Escalation Threshold Policy

Data artifactHuman OversightExperience & Human Oversightarc:EscalationThresholdPolicy

A declarative set of confidence and risk thresholds, scoring weights, and time-based default rules that encodes organisational risk tolerance for human escalation.

Responsibility. Encodes when agent actions require human review.

Also known as: Risk thresholds, Auto-approval threshold, Confidence thresholds, Autonomous approval threshold, Risk tier thresholds, Principle-derived escalation thresholds, Risk-tiered decision policy, Escalation rules, Exception criteria

Direct neighbourhood (hover for relationship types)

Relationships

configures structural

Approval Gateway Ch3.7 Ch9.2 +1
Confidence Gate Ch5.3 Ch10.2 +1
Dialog Rail Ch9.5 Ref9.01
Escalation Agent Ch10.1
Exception Gate Ch10.4
Oversight Gate abstract Ch1.1A Ch1.1B +1
Risk Gate Ch9.2 Ch10.1

is configured by structural

Constitution abstract Ch9.5

is written by dependency

receives data from dynamic

Oversight Governance Committee Ch9.2

Design guidance

SHOULD define time-boxed default actions only for non-critical decisions and communicate them clearly.
SHOULD set escalation thresholds balancing automation against risk tolerance: tighter thresholds reduce automation but increase safety.
SHOULD define approval thresholds and escalation paths that limit autonomy scope.
SHOULD reserve high-risk decisions (medical diagnoses, legal advice, large financial approvals, personnel decisions, sensitive data access) for humans.

Quantitative guidance

As stated by the sources; verify before use.

Example: auto-approve refunds under $100 and require approval above; auto-approve when confidence exceeds 90% and risk is low (Ch1.1A).
Examples: refunds under $100 autonomous with manager approval above; $1,000 autonomous threshold in refund workflow (Ch9.2).

Classification

Quality attributes: Maintainability (ISO/IEC 25010)Transparency and accountability (NIST AI RMF: accountable and transparent)

Sources

Ch1.1A: T. Nguyen, "Designing User Interfaces for Intuitive Human-Agent Interaction," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.1A. ISBN: 9798244538229.
Ch1.1B: T. Nguyen, "Human-in-the-Loop Patterns and Accessible Design," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 1.1B. ISBN: 9798244538229.
Ch3.7: T. Nguyen, "Tool Usage Auditing," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 3.7. ISBN: 9798244538229.
Ch5.3: T. Nguyen, "Self-Consistency Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.3. ISBN: 9798244538229.
Ch9.2: T. Nguyen, "Action Constraints and Permission Models," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 9.2. ISBN: 9798244538229.
Ch9.5: T. Nguyen, "Constitutional AI," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 9.5. ISBN: 9798244538229.
Ch10.1: T. Nguyen, "Conversational UI," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.1. ISBN: 9798244538229.
Ch10.2: T. Nguyen, "Proactive Agents," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.2. ISBN: 9798244538229.
Ch10.4: T. Nguyen, "Human-in-the-Loop," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.4. ISBN: 9798244538229.
Ch10.5: T. Nguyen, "Human-over-the-Loop," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.5. ISBN: 9798244538229.
Ref3.04: M. Chen, "What Is Agentic AI?," Oracle, Jun. 17, 2025. [Online]. Available: https://www.oracle.com/artificial-intelligence/agentic-ai/
Ref9.01: "AI Safety Frameworks for Agent Systems," unpublished reference note (01-AI-Safety-Frameworks.md), Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam supplementary materials, 2026. unpublished note