NotebookLM Updates: October 2025

Resources:

Join the discussion: Got something to say? Let us know on LinkedIn and network with other AI leaders


Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup

Connect with Jordan Wilson: LinkedIn Profile

Unlocking Real Value with Google’s NotebookLM: October 2025 Updates and Practical Walkthrough

Google's NotebookLM has consistently flown under the radar, yet the most recent October 2025 updates have made it an even more critical tool for business leaders aiming to leverage artificial intelligence for tangible, day-to-day outcomes. This article unpacks the latest enhancements, demonstrates their real-world application, and highlights how grounded, source-specific AI models like NotebookLM directly address central challenges in knowledge work, data management, and decision-making.

Customized AI Responses: Precision in Tone and Length

A longstanding challenge with AI chatbots in business environments is the inconsistency in response style and verbosity. NotebookLM’s new capability to customize both the tone and response length changes this dynamic. Unlike default models that might return overly verbose and bland outputs, users can now configure responses to be direct, concise, and even inject customized humor or formality. This transformation is immediate: toggling these options results in summaries that are a fraction of the original length, tailored to the unique communication styles and time constraints within professional teams.

For instance, in internal knowledge dissemination or marketing campaign recaps, stakeholders can demand pointed, minimalistic responses for efficiency, or request more expansive explanations when onboarding new staff. These configurations are accessible from the 'Configure Notebook' option within the chat interface, offering the choice between conversational styles (custom or default) and response lengths (default, longer, or shorter). This feature enables professionals to adapt AI assistance precisely to their workflow rather than adjusting to a generic chatbot’s output.

Learning Guide Style: From Summaries to Strategic Comprehension

Effective upskilling and competitive intelligence depend on smart knowledge checks. The new Learning Guide mode in NotebookLM prompts users with adaptive questions that clarify intent and objectives before producing a summary or analysis. The system queries users—“What is your high-level goal for reviewing this material?”—forcing clarity and focus on business priorities (such as technical nuances, business implications, or general overview). The AI, in turn, becomes a true thought partner, guiding teams to critically engage with content rather than merely consuming summaries.

This approach is especially valuable when analyzing competitor launches, product updates, or complex industry reports. Instead of a one-size-fits-all recap, stakeholders are challenged to articulate what matters most, ensuring outputs are aligned with organizational objectives, not just information overload. The learning guide’s inherent prompt engineering helps teams translate broad data into actionable insight.

Grounded Responses for Trusted Decision-Making

Perhaps the most significant advantage highlighted in recent updates is NotebookLM’s “grounded” architecture. Unlike most large language models, NotebookLM does not pull from its general training data or the web during analysis; it strictly analyzes and cites only user-provided source materials. For business owners, this means responses are always anchored to uploaded documents—PDFs, transcripts, reports, or proprietary data—eliminating the risk of AI “hallucinations” and speculative reasoning.

For example, marketing teams can upload white papers, executives’ memos, or research datasets and receive consistently cited, document-linked insights, ensuring that all AI-generated recommendations directly trace back to internal knowledge, not potentially erroneous external data. This boosts compliance, transparency, and auditability, crucial for regulated industries or high-stakes decision environments.

Integration with Google Drive: Streamlining Source Management

The latest update enhances discoverability of business documents through Google Drive integration, allowing for seamless import and search of company assets. Users can initiate a natural language query within NotebookLM’s interface and instantly pull relevant files or folders from connected Google Drives. This simplification streamlines onboarding of source materials for analysis, synthesis, or learning workflows, eliminating manual upload friction and ensuring up-to-the-minute relevance.

Studio Panel: Interactive Multimodal Engagement

Beyond chat and summaries, the Studio Panel in NotebookLM provides a suite of interactive tools to elevate business learning:

• Audio and Video Overviews: Automatically generate podcast-style or video explainers from source materials. These are not limited to static narration—users can select from templates such as brief reports, in-depth analyses, critiques, or debates. The “Join the Conversation” interactive audio mode enables real-time Q&A for ongoing learning.
• Mind Mapping: Visualize connections between concepts, projects, or product lines within uploaded sources, allowing teams to explore relationships and dependencies at a glance.
• Reports, Flashcards, and Quizzes: Convert resource libraries into briefing documents, training materials, and instant assessments. This is particularly useful for preparing staff for customer interactions or policy changes as reports and flashcard decks are dynamically generated and saved for ongoing access.

What's Next: Visuals, APIs, and Expanded Capabilities

Coming soon, NotebookLM is expected to integrate with Google’s Nano Banana AI image generator to create infographics and enhanced video visualizations. Plans are also in place for automatic chat history saving and public API access, which will open pathways for customized business workflows and private integrations. These enhancements are tailored for business operators looking to automate reporting, generate on-brand training modules, or build proprietary knowledge workflows.

