Detail
The technical view below shows how L1 symbolic grounding, L2 semantic consensus, and L3 alignment feedback produce the Decision API, Governed State Transition Record, evidence pack, triad, audit credential, and improvement signal.
L1
Symbolic Grounding
Normalizes the governed prompt and response into clause hits, evidence IDs, retrieval IDs, checksums, regulation/domain scope, and a buyer-readable symbolic trace.
- Ontology-grounded concepts
- Clause and fallback hits
- Evidence pack and provenance
L2
Semantic Consensus
Compares compliance, accuracy, risk, uncertainty, hallucination status, and agent disagreement before deciding whether release is safe.
- Trust and uncertainty signals
- Cold / Heat / Mercury triad
- ALLOW, BLOCK, or ESCALATE routing
L3
Alignment Feedback
Turns governed outcomes into reusable improvement signals for retrieval, policy, alignment, training data curation, and future agent decision quality.
- Gold example candidates
- Human-review outcomes
- Closed-loop training signal export
State-transition governanceHow proposed AI movement becomes a governed decision.
From Output Governance to State-Transition Governance
Agentic systems don’t just answer — they change state. OntoGuard already governs LLM outputs and is extending the same rigorous control to tool calls, memory, and agent actions. The question now is: is this proposed transition authorized, traceable, and safe to commit?
Old AI Governance Unit
Future Governance Unit with OntoGuard
Model output
Governed decision / state transition
Prompt / response pair
Before → Proposed → After state
Hallucination check
Uncertainty-aware authorization
Logging
Audit credential + improvement signal
Human review
Routed decision workflow
Fine-tuning data
Governed training signal
Agent action
Tool-call / memory / ontology authorization
Static governance weakens when AI systems use tools, maintain memory, and update knowledge. OntoGuard makes each proposed transition explicit before it becomes business state.
Governable eventsCurrent output governance and integration-path event types.
What We Govern: The 7 State Transitions
Seven event types can become ALLOW, BLOCK, or ESCALATE decisions with evidence, scope, audit hashes, and improvement signals.
Proposed ChangeOutput, action, write, update, signal, or handoff
→
OntoGuard CheckpointBM25 + semantic evidence, ontology grounding, triad, arbitration
→
DecisionALLOW, BLOCK, or ESCALATE before commit
→
Proof + LearningPacket, audit credential, human route, L3 signal
1. LLM / Model OutputAvailable Today
Risk if ungoverned: unsupported answers, misleading recommendations, unsafe release, or false confidence.
OntoGuard produces: Decision API, evidence pack, triad, hallucination status, audit hashes, and review route.
2, 3, 4, 5 & 7. Agentic Extensions Integration Path
Current Status: The same Decision API and JSON contract used for LLM outputs is designed to support these use cases.
How it works: Agent frameworks can call OntoGuard’s Decision API before executing tool calls, memory writes, ontology changes, policy updates, or agent handoffs.
Expansion Roadmap: Customer-specific production route integration can attach real tool-call, memory-write, ontology-change, policy-mapping, and handoff routes before enforcement claims are made.
6. Training / Alignment Signal ExportAvailable Today
Risk if ungoverned: bad examples, unsafe corrections, or unreviewed behavior changes feeding future systems.
OntoGuard produces: gold-example candidates only when governed outcomes are approved, corrected, and traceable.
Available today: governed LLM output release, Decision Authorization Packet generation, controlled boundary proof, and runtime-safe L3 training signal export. Agentic extensions require customer-specific route integration before production enforcement claims.
Ontology and semantic layerHow ontology grounds objects, evidence, scope, and authorization.
Ontology AI as the Semantic Layer for Enterprise AI
Enterprise systems already run on objects, relationships, and rules. OntoGuard turns that structure into Ontology AI: the Semantic Layer that lets AI reason over enterprise reality before a proposed state transition becomes a decision.
