Free account
Read this concept with a free account
Read this concept at no cost by signing in with Google.
no card needed
Core
Designing for Uncertainty: UX Patterns for AI That Is Sometimes Wrong
AI features are right most of the time and wrong some of the time, often without obvious signs. This guide covers the main UX patterns for uncertainty (sources, abstention, clarifying questions, drafts, targeted flags, consistency checks, fallbacks), why raw confidence scores rarely help, and how to match design intensity to error cost.
CONCEPT / DESIGNING FOR UNCERTAINTY
RELATED CONCEPTS
PRACTISE THIS CONCEPT
AI UX & Human-in-the-LoopOur AI feature is right about 90% of the time. How do you design the user experience around the 10%?→AI UX & Human-in-the-LoopHow would you design citations in an AI answer so users can actually verify what it says?→AI Product SenseA customer's lawyer filed a brief containing case citations our legal research assistant invented. What do you change in the product?→AI Product SenseHow would you design an AI-first spreadsheet?→AI UX & Human-in-the-LoopEvals say our AI ticket summaries are accurate, but support agents don't trust them and re-read every ticket anyway. What do you do?→AI UX & Human-in-the-LoopOur AI assistant can now take actions in users' accounts. Design how users review, understand and undo what it did.→