Gemini adds in-chat document editing and PDF tools

Direct editing in the chat

Google has updated Gemini to include smarter document management capabilities that allow users to open or tweak supported files without leaving the chat window. Previously, users primarily used the AI to generate new assets from prompts or attachments, but they often had to move to separate apps to make changes. This new integration allows for a more fluid transition between generating data and refining the final output.

The update specifically covers Google Workspace formats like Docs, Sheets, and Slides, as well as raw text and code blocks. On the web version of the tool, users can preview and modify these files directly in the conversation thread. This shift aims to boost productivity by keeping the creative process contained in a single interface rather than jumping between different software tabs.

While the feature is a modest upgrade, it allows for more sophisticated interactions with content. For example, instead of just asking the AI to write a new document, a user can now pull up an existing one and ask for specific revisions. This makes the AI feel less like a simple generator and more like a collaborative editor for existing projects.

New PDF editing features

One of the most practical additions to the update is a built-in PDF editor available on both mobile and web builds. This tool allows users to perform basic text markup and request revisions on PDF files directly within the Gemini interface. For those who do not want to pay for a dedicated subscription to software like Adobe Acrobat, this provides a free way to handle basic document adjustments.

The PDF functionality includes a feature where users can highlight specific areas of a document to request changes. This is described as a significant improvement over typing out long, detailed instructions in a prompt box because the AI can detect the exact selection. Users can also tap and select text in the PDF viewer to ask the AI to generate equivalent charts or graphs based on that specific data.

However, there is a significant limitation regarding where these PDFs come from. The built-in editor currently only works for native files that were generated inside the Gemini chat or within the Google Workspace ecosystem. It does not support arbitrary PDFs created in outside software, such as scanned images or documents exported from third-party platforms.

Platform performance differences

The experience of using these new tools varies depending on whether a user is on a desktop or a mobile device. On the web version, the desktop browser offers a faster experience for opening files and highlighting specific sections for the AI to tweak. The web interface is generally more stable and handles the rendering of documents with fewer interruptions.

In contrast, the mobile experience can be hit-or-miss for some users. Reports indicate that when previewing PDFs on mobile devices, the files can sometimes load very slowly or force the user to close and reopen the viewer repeatedly. There are also instances where the app may hang on mobile, making the desktop version the more reliable choice for heavy document work.

Despite these performance issues, the cross-device syncing remains fluid for those who need to switch between devices. A user can start a task on a phone and jump to a laptop to continue accessing the same files within the chat. This portability is a key benefit for users who need to make quick edits while away from their primary computers.

Limitations and bottlenecks

While the new features are useful, Gemini is not yet ready to replace a dedicated office suite like Google Docs or Sheets. The tool currently suffers from janky rendering and sluggish processing when trying to function as a standalone document editor. These bottlenecks create friction for users who need to perform complex formatting or high-speed editing on large files.

There are also specific quirks regarding how different file types are handled. For instance, while Docs and Sheets are manageable in the chat, attempting to export text files or spreadsheets sometimes sends the user to an external link in Drive. To get around this, some users find it necessary to convert assets into other formats to keep them editable within the conversation.

The current state of the tool is best suited for on-the-go and casual editing rather than professional-grade document production. If Google can iron out the rough edges of the rendering and improve mobile stability, the tool could become a more powerful utility. For now, it serves as a reliable companion for quick adjustments to Google-native content.

Future possibilities

The current limitations of the tool highlight several areas where Google could expand the functionality in the future. A major desired improvement would be the ability to manage and edit imported external files. Currently, the system is mostly restricted to files born within the Google ecosystem, which limits its utility for many professional workflows.

Expanding support for external PDFs and third-party files would make the tool a much more versatile assistant for people who use a variety of different software. If the AI can eventually handle these external assets with the same ease as native Google files, it would significantly broaden its appeal to a wider range of users who do not live exclusively in Google Workspace.

For now, the focus remains on refining the existing experience for Google users. The goal is to provide a seamless way to move from a prompt to a finished product. As the technology matures, the line between chatting with an AI and editing a document may continue to blur, making the AI a more central part of the creation process.