Enterprise Objects
- Customer / Account / Claim
- Policy / Contract / Case
- Vendor / Asset / Location
- Obligation / Exception / Evidence
Relationships & Rules
- Customer → owns → Account
- Claim → references → Policy
- Contract → restricts → Data Use
- Case → requires → Review Outcome
Semantic Layer
Maps prompts, outputs, policy scope, BM25 evidence, clause hits, enterprise objects, and symbolic traces into one governed representation.
Reasoning Layer
Turns semantic evidence into allowed, blocked, or escalated decisions with uncertainty, risk, arbitration, and human-review routing.
Cognition Layer
Feeds approved or corrected outcomes back into L3 training signals, policy improvements, retrieval improvements, and future governance quality.
Together, these layers form a System of Intelligence: not just model output, but enterprise-aware reasoning, authorization, evidence, and learning.
Enterprise Reality
Objects, relationships, policies, obligations, evidence, and workflow state
+
AI Proposal
Output, tool call, memory write, ontology update, policy mapping, training signal, or handoff
→
OntoGuard Semantic Layer
Grounds, reasons, routes, audits, and converts decisions into governed learning signals
Result: “We can prove it — and here is the semantic path from enterprise state to governed decision.”
Governed state-transition recordBefore/proposed/decision/audit/improvement record structure.
Governed State Transition Record
The buyer-readable PDF is one view of the proof. The Governed State Transition Record is the structured JSON contract underneath it. In the current v1.2.0 lineage, the record already carries the fields needed for audit, routing, repair, and improvement loops.
before_state
Prompt, workflow, scope, and current object context proposed_action
Output, tool call, memory write, mapping, or handoff evidence_state
Clause hits, retrieval IDs, checksums, symbolic trace triad_with_meta
Cold, Heat, Mercury, provenance, and penalties audit_credential
Decision, evidence, report, and manifest hashes improvement_signal
Review outcome and L3 gold-example candidate
Prompt, workflow, scope, and current object context proposed_action
Output, tool call, memory write, mapping, or handoff evidence_state
Clause hits, retrieval IDs, checksums, symbolic trace triad_with_meta
Cold, Heat, Mercury, provenance, and penalties audit_credential
Decision, evidence, report, and manifest hashes improvement_signal
Review outcome and L3 gold-example candidate
L3 training-signal exportRuntime-safe learning signals from governed outcomes.
L3 Training Signal Export — From Decision to Gold Example Candidate
The same governance packet that protects a workflow can also create clean, reviewable signals for better retrieval, policies, evaluation sets, and future agents.
AttemptModel, agent, or tool call proposes an output.
→
OntoGuard DecisionALLOW, BLOCK, or ESCALATE is computed before release.
→
Evidence + ReasonsTrace, clauses, uncertainty, risk, and reason codes are exported.
→
Gold ExamplesApproved review outcomes become candidate training and policy examples.
→
Better Decision MakingFuture retrieval, policies, agents, and evals improve from governed signals.
Financial-services exampleBuyer-safe example of governed enterprise output and release withholding.
What Real Enterprise Output Looks Like
This is a buyer-safe financial services pilot output based on a governed enterprise workflow. OntoGuard mapped financial-services evidence, detected a remaining SEC citation-linkage gap, failed the strict autonomous-release benchmark gate, and withheld release pending human review — exactly the commercially valuable control-plane behavior regulated buyers need.
Financial Services Pilot
Buyer-Safe Telemetry
SEC Citation-Linkage Gap Detected
Governed State Transition Record v1.2.0
Strict Release Gate Failed
DecisionESCALATE
Release ControlWITHHELD PENDING REVIEW
Primary ScopeSEC / SOX / GLBA / FINRA
CoverageNo coverage gaps
Citation GapSEC pending citation linkage
Trust81.41%
TriadCold 100.00% | Heat 6.00% | Mercury 96.67%
BenchmarkComparable corpus available; strict release gate did not pass
ROIHard-dollar ROI not calculated — buyer baseline required
Artifact Depth18-page audit PDF + 6-page sellable-lite packet + schema-rich JSON
ROI Status
Hard-dollar ROI not calculated — buyer baseline required.
