Google NotebookLM Just Got a Major Upgrade: Agentic Capabilities & a New Name!

Author: Universe of AI · Watch on YouTube

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: Google NotebookLM Just Got a Major Upgrade: Agentic Capabilities & a New Name! · Channel: Universe of AI · 09:24 · Watch on YouTube

Topics: AI agents, Task and workflow automation, Data, databases and embeddings, AI developer tools, Health and science

Summary

The video explains that NotebookLM has been renamed Gemini Notebook and describes recent upgrades to its chat and research capabilities. The updated chat can generate outputs such as spreadsheets and PDFs directly from notebook sources, while Agentic Research can begin with a loosely defined idea. Demonstrations show the tool compiling a formatted spreadsheet and creating a PDF report with a location graphic. The presenter argues that Gemini Notebook is expanding beyond a research aid into a broader tool for analyzing sources and producing practical outputs.

Key points

NotebookLM is now called Gemini Notebook, though the presenter prefers the old name.
The chat feature has been upgraded to support more interactions than simply asking questions about sources.
Users can request outputs such as charts, PDFs, spreadsheets, and images directly through notebook chat.
Agentic Research can work from a simple idea and questions, rather than requiring a detailed prompt.
A demonstration shows the tool creating a formatted spreadsheet with company details, formulas, and source-status indicators.
The presenter also demonstrates generating a market-analysis PDF and a graphic showing company locations.
The presenter suggests potential uses for students, businesses, and personal tasks.
Gemini Notebook is described as evolving from a research tool into a platform for working with and analyzing data.

Chapters

Start	Chapter	Summary
00:00	NotebookLM gets a new name	The presenter introduces the Gemini Notebook rebrand and previews improvements to research and chat.
01:02	Upgraded chat and Agentic Research	The video outlines expanded chat outputs and explains how Agentic Research can work from a loose idea.
03:06	Spreadsheet and PDF demonstrations	The presenter demonstrates generating a detailed spreadsheet, a location graphic, and a formatted market-analysis PDF.
07:49	Use cases and what the upgrade means	The video gives examples for students, businesses, and personal tasks, and frames Gemini Notebook as a data-analysis tool as well as a research tool.

Quotes

nothing has changed except for the name. — 00:00

You can create charts, PDFs, spreadsheets, images, and more directly through the notebook chat feature — 01:33

you can just start with a simple loose idea and some questions — 02:03

This is becoming more of a data and analytics platform — 08:52

Concepts and entities

NotebookLM, Gemini Notebook, Agentic Research, Google, Google DeepMind, Python, Excel

Full transcript (original language)

00:00 A lot of you guys who have watched my channel know that one of the products that I love from the Google ecosystem is Notebook LM. Well, unfortunately, Google decided to cancel Notebook LM. I’m just kidding, nothing has changed except for the name. And this is something that Google is known for, changing the names to products [music] constantly and it looks like Notebook LM was the first victim in the Gemini ecosystem because they renamed it to Gemini Notebook. Now, I don’t know if I like the name Gemini Notebook. I like Notebook LM much better. So, throughout this video, I’m going to try my best to refer to

00:31 Notebook LM as Notebook LM, but I’m going to be talking about Gemini Notebook, [music] which is the same thing. I hope you guys are not as confused as I made it all sound, but today we’re talking about Notebook LM and Gemini Notebook, which is a new name for it and the new updates that you probably have missed over the last couple of months. With new AI models and a lot of new labs and open-source development, one product that has quietly improved in the back end is Notebook LM and I’m going to show you exactly what is new. First things first, now Notebook LM is known as Gemini Notebook. So, there are a couple of

01:02 areas where Gemini Notebook has improved. The first thing is the chat feature. Originally, this chat feature was not the best. Most of the times, a lot of people would import their notebooks into the actual Gemini chat platform and chat with the notebook in there. Well, they have decided to completely change that experience for all users, which I think is probably the biggest update here because now you can talk directly to your Gemini notebooks like you would talk to a normal LLM because the chat feature is a little bit more powerful, smarter, and there are more interactions that you can take with

01:33 this compared to before. Before, you’re kind of limited to just ask it questions about your sources. Now, that is not the case. There’s also new outputs. You can create charts, PDFs, spreadsheets, images, and more directly through the notebook chat feature, which is something that before this update was an extra additional step. So, yeah, it’s a small update, but I think it makes the whole experience using Gemini Notebook much better. Then, there’s also agentic Research. Before you kind of had to have a pretty solid prompt about what you wanted to learn or what you wanted to

02:03 research about. Now with Agentic Research, you can just start with a simple loose idea and some questions and then Gemini Notebook will handle the rest. And these are some of the biggest, I would say, usability features or updates that I personally have been loving. And I’m going to show you exactly what they look like. So the platform itself hasn’t changed a lot. It’s pretty much the same platform. These are my old notebooks, but we’re just going to open up a new notebook. The first thing I’m going to do is ask you a simple question like I want to learn about AI companies that operate

02:35 across the world. I’m just going to use the fast research mode, not the deep research mode, just because I want to show you guys what’s new with the new updates that they added. Then I’m just going to press enter. So this process has not changed. The experience itself has not changed. The biggest update is the chat feature. So once we have our research and we import our sources, we can interact with the updated chat. So look, it looks like we already have our updated research. We can see that it did a quick research on all these companies. I’m going to import all these sources into my notebook. So one of the updates

