Episode Transcript
[00:00:00] Foreign.
[00:00:05] Welcome to ChatGPT Curious, a podcast for people who are, well, curious about ChatGPT. I'm, um, your host, Dr. Shantae Cofield, also known as the Maestro, and I created this show to explore what ChatGPT actually is really, though, are the files in the computer, how to use it, and what it might mean for how we think, work, create, and move through life. Whether you're skeptical, intrigued, or already experimenting, you're in the right place. All that I ask is that you stay curious. All right, let's get into it.
[00:00:38] Hello, hello, hello, my curious people, and welcome to episode 25 of ChatGpt Curious. I am your grateful host, the Maestro, and today we are talking about the best AI tool that you've never heard of. I realize, folks, that's a bit of a clickbaity title.
[00:00:56] Not my usual style, but I'm hoping it hasn't deterred any folks from listening, because it's the actual and the absolute. That was like a bit of a both word. The actual, it's the absolute and the actual truth.
[00:01:10] This, this AI, this tool I'm going to talk about it is awesome. And my guess is that you probably haven't heard of it. So, first off, welcome to 2026. I realized that last week's episode was actually the first episode of 2026. And I pretty sure in that episode, you know, we celebrated, said welcome to the new Year. Uh, but it was also the first day of 2026. So, you know, we're like one week in, and it feels like it's like I'm, um, a bit more like we can actually say welcome to 2026. So on to the full episode, or the main point of the episode, the main topic, the tool we're talking about today is one that I have used a bit and was first introduced to me by my good friends tina Tang and Dr. Jackie Fenton, separately, separate occasions. So I, I bring them up because they're fucking awesome. But also to point out that this is not brand new.
[00:02:00] People, uh, just don't fucking know about it, Right? Just not enough folks, in my opinion, know about it, and I want to change that. So the tool in question is Google Notebook lm. If you've heard of it, if you're using it, congrats, you are ahead of, like, everyone.
[00:02:17] As a bit of an aside, I know this episode, or rather this episode. Wow. This podcast is called ChatGPT Curious, but I, uh, 100% always planned on using it to also just discuss AI in general. I say, I do say that on the website, um, irrespective of who the parent company is. So, you know, in future episodes, definitely expect to hear about Claude, namely Claude Code. Um, and just whatever I get into, I know I've, I've brought up Gemini. Rachel actually was one of the first people to be using Gemini. And I brought it up just because they put out their, their Gemini 3 model and Sam Altman was like, it is a code red for us. Um, so all that to say, I do plan on talking about more than just ChatGPT and, and I have, but this, I think, is the first episode that's like, fully dedicated to something outside of chat GPT and something that's outside of OpenAI's, uh, ownership. So thank you for being okay with that. So let's talk about Google Notebook. Google Notebook LM is an AI learning app.
[00:03:17] In putting this episode together, I was trying to find anywhere, literally anywhere on the website where it said, like, explicitly what the fuck it is, like, what is this thing? And I could not find a definition anywhere. So if you go to note notebooklm.google, or you can just Google Google Notebook LM at the top of the page that comes up, the information page, you will see that it says your research and thinking partner, grounded in the information you trust, built with the latest Gemini models.
[00:03:49] So still no real idea, uh, what it is, but it does say those things. Uh, if you scroll down to the FAQ section, you will see it compared to other things. And one of the questions, um, that's in the faq, in one of the questions that's in the faq, it uses the, the term AI learning app. So that's what we're going to go with, right? This is an AI learning app. The LM in the name, right? Google Notebook LM stands for language model. And like, it does say in that header, the language model that it uses is Gemini.
[00:04:27] So in general, the information page is honestly kind of bare bones if you compare to like, basically like any other information page that's out there about anything else. But I honestly don't mind. It gets to the point you're like, all right, cool. And then you just go try it.
[00:04:41] The coolest part, and perhaps the most telling part is like the full header. So whatever before is like the subheader, the full header. And it's, it's their tagline. It's in huge font, right? Huge sized font. And it says understand anything. I love that. I love that. That right there is at the heart of Google Notebook lm. And while I, why I personally think it is the best AI tool that you have never heard of. So the way that it works is you upload sources, literally any sources. You have. PDFs, websites, you put URLs in there, YouTube videos, audio files, Google Docs, Google Slides, literally, whatever. And it will create a learning resource for you.
[00:05:23] Worth noting. And this is what I've done before. If you don't have any sources to upload, you can have it search the Internet for you and find those sources and it will compile a bunch of them. And then you go through and you decide which of those sources you would like for it to use.
[00:05:38] And then it will create a learning resource for you from those resources or those sources that you selected. Like this is. It's mind blowing to me. Mind blowing. Now this next part, this is where it gets super cool. What type of learning resource can and will it create?
[00:05:55] Pretty much any kind you could want.
[00:05:58] So once you put the resource, the sources in, you have options. And you can check these little boxes as to what kind of resources, resource you'd like for it to create and the options. As of right now. What's, uh, today's date? Today is December 31st. I'm recording on December 31st. Uh, this will come out on what, January 6th? No, January. January 8th. I guess this will come out something like that. Yeah. I don't have the calendar in front of me, so it is past that date when, when you're listening to this. But the choices as of today are audio overview, which friends? That is a podcast episode between two co hosts. It's not real people, it's AI generated. Mind blowing. It's like a male voice and a female voice.