To calculate ROI, provide:
- monthly_case_volume
- baseline_review_minutes_per_case
- ontoguard_review_minutes_per_case
- fully_loaded_reviewer_hourly_rate
- baseline_false_approval_rate
- ontoguard_false_approval_rate
- average_loss_per_false_approval
- implementation_cost_usd
- assumptions_source
ROI Results
Monthly review savings—
Annual loss prevention—
Total annual value—
Payback months—
Assumptions source—
Why this example matters: OntoGuard did not approve the AI output just because the trust score was 81.41%. It detected a citation-linkage gap, failed the strict autonomous-release benchmark gate, and escalated the decision for human review.
That is the product: not a score, but a release decision with evidence, reasons, and review routing.
That is the product: not a score, but a release decision with evidence, reasons, and review routing.
Request Audit PDF + Sellable-Lite Packet
Request Schema-Rich JSON
Download Financial Services Sample Packet
Similar artifacts have been delivered across KYC, claims, prior authorization, and benefits determination workflows.
The financial-services sample packet is available for open website visitors. Additional workflow-specific packets can be shared in sanitized or NDA-gated form.
Control-plane expansionBroader platform packaging and executive proof framing.
Current Capability vs. Expansion Roadmap
Core capabilities—including governed LLM-output evaluation, Decision Authorization, proof-packet generation, and review-gated improvement-signal export—are implemented in the current controlled product. Production route enforcement and production L5 evidence require customer-specific integration.
Available in the Current Controlled Product
LLM Output Governance + L3 Training Signals
Full runtime governance layer, buyer-safe telemetry, audit hashes, human-review routing, and closed-loop training signal export.
Integration Path
Agentic State Governance
Tool call authorization, memory write control, ontology change proposals, policy mapping updates, and multi-agent handoff governance.
Available now via API integration. Customer-specific production route integration is required before production route-completeness claims are made.
Enterprise workflow examplesWhere high-stakes AI movement can create risk or value.
💼 Enterprise Workflows We Govern Today
OntoGuard governs proposed AI state transitions in workflows where mistakes become cost, regulatory exposure, operational delay, or customer harm.
Financial Services
KYC and onboarding approvals, claims and dispute resolution, underwriting and credit decisions, advisor copilots, SEC / FINRA reporting support, fraud operations, and customer communications.
Healthcare
Prior authorization, eligibility checks, clinical documentation, coding support, patient triage, care navigation, medical chatbot review, and safety escalation.
Public Sector & Enterprise Ops
Benefits and eligibility determinations, casework, investigations, procurement approvals, customer operations, refunds, exceptions, IT automation, and change approvals.
Human Review Options:
APPROVE
REWRITE
BLOCK
Technical depthSemantic governance internals exposed at buyer-safe depth.
How OntoGuard Actually Works
OntoGuard exports the Semantic Layer and Reasoning Layer: ontology scope → BM25/semantic evidence → clause coverage → arbitration → audit hashes → human routing → L3 signal.
1 · Governed State Transition Record
v1.2.0-style structure carries before state, proposed action, evidence state, decision, triad metadata, audit credential, and improvement signal.
2 · BM25 + Semantic Candidate Bag
Current-run prompt, response, and scope anchors feed lexical BM25, semantic retrieval, clause normalization, checksums, and no-silent-drop telemetry.
3 · Primary-Scope Gap Penalty
Cold index is penalized when decision-driving regulations such as FINRA, GLBA, SEC, or SOX have zero coverage, even when supplemental regulations are mapped.
4 · Buyer-Safe vs. Internal Lanes
Buyers see clean decision evidence; internal lanes preserve repair flags, monotonic markers, schema fences, diagnostics, and repair provenance.
5 · Transparent Agent Consensus
Compliance, Accuracy, Risk, and Feedback agents expose votes, disagreement, consensus, and native arbitration computation.
