OpenAI, fresh off its impressive $6.6 billion funding round, has unveiled a new beta feature for ChatGPT called Canvas. This innovative collaboration interface aims to revolutionize how users interact with the powerful AI.
Karina Nguyen, the research lead for Canvas, described it as a “new interface for working with ChatGPT on writing and coding projects that go beyond simple chat.” Similar to Claude’s Artifacts window, Canvas provides users with a real-time view of ChatGPT’s output in a dedicated window, separate from the chat stream.
Canvas operates autonomously, automatically launching when ChatGPT identifies an opportunity for its assistance. Users can provide inline feedback on the generated content, highlighting specific lines or the entire text for revision. Direct editing is also possible, giving users more control over the final output.
The feature goes even further, enabling users to direct ChatGPT to research specific topics on the internet and integrate that new information into their projects.
For added convenience, Canvas introduces a shortcut menu filled with common tools. Users can suggest edits, adjust output length and reading level (from kindergarten to graduate-level), debug code, add emojis, and even employ a “final polish” feature that checks for grammar, clarity, and consistency.
Coding tasks benefit from their own dedicated shortcut menu, providing quick access to features like code review, log addition, comment insertion, bug fixes, and language translation. Canvas supports translation between JavaScript, TypeScript, Python, Java, C++, and PHP.
While Canvas is currently in beta, it is available exclusively to Plus and Teams subscribers. The release timeline for Enterprise and free-tier users remains unannounced.
