Research & Innovation
LLM-Generated Interfaces Pass the Brief and Repeat Each Other, With Creative Range at 0.581 Against 0.902 for Humans
Design Creativity Bench measures three things about LLM-generated UI: how distinct different models' answers to the same prompt are, how much a single model's output changes across different product domains, and how much of a brief it satisfies. Models nearly max out the third and fall far short on the first two.