Practical Takeaways

• Customized, Purposeful Communication: Direct and concise AI responses tailored per task save executive and team time.
• Trusted Knowledge, No Guesswork: Responses are always sourced and cited from uploaded content, supporting compliance and accuracy.
• Seamless Asset Management: Direct integration with Google Drive ensures source material is current and easily ingestible.
• Multimodal, Engaging Learning: Studio Panel features empower operators to turn static knowledge into interactive, actionable formats for different learning or stakeholder engagement modes.
• On the Horizon: New features such as visual generation and API access will further support customizable business intelligence and learning at scale.

NotebookLM’s October 2025 suite of updates goes well beyond incremental improvements. In everyday business terms, it equips knowledge workers, managers, and decision-makers with precisely the tools needed to convert ever-growing data into actionable, reliable insight—directly tied to their own organization’s knowledge base.

Topics Covered in This Episode:

1. NotebookLM October 2025 Feature Updates
2. Customizing Response Length and Tone
3. Learning Guide Style for Knowledge Testing
4. Google Drive Integration in NotebookLM
5. Grounded AI Responses Using Source Material
6. In-Depth NotebookLM Live Demo Walkthrough
7. September 2025 Flashcards and Quizzes Launch
8. Audio and Video Overview Enhancements
9. Upcoming Infographic and Nano Banana Visuals
10. NotebookLM API and Chat History Roadmap
11. Studio Panel Reports, Mind Maps, and Multimedia
12. Saving and Managing Notes in NotebookLM

Episode keywords

NotebookLM, Google NotebookLM, AI tool, AI chatbot, personalized AI, grounded AI, Gemini 2.5 Flash, hybrid model, source grounded, custom response length, custom tone, customized responses, learning guide, learning modes, Google Drive integration, discover sources, citation features, chat pane, conversational style, response configuration, chat history, API integration, Nano Banana, Google Nano Banana, AI image generator, infographic feature, video overview, video overviews, audio overview, multimedia learning, mind map, flashcards, interactive quiz, reports, strategic analysis, multi-version creation, multitasking in AI, upload limits, hallucination-free AI, citation transparency, prompt engineering, AI for business leaders, AI powered research, AI workflow creation, competitor analysis, AI for knowledge workers, project templates, summarized transcripts, AI for content organization, personalized learning module, custom instructions, AI workflow, AI document analysis, AI chat, AI use cases, review automation, strategic memo, OpenAI dev day, ChatGPT comparison, Claude, Copilot, Chegg, advanced AI features, AI-powered podcast, advanced note taking, AI productivity

Podcast Transcript

Jordan Wilson [00:00:47]:
One of the most underrated and underutilized AI tools keeps getting better and better. I'm talking about notebook l m from Google. Long before you ever heard a notebook l m ad on this podcast, we called it our twenty twenty four AI tool of the year, and it keeps getting better and better. We've covered a couple episodes over the last couple of months, but I think this last round of updates that just came out are worth your attention. So much so we are dedicating an entire episode to go over the new features in notebook LM, what's potentially next, and since it's been a while, we're gonna do a complete walk through. It is putting AI to work on Wednesdays, and this is Everyday AI. What's going on, y'all? My name is Jordan Wilson, and welcome to Everyday AI. This is your daily live stream podcast and free daily newsletter helping everyday business leaders like you and me not just keep up with AI developments because there's a ton, but how we can actually make sense of them and use them to grow our companies and careers.

Jordan Wilson [00:01:53]:
If If that's you, it starts here with the unedited, unscripted live stream podcast. If you wanna take it to the next level, go to our website at youreverydayai.com. Sign up for the free daily newsletter. We're gonna be recapping the highlights and what you really need to know from today's show, and all the AI news so you can stay on top of your game. Alright. Let's get straight into it. October, new updates, notebook LM, tons happening and a lot more on the horizon. So let's go over what we're gonna go over.

Jordan Wilson [00:02:24]:
We're gonna quickly recap what's new in NotebookLM with their customized features that I think make it truly personalized in the same way that you might think of GPTs, for chat GPT. This is kinda like a slimmed down version of that, but I think it's really important and really expands notebook l m's usefulness. Next, we're gonna preview what's confirmed coming next from notebook l m, whether that's in a couple of days or a couple of months. And since it's been a while, I'm gonna do a complete but very quick walk through of the entire platform. That's, I think, gonna be really helpful for beginners, but like I said, there's been some huge updates in the last couple of weeks, a couple of rounds. So even if you're using notebook l m, I don't know, once or twice a week, I think you're still gonna get a ton of value out of today's show. And stick around to the end. I put together I've got about, I don't know, 20 or 30 of these kind of bonus guides over the last three years.

Jordan Wilson [00:03:20]:
And this one y'all is by far the best. Put together a notebook l m cookbook and some recipes, use cases, and just some really, exploratory an exploratory guide on notebook l m. We have sections for beginners, intermediate, and advanced. You are literally not gonna wanna miss it. And well, you all voted for this show, FYI. So it just so happened, yeah, we have notebook l m ads, but this is not, you know, technically an advertised segment. You all voted for this. I've shown it on my screen for the podcast listeners and said, hey.

