ThreatKrusher Emblem
ThreatKrusher Assurance Center
🔍 ⌘K
Assurance Center > Restricted Companion Summaries > AI Management System (AIMS) Summary
Search Results
Public Summary / Restricted Companion Ref: ETP-AC-STD-019 Ver: 1.0.0 Owner: AI Systems Architecture & GRC Lead
⚡ Direct Answer Summary (AI Overview / Key Takeaway)
Executive summary of the eTrepid AI Management System conforming to NIST AI RMF 1.0 and ISO/IEC 42001.

eTrepid AI Management System & Regulatory Alignment Standard

Governing Standard: NIST AI Risk Management Framework (AI RMF 1.0) / ISO/IEC 42001:2023 Alignment

1. Scope & Framework Harmony

This standard specifies eTrepid's organizational AI Management System (AIMS), harmonizing the NIST AI Risk Management Framework (AI RMF 1.0) core functions (Govern, Map, Measure, Manage) with the international requirements of ISO/IEC 42001:2023.

The standard applies to all machine learning pipelines, generative AI assistants, and automated analytical models utilized in delivery of the ThreatKrusher ecosystem.

2. NIST AI RMF Core Function Implementations

flowchart TD

G["GOVERN: AI Ethics Charter, Risk Tolerance, Human Escalation Rules"] --> M["MAP: Context Identification, Data Flows, Impact Categorization"]

M --> ME["MEASURE: Hallucination Audits, Output Verification, Security Testing"]

ME --> MA["MANAGE: Ongoing Monitoring, Incident Tracking, Continual Improvement"]

2.1 GOVERN

  • Establishes executive accountability for AI deployment via the AI Ethics Review Committee.
  • Enforces strict organizational policies prohibiting public retraining on client data.

    • 2.2 MAP

  • Catalogs all AI system components, grounding data sources, and intended operational contexts.
  • Categorizes AI applications by risk tier, separating non-critical administrative tools from high-impact SOC triage assistance.

    • 2.3 MEASURE

  • Validates model grounding using retrieval benchmarks and human-in-the-loop validation accuracy scores.
  • Conducts automated penetration testing and prompt injection fuzzing against AI interfaces.

    • 2.4 MANAGE

  • Implements automated fail-closed containment when model uncertainty exceeds threshold limits.
  • Tracks performance drift and model updates through version-controlled Model Bills of Materials (MBOMs).

    • 3. Restricted Assurance Companion

    Detailed risk impact assessments, individual model evaluations, and proprietary prompt templates are documented in internal AIMS operational manuals (ETP-AIMS-OPS-2026) accessible under restricted assurance review.

    🔒 Restricted Technical Companion Available

    This document is an executive summary of a restricted operational standard. Qualified defense contractors and certified C3PAO assessment teams subject to CMMC Level 2 or NIST SP 800-171 audits may request unredacted inspection under our mutual non-disclosure agreement.

    ← Previous (Restricted Companion Summaries)Corporate SSP Executive SummaryNext (Restricted Companion Summaries) →PCI-DSS Scoping Guide