Gemini Notebook Connected App Lets Users Manage Notes in Chat

What happened

Google has spent most of 2026 tying its chatbot and its research notebook together, and the latest step moves notebooks into the chat itself. Gemini Notebook, the product formerly called NotebookLM, is now available as a Connected App inside the Gemini app. Users can create, query and manage notebooks by typing plain-language requests instead of switching to a separate panel or site.13 Typing "@Gemini Notebook" in a chat lets a user start a new notebook, add sources, save key notes or pull information from notebooks they already have. Gemini can also decide on its own to call the connector when a prompt seems to need it.13

Google's own examples show how broad the controls are. A user can create, edit or delete a notebook, add a copy of the current chat to an existing notebook, list every source in a notebook, or ask what their notebooks say about a topic.13 The connector is on by default for users worldwide except in the European Economic Area, where it starts switched off. It can be toggled under Settings > Connected Apps, and it covers Workspace customers, Workspace Individual subscribers and personal Google accounts.13 Some outlets dated the announcement October 24, a day after Google's own Workspace post, but they agree with Google on what the feature does and on the EEA exception.16

This is a small interface change on its face. In context, it finishes a merger that has been happening in stages.

A ten-month merger, one step at a time

The integration started modestly. In December 2025, Google began letting Gemini users attach a NotebookLM notebook as a source through the prompt box's attachment menu, with paid users getting access gradually.8 Mobile followed in January 2026, and Workspace Enterprise and Education accounts were added in early February.4 Google's Workspace announcement noted that an attached notebook could feed other Gemini tools, including Canvas, Veo, Guided Learning and Deep Research. Admins control the feature through the NotebookLM setting, not the Gemini app setting.3

The bigger change came on April 8, when Google added "notebooks" as a feature built into Gemini. The company described them as personal knowledge bases that hold chats, files and custom instructions, and that sync both ways with NotebookLM.1 Tech coverage agreed on how it works. A new Notebooks section appeared in Gemini's side panel, every chat got an "Add to notebook" option, and Gemini conversations showed up as sources inside NotebookLM.10 Android Authority reported that its app teardowns had tracked the feature under the internal name "Projects" before Google settled on "Notebooks."5 Paid subscribers on the web came first. Mobile and free users followed on April 30, with source limits of 50, 100, 300 and 600 per notebook for the Free, Plus, Pro and Ultra tiers.18

The naming then followed the product. On July 16, Google renamed NotebookLM to Gemini Notebook. The update also gave each notebook a "secure cloud computer" that can write and run code on its sources, and Google said the standalone product has more than 30 million users and over 600,000 organizations. Notebooks reached AI Mode in Google Search on August 19, in English and outside the EEA.15 The new Connected App means notebooks now live in Google's chatbot, its search engine and their own standalone app.

Why it matters for AI productivity tools

The main point is about context, not convenience. General chatbots have long had a memory problem: each new conversation starts empty, and users paste in the same background material over and over. Notebooks address this by keeping a curated set of sources, instructions and earlier chats that Gemini draws on alongside web search.1 One analysis described this as a project memory system and compared it with the Projects features in ChatGPT and Claude, arguing that Google's version stands out for its multimedia outputs and multi-device sync.7

The Connected App matters because it makes that memory easier to reach. Before, using a notebook meant going to it on purpose: open the side panel, pick the notebook, then work inside it. Now a user can say "add this chat to my Bio 101 notebook" mid-conversation, and Gemini may reach for notebook context without being asked.13 Google is pitching notebooks as background infrastructure under the chat window. One NewsBytes summary put it as notebooks no longer being confined to a side panel.16

This is a strong position in the market for AI research copilots and writing assistants. Google's pitch pairs grounded material with generation. A student can turn class notes into a Video Overview in the notebook app, then ask Gemini the next day to draft an essay outline from the same material.1 Early adopters describe the same benefit in their own words. One educator-writer said attaching a notebook let her mix her own draft material with recent web research, with attribution showing which ideas came from her sources and which from the web. She added that attaching several notebooks to one conversation solved the problem of each notebook being isolated.4

Note-taking turns into capture, and nearly meeting capture

The recent feature additions also bring Gemini Notebook closer to AI note-taking and meeting assistants. Its September update added an audio recorder to the mobile app for recording lectures or spoken thoughts, with the resulting notes stored next to sources so they can be cited and edited.14 Users 18 and older can also talk with their notebooks in real time in nearly 100 languages, with answers grounded in their sources and the option to interrupt by speaking.12

Google markets these features to students, and the coverage largely does too. Mechanically, though, "record audio, store it as a citable source, query it by voice" is the core of a meeting assistant. Our reading is that Google is building the pieces for meeting and lecture capture inside a research tool, not as a separate transcription product. That puts it closer to dedicated AI meeting assistants than its study-focused marketing suggests. The coverage does not show Google packaging it that way yet. One guide noted that the recorder launched for English output only, and that capturing reading material is still a gap.15

Where the coverage diverges

On the basic facts, the reporting is very consistent: two-way sync, custom instructions, web-first rollout and tier-based limits all match across Google's posts and the tech press.1569 The differences are in the details and in tone.

Source limits are one example. One April explainer said Google had not given exact numbers and guessed that NotebookLM's 300-source cap probably applied.7 Android Authority later reported a tiered structure that went as high as 600 sources for Ultra.18 The tiered figures are more recent and more specific, so readers should treat the earlier guess as outdated.

How well the integration works is another. Enthusiast coverage calls the combination a shift from generalist chatbot to work platform.7 Some early users reported rougher results. One Reddit commenter testing attached notebooks found that content from large sources seemed truncated, with Gemini sometimes reading only the beginning and end of a document.2 That was in January, before notebooks were built into Gemini, so it may not describe the current product. Still, it is a reasonable caution: "grounded in your sources" means only as much as the model actually reads of them.

Privacy is the third. At least one analysis said Google had not clearly stated whether material in paid users' notebooks could be used for training once it flows through Gemini, and advised reading the updated terms before uploading sensitive documents.7 The Connected App's default-off setting in the EEA, also seen in the AI Mode and study-notebook rollouts, suggests Google is handling European regulatory exposure carefully.131511

The tiering problem

One pattern shows up across the whole timeline: the most advanced features reach the most expensive plans first. Cinematic Video Overviews launched only for Ultra, code execution reached Ultra before Pro, and one guide argued that the gap between free and top-tier users grew even as the free tier improved.15 The Connected App is an exception. It launched immediately for every account type, including personal accounts.13 That fits our reading: Google treats the notebook layer as basic infrastructure it wants everyone using, and keeps premium features like video and code for paying users.

The bottom line

On its own, the Connected App is a convenience. Taken with the rest of 2026, it shows Google's strategy clearly. It is not trying to choose between a general chatbot and a grounded research tool. It is turning the notebook, meaning a user's curated sources, instructions and history, into a shared layer that Gemini chat, Search and the standalone app all draw on.13 For people choosing among AI productivity tools, the question is shifting from which chatbot writes best to which ecosystem keeps their working context most reliably. On that measure, Google has built the most connected setup of the major players, though it still has to show that grounding holds up at scale and that the data-handling terms can be trusted.