Jordan Wilson [00:03:56]:
On Wednesdays, we do putting AI to work on Wednesdays. You all said you wanted to go over, notebook l m's, custom features. So here's what's new in the October 2025 updates, and notebook l m has been shipping like crazy the last few months. The biggest one is the ability to customize the response length and tone in the chat response. And you might be wondering, what's the big deal about that? Well, let's be real. Up until this, notebook LM was a little dry in its responses. And I think that's okay. But I think as we've come and when I say we, just collective, you know, everyday professionals, as we start to use, large language models like Gemini, Chegg, GPT, Copilot, and Claude more and more, I think we become accustomed to them, becoming more personalized, more customizable.

Jordan Wilson [00:04:58]:
And, unfortunately, notebook l m didn't really have that capabilities. So a lot of times, you know, I might have wanted something short and, you know, I maybe would want a a pointed response or, you know, a no nonsense response, and I might be staring at a huge wall of text, that's overly verbose. Right? So this is great, to say, hey. This is how I want notebook l m to respond and then also having default for lengths. Huge. The new learning guide style as well, and we're gonna be going over these live in a couple of minutes, and that essentially tests your knowledge. So very cool. You know, Gemini and ChatGPT have come out with these learning modes, So it's kind of similar to that.

Jordan Wilson [00:05:40]:
I think notebook LM, by far is the best tool to just learn something and that's because it's grounded. And I'll tell you what that means here in a second. And then, a new another small new thing is, to discover new sources, there is a Google Drive integration. That's a small little footnote here, but the big, the big, update here, this month in notebook l m is the ability for it to respond in a custom way. And it's actually a big update in my opinion. And I love the new learning guide style as well. So enough. Let's jump in live.

Jordan Wilson [00:06:19]:
I'm gonna show you first, those couple of new updates. And let me, preface this, by telling everyone about notebook l m. So if you've never heard of it, it is much different, than your average AI chatbot. Right? It's technically powered by Gemini 2.5 Flash, which is a hybrid model. So sometimes it might give you a quick response. Sometimes it'll think and kind of plan things, step by step. Other times, it's just gonna be quick. Right? So, it is a hybrid, model, which is great.

Jordan Wilson [00:06:53]:
The other big thing to know is it is grounded. And what I mean by that, I'm sharing this on my screen now, and if you wanna see the video version of this, it's not gonna be overly visual, but it might be helpful. The video version is always on our website for free at youreverydaydayi.com. Click on episodes. Alright. So make sure you go check, today's out if you're listening on the podcast and you wanna see all this. But what it means by NotebookLM being grounded is it means it only works with your source material. So when you go into notebook l m and there's a free version and a paid version, the paid version just essentially gets you better limits, but the free version is actually really good.

Jordan Wilson [00:07:28]:
All you need is a Gmail account. Essentially, it doesn't hallucinate, because when you go and you click a new notebook, you can't even start, quote, unquote, chatting with notebook l m. You first have to upload your source material. If I'm being honest, I wish all large language models had a little toggle at the top that you could just go into, like, a notebook LM or a grounded mode. Even when you're using things, like, custom GBTs or projects in chat GBT, gems in Gemini, projects in, in Claude. Right? You think, oh, I'm gonna upload, just my documents and in the custom instructions, I'm gonna tell the model only use my documents. Doesn't work that way. Sometimes it does, sometimes it doesn't.

Jordan Wilson [00:08:14]:
That's where notebook l m shines. And that's one of the reasons aside from the, very viral, you know, AI podcast. Right? That's I think what, kind of, shown a lot of attention on NotebookLM in the earlier days, but I think the fact that it's crowded is the biggest thing. So, you know, if you upload a a bunch of information about, you know, a random example I always throw out there, if you work, in logistics and you're in marketing and you're putting a bunch of new white papers in there and then you're asking it, you know, hey, what's the weather in Chicago next week or who won the twenty sixteen MLB World Series? It'll say, I have no clue. Right? Whereas if you ask Gemini or Chad GPT or any other model, it will obviously tell you that because it's in its training data. So notebook l m is grounded. It only works on the source information that you input. So let's just do a very quick demo.

Jordan Wilson [00:09:06]:
I opened a new notebook here in notebook l m, and all I'm gonna do, I'm gonna jump over to our website. I'm gonna go, grab, yesterday's episode, and I'm just gonna copy and paste. So, I'll get more into how all this works later. Alright? And I'm just gonna paste in the transcript from yesterday's episode, where we went over, dev day updates from OpenAI. Alright. Pretty simple. So now let me show you the what's new in the, and how you can customize the response and why it's super helpful. Alright.

