Durable patterns in AI product design: suggested questions, citations, progressive disclosure, spatial context¶
Insight: Wroblewski identifies three persistent challenges across AI product evolution — capability awareness ("what can I do here?"), context awareness ("what is the system looking at?"), and walls of text (too much output to process) — along with six durable design patterns that address them: suggested questions, citations, multimedia responses, progressive disclosure, spatial context, and AI-driven onboarding.
Detail: Suggested questions drive "a huge percentage of all interactions" as entry points. Citations that allow hovering over sources build trust. Progressive disclosure collapses tool calls into expandable single-line entries. Spatial context lets users highlight objects or drag reference materials to guide model attention. AI-driven onboarding demonstrates capabilities through action (e.g., automatically generating charts when spreadsheets are uploaded) rather than tutorials. These patterns are grounded in products Wroblewski has built: Ask LukeW, Bench, Intent, ChatDB, and Rev.