03:06 that they added is a spreadsheet feature. Well, technically we kind of had that. We had a data table, which was kind of limited. You know, you would click on this, you would maybe customize this and you could export this as a spreadsheet. But now I can ask it for more detailed things like make me a spreadsheet with the list of the AI companies their products and when they were founded. go through all my sources and kind of create this. Before this chat feature was kind of very limited. What it had in the sources tab was all you could get, but now it could do a little bit more because of the

03:37 intelligence behind the chat feature now that has been updated. So I’m going to press enter. So you can see it’s analyzing the company data, compiling the spreadsheet, structuring the spreadsheet, and then we are going to get the updated spreadsheet over here. So our output is done generating our spreadsheet. And one thing you’ll notice is that every time there is a new output, it still gets stored underneath the studio panel. So, this is the Excel file, and then you can see that it had some formatting and Excel formulas that I applied. So, this is how powerful this new Gemini notebook is because creating

04:09 a spreadsheet before was you just creating a table. There wasn’t obviously Excel formulas or things like that, but we have all of these other powerful features that I added to the spreadsheet, so it can really speed up your workflow. This is a simple example, but just imagine you give it your own company data, your own company files, whatever the case is. But, give it something like that, it can understand your data, give you a spreadsheet, break it down for you, and this could speed up your workflow. So, we’ll just take a look at the spreadsheet. So, I’m so So, I’m just going to open this up. So, you can see how clean and formatted this

04:41 Excel spreadsheet is. And this is from a simple prompt that I gave Gemini notebook. We have our company names over here, when it was founded, headquarters, key products and models, which is something I didn’t ask it to do. And then notable highlights and contacts, which I liked. And then data status and everything. So, like where we got it from. You can also see that, okay, when it says standard white, it means that it is specifically grounded. Blue value is fully supported by the upload source. Yellow means that they’re inferring, and then light means that they are computing that KPI or metric. So, you can see that

05:13 it counted 12 Frontier Labs, total 18 entities, and then six chip companies, which are over here, TSMC, Samsung, things like that. Now, that was just one example. And then here’s another example. I’m asking Gemini notebook to turn my sources into a formatted PDF report about market analysis. At the end, I wanted to create an image with where they are all located, the companies. And then I’m going to press enter. So, Gemini notebook is done creating our PDF as well the image we requested. And it’s once again in the studio panel. But, this time around I

05:44 want to show you guys the thinking behind all of this that we are seeing. For example, I’m not going to go through obviously breakdown all of the thought process it had because there’s a lot over here, but you can see analyzing the data, examining global usage, reviewing sources, pinpointing headquarters, identifying key players, and then it’s analyzing the geographic data be because I asked it to, you know, create an image with where all of them are created. And when it’s actually creating these diagrams, it’s all happening through Python, creating code, things like that.

06:14 So, all of this is like code that is executing and you can even see at some lines where it says executed code. It first thinks, then it executes the code to kind of create the output that we see. But, here’s the image for example. Here’s the one image that we can see. I asked it where everything is located. So, this is what we see like all of the locations. This is probably some data centers as well as some semiconductor companies we can see over here like Tokyo. It’s pretty nice like you we know exactly areas that are of important Silicon Valley being a little bit bigger than the rest probably signaling that,

06:46 “Okay, most of the companies are operating over here.” And then we have some movement over here and it’s also color-coded. You can see that some are green, some are more, I guess, purple. I wish I added a legend to really understand what is going on, but we’ll just, you know, cuz you might understand this if you are aware of what’s happening, but if you’re not aware, then you’re just kind of confused. So, a legend would have been nice, but that’s fine. Let’s just take a look at the PDF that it created. So, look at this PDF that we get from the model now. If you were used to the

07:18 old notebook LM or sorry, Gemini notebook, the PDF or the report output that you would request before was not formatted like this. Like this is insane. This allows us to speed up our workflow as well ensure that, you know, everything is covered in our research. Sometimes you might run deep research and you have like more than 20, 30 sources. Creating a PDF report that summarizes everything not only helps you understand it, but like if you are sharing this with somebody else, this is probably going to be more useful to them than your notebook LM output, that the

07:49 generic output that we get from the studio, because this formatting just makes it more enjoyable to look at, more important, and more usable in actual workflow. So, yeah, these are some of the examples that I showed you guys. There are other examples that they actually posted on X as well. For example, you can if you’re learning something, you can create five quizzes by just asking it to just create five quizzes or turn your raw lecture notes into a formatted PDF study guide, complete with the five-question practice quiz. So, for students, this new update is going to be huge. Then you can add

08:21 all of the URLs from this class syllabus as sources, update your sources instantly. For businesses, which I think is really crazy, calculate revenue growth across each product line and plot the trends, which you can request. And maybe I’ll make a follow-up video about how you can use the spreadsheet feature and how powerful it is. Then you can also ask it things like creating a marketing brief. Personal reasons, you can ask it to like organize your receipts for kitchen renovation, make a spreadsheet, or you can help me analyze this property purchase, look at recent market performance. So, Gemini notebook,

08:52 I think is becoming a big place that a lot of people are going to use, not just for learning, which was its original intention, but now you can also work with your sources, understand your sources, analyze your sources, and create outputs that are going to help you become more analytical about your data. So, this is becoming more of a data and analytics platform, I think, rather than just a simple research tool. So, good job by the Google DeepMind team, because I am definitely enjoying this new [music] Gemini notebook. But, that’s it for today’s video. Make

09:24 sure you guys are subscribed to the channel, follow our new newsletter as well at universeofai.beehive.com, as well subscribe to the main channel World of AI, and support us on X by following the universeofai z as well. Until then, I’ll see you guys in the next video.