Jordan Wilson [00:09:44]:
So now, in the middle pane, all I'm gonna say is recap, recap this episode. Right? Very simple. In my default, it's gonna take a while because it was a pretty long episode. It was about forty minutes long. There was a ton of information on that page. So it's taking here, you can see, actually actually not too long. Took about, ten or fifteen seconds, and I got a very in-depth, in-depth recap here. And, if you are watching the video version of this, you'll see, one of the special things about notebook l m is it sites, and it gives you a footnote to every single thing it tells you.

Jordan Wilson [00:10:23]:
So if you have a bunch of different sources, it'll tell you that, more here in a second. Let's get to what's new. So if you go into the middle pane, which is the chat pane. Alright. There's three panes. The middle pane is the chat pane, and there's a little toggle there that says configure notebook. Alright. If you click on that, here's what's new.

Jordan Wilson [00:10:45]:
So you didn't really have any way to customize notebook l m's response before. So you kind of always got an overly wordy, usually very dry, but always factual response. That's not for everyone. Sometimes you need a large language model to kind of adapt to how you want to learn. So, you can choose the conversational style and the response length. So I'm gonna go to custom first. Alright? I should've typed this out before so I could copy and paste it. So I'm gonna say, be direct as short as possible, a little humorous, and Kurt.

Jordan Wilson [00:11:29]:
I'm gonna say, don't waste words. Alright? Be direct a f. Alright. There we go. And then for response length, I'm going to choose shorter. So one thing that I hope notebook l m builds into the future, and I'll I'll reach out to some of my, friends there is, to be able to save these. These. Right? This is just a very quick example.

Jordan Wilson [00:11:55]:
I've made some of these where I build them, you know, I really take my time and and do a little testing to get it exactly how I want it. I wish you could save them. You can't do that, right now. Maybe in the future, you will be able to. Alright. Just did some misspellings there, and then I'm gonna save. Alright. So now, the other thing is you also don't know, that you are chatting with that persona, if that makes sense.

Jordan Wilson [00:12:23]:
Alright? So now I'm I'm I'm just gonna go in. I'm gonna click, I'm gonna go ahead and refresh this chat. Alright. I'm gonna double check. There's my, there's my custom, conversational style, and then I chose shorter for the length. And now I'm gonna say the same thing. I'm just gonna say recap this episode, and we'll see here. It's gonna be very short, hopefully, maybe a little humorous, kind of curt, so almost to the point of it, you know, being rude.

Jordan Wilson [00:12:51]:
But, yeah. Right away. It is about a fourth of the length, which I love. Right? And I like, for the most part, large language model responses to be straight to the point. I don't need fluff. I'm busy. I'm in large language models depending on the day, two to twelve hours. Right? I don't like staring at a wall of text and I also want the model to respond in a way that is, that I'm receptive to.

Jordan Wilson [00:13:16]:
Right? It's like choosing, right, how someone communicates to you. It's it's like, what if you could go talk to your boss and be like, hey, boss. You know, I want, you know, you to always tell me basketball references, you know, throw in a funny rhyme once in a while and always be optimistic. Right? I mean, you can do anything, with this new custom response. So very cool. So I'm looking through here. I'm gonna see if it actually, did anything funny. So it said OpenAI dropped the agent builder part of agent kit, a visual drag and drop, that allows non developers to create and deploy AI workflows.

Jordan Wilson [00:13:51]:
Alright. So pretty good. It's talking about the, you know, how ChatGPT is trying to be your new operating system. So I don't see anything necessarily funny. I'll have to, read through this, a little bit more, but it is to the point. Alright. I I kind of see a a couple of things. It's starting to, you know, throw in a little bit of humor here, but great response.

Jordan Wilson [00:14:15]:
And you might be thinking, like, okay. What's the big deal? If you've used notebook l m heavily like me, you like, you know this is a big deal. Because like I said, what are the downsides to notebook l m is the response lengths were just so super long. Alright. Let me show you the one other, or two other new, things. So one is when you add sources, you add sources on the left hand side, and you go to discover sources. Now you can find and discover sources from your Google Drive. So small little thing, but you can just search in natural language and it'll actually pull those things from your Google Drive.

Jordan Wilson [00:14:51]:
There's other places where you can add Google Drive. That's just a small one. Alright. Let's do, the next one. So I'm gonna go ahead and refresh the chat history here in the middle. Again, the only source information I have here is my podcast transcript from yesterday. Now I'm going to go to the configure chat, and I'm gonna go to learning guide. Alright.

Jordan Wilson [00:15:09]:
And I'm gonna keep it shorter. So now I'm gonna do the same thing, and I'm gonna say recap this episode. And you'll see here, this is where you can see NotebookLM is a true learning powerhouse. Alright. So okay, interesting. So it it's it did something different this time. It actually did a a pretty in-depth, pretty in-depth recap here. I'm actually gonna do a a a hard refresh, and I'm gonna try this one more time.

