Cognition · Software component

Confidence Calibrator

Software componentCognitionCognition & Memoryarc:ConfidenceCalibrator

A component that adjusts raw model confidence to match empirical accuracy for the scenario type, lowering presented confidence where the model is known to be overconfident.

Responsibility. Converts raw model confidence into calibrated confidence for presentation.

Also known as: Calibrated confidence display

Direct neighbourhood (hover for relationship types)

Relationships

receives data from dynamic

sends data to dynamic

Confidence Indicator Ch10.5

Design guidance

SHOULD adjust presented confidence downward to empirical accuracy rates for scenario types where the model tends toward overconfidence.

Classification

Patterns: Post-hoc confidence calibration by scenario type
Quality attributes: Interaction capability (ISO/IEC 25010)Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)
Risks mitigated: Overconfident models misleading usersConfidence score misinterpretation

Sources

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.