Components Transparency & Disclosure Uncertainty Indicators
TD04 · Transparency

Uncertainty Indicators

Purpose Makes AI confidence visible and meaningful at the point of output, distinguishing between outputs suitable for automation and those requiring human judgment.
Use when Any AI output that drives decisions, recommendations, or automated actions downstream.
Do not use On every output indiscriminately — uncertainty UI should be triggered by threshold logic, not shown universally.
Failure mode prevented Users treating low-confidence AI outputs as authoritative because uncertainty is not surfaced. Also prevented: over-engineered confidence UIs that confuse users with unexplained numbers rather than actionable guidance.
Example Dark Academia
Dark Academia
The Atelier
Neon Tide
The Archive
Candlelight
The Conservatory
Output Confidence 94%

High confidence. Pattern matches 94% of analogous historical cases. Eligible for supervised automation within defined scope.

confidence-low — below threshold, blocks automation, prompts human reviewconfidence-high — above threshold, eligible for supervised automation
Design rule

Confidence must enable judgment, not just acknowledge doubt. Low confidence must block or flag automation — not merely appear in fine print.

Override these CSS custom properties to apply your own visual treatment without touching the component's behavioral contracts.

Shared semantic tokens

Token Controls
--rad-surface-card Card background
--rad-border-subtle Default card border
--rad-color-text-meta Label text
--rad-radius-card Card border radius

Component-scoped tokens

Token Controls
--rad-confidence-low-color Low score value, low fill, note strong text
--rad-confidence-low-border Low variant border
--rad-confidence-high-color High score value, high fill
--rad-confidence-high-border High variant border
--rad-confidence-note-color Note body text
--rad-confidence-track-bg Progress track background

"*Calibrated uncertainty. You know the feeling.*"

← Previous Bias Check Prompts Next → Confidence Threshold

Intellectual Property

RAD is the original work of Jackie Curry. All rights reserved. No portion may be reproduced, adapted, or incorporated into any product or system without express written permission.

Permitted: citation in academic or editorial contexts with full attribution.

Licensing

© 2025 Jackie Curry. All rights reserved. Publication date: 2025.

For licensing inquiries, connect on LinkedIn →