Jordan Wilson [00:15:40]:
I'm gonna make sure that I have the, the learning guide. I'm I'm a put the response on default just to say. So I'm gonna say, recap this episode, and we'll see if that happens again. It gave me a a much longer, recap than I would have thought. Here we go. The second time, it kinda worked. I think maybe I just had to do a refresh. I think it was still pulling, from my custom instructions.

Jordan Wilson [00:16:02]:
Could have been a caching issue. So here we go. This is the, the new learning guide conversational style, which I really like. I just said recap this episode. Notebook l m responds back. That's a great request. This episode covers a lot of ground. And then I said, before we dive in, how much do you know about OpenAI's dev day announcements, and what is your high level goal for reviewing this material? Are you focused on the business implications, the technical updates, or something else? Y'all, this is like prompt engineering one zero one, but spit back at you by the large language model.

Jordan Wilson [00:16:39]:
Right? Going through and iterating and really turning AI into a thought partner. Here, it forces you to do that. And again, think of you might not think, right, by using the the the custom, the custom tone, you might not think to put all of these things in here. That's why I love this new, option for the learning guide, because it forces you to think, okay, well, what do I actually want out of this? Put it through a different perspective. Think that one thing that you're trying to learn. Let's say there's there's a huge new product line that one of your competitors, is putting out. They they they have a big, you know, PR push around it. They have videos, press releases, blogs.

Jordan Wilson [00:17:27]:
They're doing interviews, all this information. And all your customers are saying, hey. Why aren't you putting out product x like your competitor? And you need to really know it. You need to understand it, and you need to tell your current clients or current customers that, hey, ours is still better for this reason. Right? So think you have all this information, as your sources, and then you can use notebook l m to work with it. And you can keep going in, and you can do a learning guide to actually learn a little bit more first about what your competitor is putting out there, and then you can do the same thing and chat, with the custom version as well. Alright. So that is a, very quick overview of what's new.

Jordan Wilson [00:18:13]:
Now let's get back and talk about some other updates. So if you haven't tuned in recently, we're not gonna go over all of these individually, but I will quickly show you them, as we do a complete walk through of NotebookLM. So in September 2025 was probably the biggest month of updates for NotebookLM. So here's what came out, in September 2025. They came out with customer ports. I'm gonna show you all of these things, where you can control the structure, tone, language. It has dynamic templates really, really good. There's now flashcards and quizzes that were released in September, you you know, where you can get instant feedback from your sources, you know, have to really learn them, and then new audio overview formats.

Jordan Wilson [00:19:02]:
So this is one of those things that originally made NotebookLM go boom, was the kind of, deep dive AI podcast host. Right? You could upload, you know, hundreds of pages of documents, click one button, and you get these two very human sounding AIs that kind of riff, on your content and create a personalized podcast for you. So they came out with the new formats, so the brief, the critique, the debate, and then the, deep dive as well as the customizable length and tone, which I really love. And I did go over that much more in-depth on episode six zero eight. So if you wanna know more about that, go listen to that episode. Then in July, there was a ton of updates as well. The biggest one was video overviews. I have no clue how this technology even exists, and I did a whole episode on that so you can go back and watch it, but same thing.

Jordan Wilson [00:19:59]:
Upload any of your source materials the same way you can just click a button and get an audio overview. Well, you can get a video overview as well, And the videos are actually really good. Alright? Especially for the first iteration of a product. A lot of times, the first iteration of a new feature that's technically groundbreaking is absolutely terrible. The video overviews from notebook l m was one of those moments where my jaw hit the floor and I'm like, how is this possible? Right? So the video overviews came out in July, a new refreshed studio UI, which I'll show you when we do our full walk through, multi version creation. So, essentially, you can create multiple versions of that where before you can only create one, and then multitasking. So even when you're watching or listening to the audio overview, you can continue to work or chat inside of notebook l m. Alright.

Jordan Wilson [00:20:52]:
And if you want to know more on that, go listen to episode five seventy eight. Covered it all in-depth. Alright. Before we go any further, it just so happens a word from our partners, which is NotebookLM.

Steven Johnson [00:21:06]:
This podcast is supported by Google. Hey, folks. Steven Johnson here, cofounder of NotebookLM. As an author, I've always been obsessed with how software could help organize ideas and make connections. So we built NotebookLM as an AI first tool for anyone trying to make sense of complex information. Upload your documents, and NotebookLM instantly becomes your personal expert, uncovering insights and helping you brainstorm. Try it @notebooklm.google.com.

Jordan Wilson [00:21:37]:
Y'all, I swear it wasn't planned. I asked you all the newsletter, what do you want for the AI at work on Wednesdays? And you said notebook l m. Alright. Just putting that out there. So I told you some of the recent, including the October updates, the September updates, and the July updates, but what's next? What's coming around the corner? So there's been a lot of kind of confirmed leaks, and notebook l m itself, at least online, has kind of hinted at what's next. So let me tell you what is coming next in notebook l m. Don't have dates on anything yet, but one feature that I'm really looking forward to is an integration with Google's nano banana, the AI image generator. So, there will be an infographic feature, that uses, Google's extremely powerful, the best AI image model in the world, Nano Banana.

