Gradual Generation of User Interfaces as a Design Method for Malleable Software
Bryan Min, Peiling Jiang, Zhicheng Huang, Haijun Xia
TL;DR
Gradual Generation of UIs introduces a design method that structures GenUI customizations into sequential intermediate layers that AI gradually loads during interface generation. By exposing distinct customization features at each layer and allowing users to wind back to earlier stages, the approach aims to balance expressiveness with discoverability and maintain visual simplicity. The paper details the four design tasks, illustrates with three prototype applications (video feed, scheduling calendar, and course management), and discusses the evolving role of UI/UX designers, end-user contributions, and cross-platform scalability. The work provides practical guidelines for integrating malleable customization features into GenUI systems and identifies future research to validate and extend the method in real-world settings.
Abstract
AI is growing increasingly capable of automatically generating user interfaces (GenUI) from user prompts. However, designing GenUI applications that enable users to discover diverse customizations while preserving GenUI's expressiveness remains challenging. Current design methods -- presenting prompt boxes and leveraging context -- lack affordances for customization discovery, while traditional menu-based approaches become overly complex given GenUI's vast customization space. We propose Gradually Generating User Interfaces -- a design method that structures customizations into intermediate UI layers that AI gradually loads during interface generation. These intermediate stages expose different customization features along specific dimensions, making them discoverable to users. Users can wind back the generation process to access customizations. We demonstrate this approach through three prototype websites, showing how designers can support GenUI's expanded customization capabilities while maintaining visual simplicity and discoverability. Our work offers a practical method for integrating customization features into GenUI applications, contributing an approach to designing malleable software.