6 · Portable Proof
Financial pilot artifacts expose 87-key JSON depth, hashes, coverage gaps, evidence anchors, schema-constrained fields, and L3 improvement readiness.
Differentiation: OntoGuard does not bury meaning in notebooks or pipeline glue. It exports the semantic chain as a buyer artifact.
Regulatory readinessIndustry/regulatory coverage examples and limitations.
Regulatory Strength as a Secondary Superpower
OntoGuard is broader than compliance tooling, but regulatory readiness remains a powerful entry point. The platform can expose whether the regulations that matter for a workflow are actually covered — not merely whether some evidence was found.
Financial Services
FINRA, GLBA, SEC, SOX, credit, advisory, customer communications, trading support, regulated reporting, and primary-scope coverage gaps.
Healthcare and Life Sciences
HIPAA, PHI handling, clinical support workflows, prior authorization, medical chatbot review, patient communication, and safety escalation.
Privacy and AI Governance
GDPR, EU AI Act readiness, privacy obligations, risk disclosures, human-review routing, and evidence-backed decision records.
Public Sector and Operations
Eligibility, procurement, exception handling, casework, audit trails, reviewer accountability, and policy-change governance.
We do not just check boxes. We surface coverage gaps in the laws, policies, and enterprise rules that actually drive the workflow.
EU AI Act timing — as of July 10, 2026: The Act becomes broadly applicable on August 2, 2026, subject to phased exceptions. Following the 2026 political agreement on simplification, specified high-risk-system rules are scheduled to apply from December 2, 2027, while requirements for systems embedded in regulated products are scheduled for August 2, 2028. Applicability must be confirmed for each system, role, and workflow by authorized counsel.
Remediation roadmapHuman next steps tied to L1/L2/L3 improvement loops.
Governance Feature Remediation Expansion Roadmap
Every packet can expose a buyer-readable next human step tied to the exact L1, L2, or L3 capability that needs remediation.
L1 Evidence Remediation: add or verify missing evidence, citation, retrieval ID, clause mapping, or ontology scope.
L2 Consensus Remediation: resolve trust, risk, uncertainty, hallucination, or semantic disagreement before release.
L3 Feedback Remediation: convert reviewer outcome into policy, retrieval, evaluation, or gold-example improvement signal.
Buyer valueCore buyer questions answered by the proof packet.
Why This Matters to the Buyer
Buyers do not need another opaque score. They need a defensible answer to four questions: should this AI change commit, why, what evidence proves it, and what happens next?
Commit?ALLOW, BLOCK, or ESCALATE
Why?Reason codes, triad, risk, uncertainty
Proof?Evidence IDs, retrieval IDs, hashes, trace
Next?Human route or L3 improvement signal
Visual appendixDiagrams, visual summary, and NDA-only material.
How It Works
OntoGuard uses a three-layer Semantic Governance Stack: L1 Symbolic Grounding, L2 Semantic Consensus, and L3 Alignment Feedback. The result is not just a score — it is a governed release decision with evidence, traceability, and reusable learning signals.
1. From Semantic Retrieval to Decision Authorization
2. Ontology-Grounded Evidence Reuse
3. The AI Trust Pipeline
4. Executive Summary
📄 NDA Access Request
Not Just Ontology — Runtime Authorization + Learning Loop
- 🧠 Decision API: ALLOW, BLOCK, or ESCALATE with reasons, confidence, trace ID, evidence, and release status
- 🔁 Training Signal Export: governed decisions become clean examples for better future agents
- 📊 Semantic Governance Triad: Cold, Heat, and Mercury explain grounding, volatility, and trace fidelity
- 🛡️ Audit-Ready Proof: governed response, evidence, hashes, hallucination status, uncertainty, and human-review task
OntoGuard is protected by U.S. Patent Application 19/444,521 — Track I prioritized examination granted May 2026; the technology produces runtime authorization decisions backed by evidence, routing, auditability, and improvement signals.
Public sample packet available now. Full technical details, claims mapping, and private demo assets available under NDA.