Agentic AI Reference Architecture

Experience · Software component

Explanation Expansion Controller

Software componentExperienceExperience & Human OversightVariation point (abstract)arc:ExplanationExpansionController

An abstract presentation-control component that decides when collapsed explanation layers are expanded to reveal deeper reasoning, uncertainty or data detail.

Responsibility. Determines when deeper explanation layers are revealed.

Also known as: Disclosure trigger

sends data tois specialized byis specialized byExplanation Presenter: sends data toExplanation PresenterProactive Explanation Expander: is specialized byProactive Explanation Ex…On-Demand Explanation Expander: is specialized byOn-Demand Explanation Ex…
Direct neighbourhood (hover for relationship types)

Variants

VariantWhen to choose
On-Demand Explanation ExpanderChoose when users can be expected to seek detail themselves (low time pressure, routine or low-stakes decisions); insufficient alone because users rarely seek additional information spontaneously.
Proactive Explanation ExpanderChoose when confidence may fall below thresholds, decisions may deviate from historical patterns, or high-stakes scenarios arise, so critical detail must surface without user initiative.

Relationships

sends data to dynamic

Classification

Patterns
Progressive disclosure
Quality attributes
Interaction capability (ISO/IEC 25010)Transparency and accountability (NIST AI RMF: accountable and transparent)
Risks mitigated
Cognitive overload

Sources

  1. 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.