Patterns
Explore trust AI interface design patterns.
Progress Steps
Collapsible thinking and tool traces
Citation Tooltips
Hover for source
Confidence Score
Probability UI
Knowledge Graph
Visualizing RAG
Citations
Footnotes
Privacy Filters
Masking PII
Source Browser
Side-by-side doc
Chain of Thought
Show reasoning
Confidence Indicators
Visual confidence levels for outputs
Data Ownership & Control
User control over AI data usage
Bias Detection
Flag potentially biased outputs
Fact-Checking Indicators
Real-time fact-checking status
Source Quality Scores
Rate source reliability
Audit Trail
Complete log of AI decisions and data usage
Transparency Report
Periodic reports on AI behavior/accuracy
Scope Disclosure
Plain-language agent permissions
Permission Drift Indicator
Surface accumulated agent permissions
Failure Disclosure
Honest signaling of AI limitations
Duration-Bound Consent
Permissions that expire by default
Granular Consent
Per-capability toggles, not bundled grants
Authentication Chains
Legible identity trails across agent actions
Responsibility Attribution
Trace which agent or human caused each action
Agent Identity
Stable name, version, and capabilities per agent
Revocation Affordances
One-click revoke beside the grant