AI UX DAILY
Saturday, August 29, 2026
5 stories · curated for designers
The stories
Today in AI Products
 Nielsen Norman Group |
Aug 28 |
NN/g names the pattern: teams are shipping AI-generated designs faster than UX can evaluate them
NN/g's new article frames this as a 'custodial era' — UX teams spending more time auditing AI output than shaping it. They offer three responses: build shared judgment, speed up evaluation, and guide AI generation upstream.
| “ |
Run a two-hour team session this sprint mapping which AI-generated screens have shipped without a UX review, then prioritize one for an audit. — Designer's Takeaway |
| Nielsen Norman Group |
Aug 28 |
NN/g: AI can organize your empathy map data, but it cannot generate the research behind it
NN/g's article draws a hard line: AI is useful for clustering and summarizing research you already have, but feeding it no real user data produces a map that reflects AI training data, not your users.
| “ |
Before using AI to build an empathy map, audit your source material — if you cannot cite a real session or quote behind each quadrant, schedule the research first. — Designer's Takeaway |
 Figma |
Aug 28 |
Figma publishes findings from its 2024 study on how AI is changing design and product work
Figma kicked off a longitudinal study in 2024 to measure AI's effect on design and product development. The new post shares early findings and notes that AI is now helping run the study itself.
| “ |
Use Figma's published methodology as a template to set up a simple before/after measurement for your own team's AI-assisted design work this quarter. — Designer's Takeaway |
| Sidebar | Aug 27 |
A new visual archive catalogs how AI interfaces have looked and behaved across the past decade
Kyle Jeong's AI Interface Museum collects screenshots and interaction patterns from AI products over time, organized as a browsable visual history of how humans have learned to talk to machines.
| “ |
Browse the archive before your next AI feature kickoff to spot which interaction patterns have already been tried, failed, or iterated on by other products. — Designer's Takeaway |
| TLDR Design / UX4dot |
via TLDR Design |
Opinion: generative AI delivers answers before designers have explored the problem
A UX4dot essay argues that AI's speed removes the productive detours where creative judgment develops, and proposes designing for 'Learning Experience' alongside User Experience — especially for junior designers.
| “ |
Add one constraint to your team's next AI-assisted design session: generate nothing until you have written down at least three problem framings by hand. — Designer's Takeaway |
Today's Idea
AI speed creates a UX debt you have to budget for
Faster generation means more unreviewed output, synthetic research artifacts, and patterns borrowed from AI training data rather than your users. Build evaluation time into every sprint and treat AI output as a first draft, not a final one.
Stop shipping AI slop
Turn your design into Claude skills
Drop a screenshot. See which of the 38 patterns you are missing and take them away as Claude Code skills. Free, no signup for the first audit.