OpenAI has launched a new feature for ChatGPT called “canvas,” aimed at improving user engagement with the AI model which creates a dedicated workspace alongside the standard chat window, allowing users to write and code more effectively.
Within the canvas, users can generate text or code, highlight specific sections for editing, and receive real-time suggestions from the model.
The canvas feature is currently in beta, available to ChatGPT Plus and Teams users starting Thursday, with plans to roll it out to Enterprise and Edu users next week.
As the consumer AI landscape evolves, many companies are moving towards editable workspaces to better utilize generative AI. ChatGPT’s canvas shares functionalities with competitor offerings like Anthropic’s Artifacts, introduced in June, and the popular coding tool, Cursor. This development reflects OpenAI’s efforts to enhance its services and grow its paid user base.
Traditional AI chatbots often struggle to handle large projects based on a single prompt. The canvas, however, allows users to modify the AI’s outputs without needing to rephrase their initial requests or start over.
“This is just a more natural interface for collaborating with ChatGPT,” said OpenAI product manager Daniel Levine during a demonstration with TechCrunch.
In the demo, Levine illustrated how to activate the canvas by selecting “GPT-4o with canvas” from the model picker. OpenAI indicated that canvas windows will automatically open when the system detects that a separate workspace would be useful, particularly for lengthy outputs or complex coding tasks. Users can also prompt the system by typing “use canvas” to initiate a project window.
Levine showcased the email-writing feature, where users can instruct ChatGPT to generate an email that appears in the canvas. The interface includes a slider to adjust the text length and options to request changes, such as making the tone friendlier or adding emojis. Users can also ask ChatGPT to rewrite the email in another language.
For coding tasks, the canvas provides specific functionalities as well. In the demonstration, Levine prompted ChatGPT to create an API web server in Python, which then appeared in the canvas. Users can click an “add comments” button to receive in-line documentation that explains the code in plain language. They can also highlight sections of code to request explanations or bug reviews. A new “review code” button will offer suggestions for edits to any code in the workspace, facilitating collaboration on code corrections.
OpenAI plans to make the canvas feature available to free users after the beta phase concludes, further enhancing accessibility for all ChatGPT users.