[00:06:40] You can have it create a video overview. You can have it create a mind map. You can have it create reports, flashcards, quizzes, infographics, slide decks, or a data table. That is fucking insane to me.
[00:06:53] Insane. I have only used it for a mind map, an infographic and a slide deck. But what it created for each of those was amazing. So I have no doubt that the rest would be just as good. I also know that Tina used it to create audio overviews, AKA the podcast episodes. Uh, and she told me she really liked it. And that's actually how I found out about Google Notebook from her. Um, that was her use case. That's like mind blowing to me. Mind blowing.
[00:07:20] So what makes this so awesome is that it can't hallucinate hallucinations. Remember? That's not, it's not spitting out wingdings hallucinations when it's when it spits out perfectly plausible sounding answers. Uh, but they're wrong, right? They're not based in truth. It can't do that because the answers it gives you are only pulled from the information from the uploaded sources and it provides citations for its work and it will show you the exact quote, the exact place that it pulled from this, that then the source that it pulled from. That's insane.
[00:07:54] That's insane. This thing is awesome.
[00:07:57] I personally really see the utility with what I'm going to call internally facing use cases, AKA when you use it for yourself and use it to learn things. If you want to summarize a long article or a long podcast episode I've done, I've seen this done actually quite a bit like in the online space, in the, um, social media space, kind of in the. It's been shared like largely, like from Twitter over to threads and things like that, where there's like big long podcast episodes that come out and people are like, I can't get through this thing. So I fed it to Google Notebook and then had it spit out out a, like a one sheet. And the people that were interviewed, the person that was interviewed in the podcast episode that they were people who were trying, like, I can't get through it was like, yeah, these are the main talking points. This thing is correct. Like it got validated. Like, this is insane to me. So you want to summarize a long article or a long podcast episode. Have you that? And I say, when I say summarize, not just like, you know, summarize it actually put it into a resource that you can learn from. So it could be a one sheet, it could be that it, uh, makes it into a mind map. I really like mind maps. Um, you have, you have so many options with this if you want to learn something and maybe you want to understand the components. Right? That's a good segue from mind maps because that's how I used it. I did this with some of the foundational concepts of computer science because I was like, I wanted to understand how the pieces and parts fit together. I'm very visual and like understand how this comes from this, this connects to this and it will make a mind map for you. And if you don't really understand what a mind map is, my suggestion is to go and make one. And so you can see, because it's not just like a pictorial thing, like you can click on the things, it's pictorial so you can see how each part is made above these other parts. But when you click on the, the parts, it expands it into like full text so you can learn about those things. It's, it's incredible. Incredible.
[00:09:37] So, you know, no matter what your learning style is, this thing can help you. And this, this concept really taps into the general excitement that I have around AI and how I think it can democratize so much. Just, just so much. And in this case, learning. Imagine what this can do for, for schools and, and for kids in school and all the different learning styles. Everyone can learn everything.
[00:10:01] That is amazing. You're studying for something you're trying to understand. I get it most schools and in colleges that you're trying to memorize stuff, but which this can still help you with. But if you're really trying to learn and understand, this is amazing.
[00:10:16] It doesn't matter how you learn. You're an auditory processor, your visual, whatever, it can teach it to you in that way. That's incredible to me.
[00:10:26] So the flip side of this, right, we talked about, what we just talked about was more of those internal facing use cases. So externally facing use cases. So like when you would use a submit and like distribute it, yes, you could absolutely use it to generate presentation slides, but as of right now, you can't actually edit those slides. So it's like a little bit like, uh, I tried doing this and the slides were really good, but I couldn't present from them because for me, the way my brain works, it's making the actual slides that like, that's how I practice giving the presentation. That's how I organize the things and that's how I remember the things and what I want to say, what I want to emphasize. So it's not helpful to have them be built. For me, this is actually something. So I used to teach for a company called Rock Tape. Perhaps some of you listening to this have been here, been with me since the Rock Tape days. If so, you're the best.
[00:11:11] Uh, but with Rock Tape they would give you the decks. And that was like one of the hardest things was being like, I gotta tell someone else's story and try to make it mine and like, figure out the timing of this. Um, so it's like, I can do it, but it's not ideal. I'd always much rather make my own. But for those of you that maybe don't care about that side of things and don't want to make your own, this could be a phenomenal, phenomenal resource. The only issue again, being you cannot edit it. At least now. Um, I foresee that changing, but as it stands right now you can't edit them.
[00:11:44] Okay. Um, but what it makes. Like, they're really good looking too. Like, my slides are always boring as. Like, I just have the words on them and I don't care. But for those of you that are like, I do care. I want to look better, this can be a very viable option, and I think will it will become an even more viable option. So, uh, what it was really good at that I did, in fact use it for, um, and I distributed, was making a one sheet and a one sheet, at least the way that I think of them, they just shares the most important takeaways of the presentation. So I had to make one for my chatgpt for Online Business owners webinar. And all I did was upload my slide deck and I asked it to make one sheet. And I gave it a few, like, stylistic instructions and, like, the kind of the color that I wanted. And the third iteration was a goddamn banger. So it. Yes, I did go through a few because I didn't know what it was gonna make at all. And the first one it made, I was like, holy. Like, this is really good. But it kind of.