Jordan Wilson [00:22:27]:
I'm looking forward to that. Chat history will soon have the option to be saved by default, which is nice. I'll kind of show you kind of how that works. Like I said, notebook LM, little different than your, Gemini Chat GPTs of the world. Nano Banana visuals are gonna be coming to video overviews soon. Not like the visuals in there are bad. They're not. They're pretty good.

Jordan Wilson [00:22:49]:
They're actually well designed. I like the the bright popping colors, but apparently they're gonna be getting even better. And at some point, hopefully soon, there will be an API for notebook l m. And at that point, if if you're an entrepreneur, if you're trying to start a side hustle, when notebook l m comes out My gosh, what you can create, and to, to sell to people it's, it's gonna be absolutely bonkers. All right. So let's now do a quick walk through. I'm putting myself on the clock here. A lot of you are saying, hey.

Jordan Wilson [00:23:26]:
The episodes are a little too long or, hey, Jordan. I'm on the treadmill. I can't do forty five minutes every day. Alright. So we're gonna try to go a little quick here, in notebook l m. I'm gonna do a complete walkthrough, from the beginning. So, here's what we're gonna do. I'm going back sharing my screen, live stream audience.

Jordan Wilson [00:23:48]:
Let me know. Hopefully, you could see it there. Alright. So when you are in notebook LM, you're gonna click create new notebook. So like I said, notebook LM is grounded. It only works with your data. Okay? And there's different ways that you can upload your sources. Alright? And then your sources are gonna go on the left hand side.

Jordan Wilson [00:24:07]:
You can click discover sources, and you can just type in anything. Right? So I can just type in LLM news, click, submit, and it's gonna search oops. I should probably click the actual, button there, and then it's gonna search on the web, and then I can, choose sources one by one to bring into my notebook. Alright. It's taken a quick second here. I can select them all. I can just select one. Here we go.

Jordan Wilson [00:24:36]:
I'll just select this random one right there and click import. Alright. So that is how you can discover sources. You can also add them by uploading them. It supports PDF text markdown and audio. Yeah. You can upload an MP3 and it's gonna automatically transcribe it and you can talk to it. Y'all for someone like me, a podcast host, that's amazing.

Jordan Wilson [00:25:02]:
Right? To be able to talk to my entire catalog of episodes, it's feels illegal. That's why I mean, one of the many reasons why notebook eleventh was our tool of the year in, in 2024. Alright. You can also click to connect your Google Drive. You can select certain folders. You can bring in links, to a website, links to a YouTube video, or you can copy and paste text. So that's the different ways that you can bring in different sources. So I'm gonna go ahead.

Jordan Wilson [00:25:31]:
I'm gonna paste text here just so I can have, two different things. I'm gonna paste in what I pasted in earlier, which is the transcript of the dev day episode that we did yesterday. So now I have two different sources. Okay. And hopefully, even if you're listening on the podcast, hopefully this can make sense. The panes are important. Okay? There's essentially three panes or three different areas inside of notebook l m. And the cool thing is is you can adjust them.

Jordan Wilson [00:26:02]:
So you can, collapse the source panel to give more room to the chat. You can also, collapse the studio panel on the right hand side. So if you want a little bit more room, the middle pane is the chat pane, and that's where you do, you know, a lot of your work. So in that instance, you are talking to Gemini 2.5 Flash. Alright? But only to the data that you upload, and you can upload a ton of sources. I'm on a paid plan, and I can upload 300 sources. I did the math on this. I forgot, but it's something like, six 60,000,000 words or something like that.

Jordan Wilson [00:26:42]:
It's absolutely bonkers. Check check the newsletter. I'm gonna go, fact check myself, but it's just a ridiculous amount of content that you can put in there. Alright. So before we go on anymore, just think of that. Just think of that. Think of maybe how much time you spend using other large language models, because all the other ones work the same. And I said, even if you're using projects, GPTs, gems, etcetera, and you upload your files, none of them have this kind of upload limits.

Jordan Wilson [00:27:11]:
And also, even if you tell them in the custom instructions, hey. Only look at my files. Don't look at anything else. It's it's it's not reliable. Oftentimes, it still will look at its own training data or it will browse the web even when you don't want it to. So right there, just think of the huge advantage that you have in notebook l m. That you can only bring in your company's data, something that you're trying to learn, a hobby, a passion project, a side hustle, whatever. This is your partner that is essentially it's not a 100% hallucination free, but it is essentially hallucination free and always sighted.

