Google Search rolls out Gemini’s Canvas in AI Mode to all US users
Google has expanded access to Canvas in AI mode to all US users in English, after first launching the feature last year as part of its Google Labs experiments.
Canvas in AI mode is designed to help users organize and plan projects or conduct deeper research. The feature now supports the ability to compose documents or create custom tools within Google Search, the company said in a blog post.
Google previously suggested using Canvas for tasks such as building a study guide by uploading class notes and other resources; the feature can also perform other tasks, such as turning a research report into a web page, quiz, or audio summary, which has some overlap with Google’s research tool Notebook LM.
Users can describe an idea on Canvas and watch Canvas generate the code to turn that idea into a shareable app or game. The feature can also be used to refine creative writing concepts and get feedback on projects.
Canvas is already available in Gemini, where Google AI Pro and Google AI Ultra subscribers have access to the latest model, Gemini 3, and a larger context window of 1 million tokens for more complex projects.
More people will be exposed to Canvas now that it’s available to all US users through Google’s AI search known as AI Mode, including those who haven’t yet interacted with Gemini’s capabilities. That’s one of Google’s advantages in the AI race: Google Search’s reach gives it the power to get its products in front of billions of users.
To use Canvas, users in AI mode select the new Canvas option from the tool menu (+) and then describe what they want to create. This opens a Canvas side panel where users can bring together information from the web and Google’s Knowledge Graph. As users build a prototype or app, they can test its functionality, toggle to see the underlying code, and fine-tune how the app works by chatting with Gemini.
Canvas competes with similar tools from rivals like OpenAI and Anthropic. However, ChatGPT’s Canvas feature will be activated automatically based on demand, while Google’s Claude and Anthropic require more direct interaction. Both also allow users to get help with writing or turn ideas into projects.
