ChainThink report, April 23: According to public information, former OpenAI researcher Zain Shah and his team have launched an experimental prototype called Flipbook, a project that uses AI models to directly generate screen pixels, replacing traditional web technologies such as HTML and CSS.

Each "page" in Flipbook is an AI-generated image; users can click any region of the image to generate a new image for continuous interaction. The entire interface contains no HTML code, fixed links, or predefined buttons—text itself is rendered as pixel data within the image.

The video mode is based on LTX Studio, an open-source diffusion Transformer video generation model developed by Israeli company Lightricks, optimized to enable real-time WebSocket streaming at 1080p resolution and 24fps. The backend runs on Modal Labs' serverless GPU infrastructure.

Zain Shah previously worked at OpenAI on AI and robotics research, then served as a Creative Technology Expert at Samsung, and is also a YC S13 alumnus;

Team members include Eddie Jiao, former engineer at Humane and Slack, and Drew O'Carr, former Apple engineer.

Shah stated that Flipbook currently has limited functionality and is designed around visual interpretation, but this prototype demonstrates a potential future direction. As model accuracy and state management capabilities improve, it could scale to structured UI scenarios, including domains related to programming.