I have written numerous AI articles over the past year. These cover a wide range of topics from descriptions of what Chatbots, Artificial intelligence (AI), and Large Learning Models (LLMs) are, how they can be used, which are the best depending on your particular use case, and how to roll them out in your organization.

This article covers NotebookLM from Google – the best AI tool I have used. I use ChatGPT, Gemini, Claude, and Copilot daily. I tend to use Gemini most often as the data is current. But as discussed in the article “Practical Guide to Using AI Chatbots“, each is best for a set of use cases. NotebookLM is not a general-purpose AI Chatbot. It is focused on a use case that I have come to depend on.

What Is NotebookLM?

NotebookLM (notebooklm.google.com) is an AI-powered app from Google that helps users analyze research material and extract key themes. It’s designed to help users synthesize information from multiple documents, making it useful for students, scholars, journalists, analysts, and other knowledge workers. 

NotebookLM works by asking users to create a virtual notebook to store all their text, documents, and other research. Users can then upload their own sources, such as Google Docs, PDFs, or copy/pasted text, and NotebookLM becomes an expert in those sources. Users can then use NotebookLM to transform their notes into an outline, blog post, business plan, and more. 

NotebookLM is an experimental product built by a small team in Google Labs. It is available to anyone through Google’s experimental Labs service and is currently available in the U.S. to ages 18 and up. Users can join the waitlist to access it with a personal Gmail account.

Take a look at the excellent article “Google’s Notebook LM – Was it written for ME?“, written by Deena A. Mims. She does a great job of summarizing the capabilities of NotebookLM.

What is it Best For?

NotebookLM shines in tasks that involve integrating your research and knowledge with the power of AI. Here are some areas where it excels:

  • Research Assistant: If you’re working on a project that requires gathering information from various sources (like articles, books, or PDFs), NotebookLM can be a game-changer. Upload your materials, and it will summarize them, identify key points, and help you build connections between them.
  • Writing and Brainstorming: NotebookLM can help you transform your research into well-structured writing. Use it to turn notes and quotes into outlines, or brainstorm new ideas based on the information you’ve gathered.
  • Personalized Knowledge Base: Unlike traditional search engines, NotebookLM tailors its results to the specific documents you provide. This allows you to create a personalized knowledge base around your research topic.
  • Understanding Complex Information: NotebookLM can help you break down complex concepts by summarizing information and highlighting key points.

Here are some things to keep in mind:

  • Note-taking and Organization: While NotebookLM can process your notes, it’s not a traditional note-taking app. There are better options for basic note-taking and organization.
  • Direct Question Answering from Document Sets: NotebookLM isn’t currently designed for directly answering questions from a defined set of documents like some Generative Question Answering systems. However, it can help you prepare the documents for Q&A by summarizing and extracting information.

What Do I Use It For?

I use NotebookLM for many different use cases. I find that it saves me a tremendous amount of time, and in many cases provides a perspective that would be difficult to develop. Below are three examples.

Research Assistant:

I am very interested in health and fitness-related topics. I have had a hard time separating opinion from fact. To address this, I decided I would base my personal decision on research. I search for research on a particular topic that I am interested in. I create a notebook in NotebookLM on the topic and load it with PDFs of the research papers that I find. The first thing that I then do is ask which have conflicts of interest. I remove them (uncheck the documents in NotebookLM) and begin to ask my questions.

An example of this is whether coffee is healthy. I have found a large number of research papers on this. I load them into a notebook. I select all of the documents and ask “Are there conflicts of interest in any of these papers.” I then unselect any that have conflicts. I then begin my questions. I start with questions like “Summarise the results across all of the research documents” and “What does the research say about coffee and longevity?” Depending on the answers, I will ask more questions to clarify. I use the capability to pin responses to save things that I think are most important.

Help Parsing Long and Complicated Documents:

What do we typically do when we receive long and complicated documents, things like insurance coverage, contracts, software licenses, and so on? We don’t read them – we just hope that they are right. NotebookLM can help with this. Load the documents into a notebook and ask the questions that you would otherwise want answers to.

An example of this is my home insurance. I should have read these in the past, but I didn’t. I kind of knew what was covered, but it was not until I had a problem that I found out exactly what was and what was not covered.

We recently had a flood in our basement from a large storm. I didn’t expect to get any money as I thought I was not covered. I was only partially right. I had coverage on my sump pump that gave me a few thousand dollars back.

What I have now done is loaded these documents into NotebookLM in a separate notebook. I have asked several questions and now understand my policy better. For example, when I ask about my previous situation, it explained exactly what was and what was not covered.

Rules and Policies.

There is nothing more boring than reading rules and policies (at least for me). I just don’t read. When I have a question I end up spending too much time trying to find the answer and often just give up and hope I am doing the right thing. NotebookLM has helped me with this.

An example of this is my HOA rules and regulations. I live in a complex with a HOA. There are lots of rules and regulations. Most don’t effect me. I know where the documents are and at times have tried to look up information. It was hit and miss as to whether I found what I was looking for. I have now loaded these into a notebook in NotebookLM. I ask questions and immediately find what I am looking for. For example, the other day I wanted to know what time the pool closed at night. I asked and it gave me the answer. Super simple. I was also interested in whether we could replace a plant in the front yard. I did the same and got my answer. Super nice?


This is a new and developing space. I have found two options for NotebookLM that are specifically focused on providing a similar solution. You can also similarly use generic LLMs, but my results have been mixed. The best is Claude. You can load multiple documents and query them. The other is Gemini, which provides a mechanism to include documents stored in Google Drive.

Dot (dotapp.uk)

Dot is a groundbreaking application developed by BluePoint, aimed at revolutionizing the way users interact with their documents on a local computer. Its innovative approach seamlessly combines document management and communications providing users with a unique and efficient way to engage in real-time conversations with their files.

The pros of using Dot are that it is free, open source, and evolving. The cons is that it is open source and therefore may not have the level of attention required to address bugs and add new features.

ChatRTX (www.nvidia.com/en-us/ai-on-rtx/chatrtx)

ChatRTX is a demo app that lets you personalize a GPT large language model (LLM) connected to your content—docs, notes, or other data. Leveraging retrieval-augmented generation (RAG), TensorRT-LLM, and RTX acceleration, you can query a custom chatbot to quickly get contextually relevant answers. And because it all runs locally on your Windows RTX PC or workstation, you’ll get fast and secure results.

The pros of ChatRTX are that it is free and provides very similar capabilities to NotebookLM. The cons is that it is a demo app and therefore not supported. NotebookLM is also not production, but it is not described as a demo.


AI Chatbots are incredibly valuable for a broad range of tasks. But if you want to limit the scope to only what is contained in a set of documents, NotebookLM is a far superior solution.

There are two caveats to using NotebookLM. One is that it is still experimental. Google may continue the project or may not. If they do continue it, they may start charging.

The second is data privacy. NotebookLM keeps your data private. But this is an experimental platform. It is not yet a product and therefore there are no guarantees.

My recommendation is to use NotebookLM for things that are not confidential. If you have confidential material that you do not want to leak, consider using Dot.

See Also

Google’s Notebook LM – Was it written for ME?

Categories: Tools


Leave a Reply

Avatar placeholder

Your email address will not be published. Required fields are marked *