Google recently announced a major feature upgrade to its conversational AI large model Gemini, officially launching a new ability to edit files directly within the file. Starting today, all Gemini users on both web and mobile can generate, preview, and edit a variety of mainstream file formats in real time and in full screen, breaking the previous limitation of only being able to export files and further improving the interactive fluidity of AI-assisted productivity.
According to details released by Google, this upgrade fully covers PDF, Google Docs, Google Sheets, Slides, Markdown files, TXT text, and code files. Among them, PDF preview and editing are supported on both web and mobile, while the remaining formats are currently mainly available on the web. In practice, users can directly make in-depth edits to generated file content without leaving the app. It not only supports full-screen editing and asking targeted questions about selected text, but also enables advanced features such as interactive table-to-chart conversion.
This feature upgrade marks that large models are accelerating their evolution from “content generation tools” to “immersive collaboration workbenches.” Previously, Google had already deeply integrated related capabilities into the Workspace ecosystem and sidebar, while competitor Anthropic had also previously launched similar document and presentation editing features in the Claude chat window.
As major mainstream AI vendors continue to strengthen the closed-loop experience of human-computer interaction, conversational office work is gradually becoming an important industry trend for deploying large models and is expected to further reshape users’ daily digital workflows.
