Google Labs announced updates to Stitch, its AI-powered design tool that converts natural language descriptions into user interface designs. The company said the platform now features an AI-native infinite canvas and new collaboration capabilities.

The updated Stitch allows users to create high-fidelity UI designs from text descriptions without starting with traditional wireframes. Users can input business objectives or describe desired user experiences to generate design concepts.

"AI can be a creativity multiplier, helping people explore many ideas quickly," said Josh Woodward, VP of Google Labs.

The platform introduces a redesigned interface with an infinite canvas that accommodates various input types including images, text, and code. A new design agent can analyze entire project histories and track multiple design directions simultaneously through an Agent manager feature.

Google Labs expanded the design system functionality with DESIGN.md, a markdown file format that enables users to extract design systems from URLs or transfer design rules between projects and external tools.

The update includes interactive prototyping capabilities that convert static designs into clickable prototypes. Users can connect screens and preview app flows through a "Play" button feature. The system can automatically generate subsequent screens based on user interactions.

New voice capabilities allow users to speak commands directly to the canvas for real-time design modifications and critiques. The platform supports requests such as generating multiple menu variations or displaying screens in different color schemes.

Stitch integrates with external development tools through a Model Context Protocol server and software development kit. Users can export designs to platforms including AI Studio and Antigravity for developer handoff.

The information comes from a Google Labs announcement about the Stitch platform updates.