AutoDrawAutoDraw 

work

Google's AutoDraw was a viral hit based on the provocation: "What if we could make a drawing tool that guesses what you are drawing?" Launched in April 2017, it became an immediate success, reaching millions of users and being named Product Hunt’s #1 Launch of the Year.

The project leveraged a custom-trained machine-learning classification model originally developed for Quick, Draw! and based on Google's handwriting recognition technology. By re-training these models to recognize hundreds of different objects, we were able to create a bridge between a user's rough sketch and professional, pre-drawn artworks.

From Experiment to Product

What started as an Experiments with Google project has since seen its core functionality integrated as product features across the Google ecosystem, including Google Jamboard, Docs, Slides, and Sheets. This transition from a provocative experiment to a utility used by millions highlights the project's impact on how we think about AI-assisted creativity, even in a pre-generative AI world.

Designing AI Interfaces

AutoDraw marked a key moment in my professional development, initiating a thread of exploration into the "interface for AI tools." This work paved the way for future projects like Teachable Machine and Creatability, where the focus remained on making complex technology feel simple, playful, and human. Today, this lineage continues in my work with multimodal LLMs and the Gemini Live Web Console, as I continue to invent future UIs for creative expression.

Awards & Recognition

• Product Hunt: #1 Launch of 2017
• TASCHEN: Featured in the book - The History of Web Design (read about it on It's Nice That)

Supporting Materials

• AutoDraw.com
• A.I. Experiments: AutoDraw
• Haptic-Data: AutoDraw

Team

• Alexander Chen
• Kyle Phillips (Lead Creative Technologist)
• Built by Google Creative Lab in collaboration with talented artists.