NotebookLM Update: Transforming Chat into a Source Repository with Gemini 3.5

TL;DR

• Google has updated NotebookLM so users can start a chat about a project and let the tool help build a source repository by suggesting relevant materials from research and Google Search.
• The app is also moving to Gemini 3.5 as its default model, which Google says should improve quality, performance, and context handling.
• The changes are rolling out first to Google AI Ultra users and eligible Workspace customers, with broader availability planned later.

Google’s latest NotebookLM update is aimed at making the product feel less like a note-taking app and more like an active research partner. The biggest change is a new workflow that starts with conversation: users can open a chat about a project, and NotebookLM will help assemble a knowledge base by recommending sources through its research capabilities and Google Search.

The company is also switching NotebookLM’s default model to Gemini 3.5, replacing the previous baseline with a newer version designed to improve answer quality, speed, and contextual understanding. In practical terms, that means the assistant should be better at following complex threads, grounding responses in the right material, and keeping long research sessions coherent.

Chat as a source-finding workflow

NotebookLM has long let users add sources manually, but the new update pushes the product toward source discovery inside the chat itself. Instead of requiring users to first collect documents, web pages, or files before asking questions, the app can now help identify what should go into the notebook as the conversation unfolds.

That matters because research often begins with broad questions, not curated libraries. Google’s approach appears designed to reduce the friction between “I have an idea” and “I have a usable source set,” especially for projects that start with incomplete information.

The feature also fits into NotebookLM’s broader source-management tools. Google already supports adding or discovering sources from the Sources panel, including web-based research queries and Deep Research mode. The new update extends that discovery behavior into the chat experience, making the workflow more conversational and less procedural.

Why Gemini 3.5 matters

The move to Gemini 3.5 is important because NotebookLM depends on retrieval, synthesis, and context management more than generic chatbot behavior. According to Google and early reporting, the upgrade is intended to improve performance and the quality of responses across longer, source-heavy interactions.

That should be especially useful for users working on research notebooks with many documents, where the model has to identify relevant passages, maintain context over time, and answer with tight grounding in the provided material. CNET reports that NotebookLM’s chat now supports a 1 million token context window for Gemini, which helps it handle much larger amounts of source material in a single workspace.

Taken together, the model upgrade and the new source-discovery workflow suggest Google is trying to make NotebookLM more useful for sustained research projects, not just quick Q&A sessions.

New outputs and editing flexibility

The update is not limited to finding sources and improving chat quality. Yahoo’s reporting says Google is also adding more control over how NotebookLM produces output, including the ability to give specific instructions for different formats and then edit the generated result.

According to that report, NotebookLM now supports exports in formats such as charts and visualizations, documents, images, structured data, and office files including Excel and PowerPoint. If accurate, that broadens NotebookLM’s role from a research assistant into something closer to a multi-format content production tool.

That direction is consistent with Google’s recent broader push to make NotebookLM more flexible for studying, writing, and presenting rather than simply summarizing source material.

Who gets access first

Google says the update is available starting today for Google AI Ultra users and for Workspace business customers with AI Ultra Access and AI Expanded Access. The company is framing this as an initial rollout, with broader availability planned later.

That staged release matters because NotebookLM has a large audience across free and paid tiers, but the newest capabilities are often introduced first to premium and enterprise users before they spread more widely.

What it means for NotebookLM’s future

The new update signals that NotebookLM is evolving into a more autonomous research environment. Instead of treating sources as something users must gather elsewhere, Google is making the assistant participate in the process of building the notebook itself.

At the same time, the shift to Gemini 3.5 suggests Google wants the product to feel sharper, more capable, and better suited to long-form work. If the rollout performs as advertised, NotebookLM could become a stronger option for students, analysts, and knowledge workers who want one place to discover sources, organize evidence, and turn research into usable output.

AndroGuider Team

Articles written by the AndroGuider team. We try to make them thorough and informational while being easy to read.