Jordan Wilson [00:27:45]:
So when you, have, you know, 50 different sources on the left hand side and you're asking a question of the model in the middle, it is going to cite and you can click on it and see the exact, not just the source that it came from, but the exact spot in this, in that source. So talk about one of the biggest downsides of large language models in general. Well, they're kind of a black box. Right? And there's not always a lot of trust or transparency. And notebook l m, I think, flips that script on its head. Alright. So now we've got essentially the two main panels. So we have the sources on the left.

Jordan Wilson [00:28:22]:
You can add, you can discover, and then you also have the chat in the middle. And then the new updates that I just went over, that's the configure section in the chat where you can go, do the learning guide or the custom and then set the default length to default or or to the respond, set that, response length to default longer or shorter. And then there's, some little shortcuts at the bottom of the chat window, but those are technically just shortcuts to the studio. Okay? So the studio panel on the right hand side is actually pretty amazing. Alright. So that's where you kinda get the audio overview, and I'll just go ahead and, I'm gonna click an audio overview so you can see kind of the the options here. Because I think this is, a big reason why many people started using NotebookLM in the first place because it's really not just, a cool piece of technology. I'll actually say it's a transformational piece of technology.

Jordan Wilson [00:29:20]:
It's changed how I personally learn. I prepare for most of my podcast with NotebookLM. If I'm trying to learn new, something about a guest, you know, I don't wanna, you know, when a new feature comes out, I don't wanna have a 100 tabs open. It's it's hard. It's distracting. I love using notebook l m to not just learn in a personalized way, but to also interact, because you can, in these customized audio overviews, you have the deep dive, the brief, the critique, the debate, but you can also make them interactive. So I'm just gonna do, the brief. You can do custom instructions, all that thing, but I'm just gonna click generate.

Jordan Wilson [00:29:55]:
Alright? I'm not sure if I'm gonna, do this little tutorial long enough, but I'm gonna go ahead and click the same thing for video overview. So, we'll give it a couple of minutes. Like I said, I don't want the episode to run too long, but let's see what else we have in the studio. So in the studio on the right hand pane, it's just different types of multimedia that can personalize your learning. Because if you're like me, I I hate I hate learning from big blocks of text. That's not how I do it. Right? I love learning by listening, by seeing, by understanding visuals, but by actually interacting with it. Right.

Jordan Wilson [00:30:34]:
It's actually hard for me, which maybe is not good for me as a human being. It's actually hard for me to now, learn at the level that I know I'm capable, learn a new subject, learn about a person, you know, a new skill set. It's hard for me now to do it with just text. And I think, yes, you know, online has changed. Right? The Internet has changed. So much of what we see on online now is is video or interactive media. But I actually think notebook LM might be to blame in a good way for that. I've become very, accustomed to learning new things specifically from the audio overviews and the video overviews.

Jordan Wilson [00:31:16]:
Also, you have a mind map. So I'm just gonna go ahead and click and generate that. So now on the right hand side for our live stream audience, you see that it's actually generating in the studio. It's generating three things at once. It's generating the mind map. It's generating the audio overview and the video overview as well. Like I said, I went over these, in much more detail. Recently, you also have the reports.

Jordan Wilson [00:31:40]:
The reports are great. These are essentially, they save them they're saved as notes. Okay? So you have these default kind of docs that it will create for you. Alright. So a briefing doc, a study guide, a blog post, case study, key concepts, investment memo, or strategic analysis. And actually, the suggested format, so those last four I read, those are dynamically created based on the source material that you upload. Alright. So those are just simple reports.

Jordan Wilson [00:32:11]:
Maybe I'll just generate a briefing doc, then you have the flashcards. I'll I'm just overloading. I'm gonna crash notebook while I'm here, trying to do all these things on the live stream. What could go wrong? Well, none of them could work, and then I'm gonna do a quiz. Alright. So I'm technically generating now six different things in the studio all at once. It's just based on these two random sources. So hopefully, it wouldn't take too long.

Jordan Wilson [00:32:34]:
Alright. The good thing is is, well, we have the mind map done. So I can click on the mind map. Very cool. And it's also interactive. So it broke out, kind of these two unrelated sources. Right? One was about large language model news. The other one was about OpenAI's dev day.

Jordan Wilson [00:32:50]:
So they're kinda related, but kinda not. And it did a pretty good job of breaking this into a mind map, and then I can, click on these. And it's it's, you you know, putting out more and more, kind of nodes or modules. So now I'm, like, four layers deep. And then the cool thing is you can actually click on these and then have a conversation with that module as well. Alright. Let's go back to the studio. That's the mind map.

Jordan Wilson [00:33:14]:
Really cool. The quiz is done. It's giving me a quiz, just like that. According to the LLM market landscape 2025 report, what is anthropic's key competitive strategy? Oh, I didn't read that. Well, I can guess. Let's see. I'm gonna guess c positioning its models around a constitutional AI, methodology and a safety first narrative. That seems very much like anthropic.

