Google’s November 5, 2026 blog post describes a Gemini Canvas workflow for turning a photographed sketch into an app prototype. A Google Workspace video published October 27, 2025 had already shown the core sequence: upload a sketch, generate code, preview the result, and request changes.
From a sketch photo to a Canvas prototype
Google’s described workflow starts with a photo of a drawing. Upload it to Gemini, select Canvas, and ask Gemini to code the app. Canvas generates code and opens a preview; you can then ask for changes to colors, fonts, or features. Google’s blog post describes that sequence.
Use the preview to explore an app idea and refine its interface before treating the result as finished software.
Google Workspace showed the workflow in 2025
Google Workspace’s video, published October 27, 2025, shows a paper sketch being uploaded, code generated, and an app preview revised through conversational requests. Google’s November 5, 2026 blog post later described the same core sketch-to-prototype sequence. The video’s earlier publication establishes an earlier demonstration, not a broad rollout date.
Workspace preview and sharing depend on the data and account
Google Workspace’s May 20, 2025 update says a Canvas app that contains Workspace data cannot be previewed; an app without Workspace data can. The same update says Workspace business and education users cannot share content created in Canvas. They can open a link shared from a personal account, but cannot take further actions such as asking Gemini to refine the content. Google Workspace’s update sets out these conditions.
