Designing for AI Response-Time Variance: Trust, Not Just Latency
Understand why variance in AI response times erodes user trust and how to design progress states that preserve it — covering anchoring bias, process vs. time anticipation, and leading indicators for trust erosion.
3 sections · 5 lessons
Course outline
Why AI Variance Breaks Trust
- The Anchoring Trap: Why Users Judge AI by Their Fastest Experience
- Two Common Fixes That Don't Work: Generic Spinners and Artificial Floors
Designing Progress States That Hold Up
- Time Anticipation vs. Process Anticipation: Naming the Work Instead of Promising a Duration
- A Practical Heuristic for When Time Estimates Are Still Safe
Measuring Trust Before It's Too Late
- Why Trust Erosion Has No Dashboard (and What to Track Instead)