Jordan Wilson [00:33:42]:
There we go. I got it right even though I didn't read the source material. So there you go. It's creating a quiz and then it's gonna give you feedback. So if you get it right or if you get it wrong, the quiz gives you feedback as well. That's the quiz module. Briefing document straightforward. Right? It just, these are just kind of, think of them as prepackaged templates for different types of reports that you can do based on your sources.

Jordan Wilson [00:34:01]:
And you can see how this is actually extremely useful, especially if you have multimedia sources, if you have m p threes, if you have YouTube videos, if you have PDFs, if you have, you know, web pages, to be able to bring all of this into a standardized and templatized, report is actually huge. Alright. Let's see what else we have done here. Then we have the flashcards. Flashcards cool. They're just kind of, visual flashcards. Right? You know, there's a front and a back. You read the front.

Jordan Wilson [00:34:31]:
So it says according to the 2025 market analysis, what percentage of consumer chatbot market does OpenAI, consistently hold? Alright. I I think it's in the mid seventies percent. Depends on how old this analysis is. The last one I saw was in the mid seventies. Oh, wait. I have to click see answer. There we go. Yes.

Jordan Wilson [00:34:51]:
I got it right. 74%. Now this maybe me means I spent too much time that a report that I hadn't read. I'm guessing things right. Alright. Those are the flashcards front and back old school y'all. I loved flashcards back. I was a geek.

Jordan Wilson [00:35:06]:
Yes. When I had free time, I did math flashcards. I memorized, you know, everything at a young age. Right? Okay. So the cool thing is is we also have our This is the brief on the rapidly evolving LLM market and OpenAI's latest strategic moves. So there's new analysis outright. It shows a super competitive landscape and transformative new tools taking large language models, you know, LLMs, the tech So there you go. We don't need to play the whole thing, but that was the the brief.

Jordan Wilson [00:35:36]:
So that one is very short. I chose that one so it would generate a little more quickly. If you do the, the deep dive, there's also this interactive feature where you can join the conversation, which I do all the time. Sometimes if I'm out taking a walk, I have notebook LM on my phone, and I'll just ask questions. So of the two hosts in the deep dive, I'm listening. If I don't understand something or if I want something rephrased in a different way, you can literally just join, quote, unquote, join the conversation. It's like you're calling into a talk show, and then you can ask questions and they'll answer it based on the source materials. Alright.

Jordan Wilson [00:36:11]:
So it looks like the only thing that we didn't get generated in time, which is fine, the video overview, but we've done shows on that before. Alright. So there you go. That is notebook l m in a nutshell. You add your sources. Your sources on on the left side, it's grounded. So when you chat with it in the middle pane, it's only gonna pull answers from your sources. It's gonna give you citations that you can click on and see the exact spot that that is pulling that information from to increase trust and transparency.

Jordan Wilson [00:36:41]:
And then in the middle as well in the chat, right now, it doesn't save anything. So any response that comes, you can click save to note, and then that note will be sent over to the studio. So keep that in mind, that feature is rolling out soon where it's gonna save, kind of your chat history. But for now, once you kind of refresh or leave the page, it's gonna be gone. So if you're working in the middle pane, and you get a lot of useful information, make sure to click save to note to save the note. So that is the middle and then the new updates that we went over, being able to, configure and customize the conversational style and the response length. Alright? And then on the right hand side, that's the studio. That's the multimedia, the audio, the video, the mind map, the reports, the flashcards, the quiz, and any notes you save are all over there on the right side.

Jordan Wilson [00:37:29]:
So that is notebook l m in a nutshell. Fast and furious complete walk through with all of the new, features and content. So I hope this was helpful y'all. We went over what's new, reportedly what's next, and gave you a complete walkthrough. So whether you're a beginner, or an expert, I hope that you gained some value from today. And, well, you probably just regardless should go repost this show on LinkedIn. So if you are listening on the podcast, check the show notes. I always put a link to this specific LinkedIn post, alright, where we put our livestream version of the podcast.

Jordan Wilson [00:38:08]:
So go repost this and I'm going to share. I put together this resource. I spent a ton of time on it, and it is absolutely bonkers good. I put together a lot of these guides when I I have so much extra information and I'm like, I can't make this a five hour podcast. What am I gonna do with all this research? What I'm gonna what am I gonna do with all this these high quality materials? I can't put in the newsletter either. I'm not gonna make you read that 800 page newsletter. So, this notebook l m cookbook, it is great different use cases across different industry, all the different, features, in in in, the utility taking full advantage of notebook l m for everyday business leaders. Go repost this show.

Jordan Wilson [00:38:50]:
I'll send it to you, you you know, in in the DMs on LinkedIn. It's amazing. Make sure you go do it. And then when you're done with that, if you haven't already, make sure you go to your everydayai.com. Sign up for the free daily newsletter. We'll see you back tomorrow and everyday for more everyday AI. Thanks, y'all.