AI UX DAILY
Wednesday, September 2, 2026
4 stories · curated for designers
The stories
Today in AI Products
 open-design-system-bench |
via TLDR Design |
A new benchmark scores whether AI agents actually use your design system
open-design-system-bench graded 898 AI-generated UIs against four design systems. Top score: 63.4 out of 100. Some models ignored the system entirely; others invented component names that don't exist.
| “ |
Run your design system through this benchmark before assuming any coding agent will respect it, then audit the components agents hallucinate most often. — Designer's Takeaway |
 Figma |
Sep 1 |
Figma generative plugins now support animations, interactions, and publishing
Generative plugins and shaders built by prompting the Figma agent can now include animations and interactions, be published to the community, and expose their underlying code for inspection.
| “ |
Build a team generative plugin for your most repeated UI pattern this sprint, then publish it so the whole org shares the same AI-assisted starting point. — Designer's Takeaway |
 Google Pics |
Sep 1 |
Google launches Pics, a prompt-first image editor for Workspace
Google Pics is a new Workspace tool built on Gemini and the Nano Banana model. Users tap objects in an image and prompt to edit or generate them, replacing manual selection-based workflows.
| “ |
Map how tap-to-select plus prompt replaces traditional marquee selection in your own AI editing flows, and test whether users prefer it over traditional toolbar controls. — Designer's Takeaway |
| Android |
Sep 1 |
Android's latest update adds motion-sickness reduction and new accessibility controls
Google's Android update includes motion-sickness mitigation settings and expanded accessibility options, some powered by Gemini. Several features mirror existing iPhone capabilities.
| “ |
Review your app's animation-heavy screens against Android's new motion-sickness settings to confirm your transitions don't fight the OS-level accommodations users have turned on. — Designer's Takeaway |
Today's Idea
AI agents ignore your design system more than you think.
The benchmark data puts a number on a gap designers have been sensing: AI coding agents score below 64 out of 100 at following any real design system, and some hallucinate component names outright. Audit what your agents actually generate, then tighten token naming and documentation to close the gap.
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.