https://www.youtube.com/watch?v=_i1RMwqmQm0 The video highlights five new upgrades for NotebookLM that enhance its power and integration with other Google tools.

1. Data Table Generation

NotebookLM now allows users to turn any source—including YouTube videos, websites, and uploaded files—into structured data tables. You can customize these tables by defining specific columns and types of information to extract, such as movie rankings, directors, or major awards from a film review transcript.

2. Enhanced Exporting and Source Integration

Notes created within a notebook can now be:

  • Exported directly to Google Docs or Google Sheets: Useful for further formatting or data analysis.
  • Converted into sources: Individual notes or groups of notes can be added back into the notebook as new source material, allowing the AI to reference and build upon your previously generated summaries.

3. Notebook Integration with Google Gemini

You can now import entire NotebookLM notebooks as sources into Google Gemini. This enables:

  • Combining multiple notebooks: Synthesize information from different projects.
  • Asset generation: Use the knowledge from your notebooks to have Gemini generate code for landing pages, web apps, or even images through its built-in tools like Canvas.
  • Deep research: Augment your notebook data with real-time online research via Gemini.

4. Powered by Gemini 3

NotebookLM’s search and chat functionalities have been upgraded to run on Gemini 3. This leads to significantly faster performance, more thorough and accurate analysis, and more natural, “human-like” interactions when chatting with your sources.

5. Advanced Chat History and Sharing Controls

Several management features have been added to streamline the user experience:

  • Chat History Management: The ability to delete chat history within a notebook.
  • Finer Sharing Permissions: Choose between sharing full notebook access or limiting collaborators to “chat only.”
  • Configurable Chat Behavior: You can now set specific goals for the chat (e.g., “Guided Learning” or a custom persona like “PhD Student”) and define your preferred response length.