[00:12:40] All the information was correct, but it broke, like, the words apart. Like, so I have an acronym that I came up with, and it, like, broke it apart. And I was like, that's not helpful visually. And so I just. When I re prompted it, I was like, you have to keep this together. And gave it the same, like, stylistic instructions as before. And by the third one, I was like, yo, this is insane. Like, I am just. I could never have done that. Just. I know I could not. I'm not that visual person.
[00:13:07] Uh, and it was just so good. And it had my colors in it. And then I just took it into Canva and I put my Maestro logo on it. And the thing is phenomenal. And it took my words and summarize them. It was so good, honestly. So, so, so good. Um, so I will never not use this thing to make one sheets. It's that good. But something to highlight here. And what makes it so good is that it uses Gemini, right? It uses a large language model. As such, it is phenomenal at identifying and identifying and abstracting patterns, which is how we can come up with the mind maps and these one sheets. And they're so good because it just pulls out the patterns and the things that are the most important or the things that keep coming up the most.
[00:13:45] So, you know, keeping with this. This concept of patterns, because this is their use case. One of the ways that I have yet to use Google Notebook because I didn't even realize you could do this until I read that information page. Trying to make this episode is that you can upload brainstorming ideas. So you put them in a, on a Google sheet and you can upload it.
[00:14:03] Uh, you can upload market research, you know, whatever, and you can ask it to identify trends or to generate new product ideas or to uncover hidden opportunities. Yes, you could do these things with Chat or chatgpt, but based on the experience that I've had with Google Notebook, I feel like it might produce better results.
[00:14:21] Um, I think maybe I spoke about this in the previous episode, in the last episode, but I'm not sure. Just that as we continue to use AI, I think our tech stacks will grow and we'll have, you know, the daily driver that we use for most things and then we'll have these kind of more specialized models that we use for other things. And this is a perfect case like using Google Notebook to make one sheets or to identify these patterns and uncover hidden opportunities. This may be the use case for this. And you really lean on, on this on Google Notebook as opposed to Chat. Um, but I, I can, I can I foresee this continuing to be the general trend in how we, how we use these, these language models. So for what it's worth, I do not really understand or know fully how Google Notebook works. I couldn't find it and honestly I didn't really feel like, you know, doing any more research on it. I am happy with my cursory understanding that it's a language model that provides responses solely based on the resources that it's even been provided. In my mind, that works largely the same way as when you make a custom GPT, right? The answers that you're looking to give it if you, if you remove the permissions of like Internet searching, is that it's only gonna respond based on what it's, you've, you've fed it and the files you've uploaded. So that understanding is good enough for me. And yes, because it's a language model, right? Uh, you can use natural language with it the same way you would with ChatGPT. You could just talk to it normally and it interacts with you in very much the same way. It's just, just incredible to me, folks. So the price, you can use it for free, it has a free tier. How good or bad that free tier is compared with the other tiers, I don't know. Right. If it's anything like ChatGPT, then you'd assume that the paid tier would be better, but I cannot say for certain because I've only used the paid tier not because I paid for it, but because I have a Google Workspace account. So if you have a Google Workspace account, the pro tier of Gemini and of Google Notebook is included.
[00:16:22] I think without that, it's like 19.99amonth. You know, most of these LLMs are 20amonth. Um, but they also seem to want to like, push you into their Google One AI plan, which is Google One. Google One is like extended cloud storage, which I also have because I don't delete anything. So that's like with your, um, personal. If you have a personal, like, Google Account, Gmail account, um, and then it has the AI plan added. So Google One would be basically adding Gemini, the $20 a month thing to, uh, your, your Google One plan. And then that would give you access to the Google Notebook Pro. You know, the naming isn't great for these things. And that's one of the things that I learned early on when I'm starting to get all into like, AI encoding, is that these people are incredibly smart, but, like, incredibly bad at naming things. And I'm like, what the fuck is what? If you folks are using Claude, then maybe you've experienced that where they're like, we have Opus and we have Sonnet. And I'm like, what is the difference between these? And even with ChatGPT, the numbers are kind of backwards. You're like, wait, oh, three, is that better or worse than GPT 5 or GPT 4? Like, not great at naming things. Um, but in order to avoid all of the confusion, my suggestion is go to the website and, and just look at it for yourself. Right? I have no influencer discount code or anything like that, but I really do hope that you'll go check it out. Um, because I think that Google Notebook is truly the best AI tool that you've never heard of. Except now you have heard of it. All right, let's wrap this up. How I use Chat GPT this week, and then we'll, we'll get out of here. So each episode I include a section where I briefly discuss how I use ChatGPT this week. If you're new to the show, this is what the section is. So this week I use ChatGPT, a literal metric shit ton, to help me vibe code my first web app. I made a pill tracker for my sister. Um, and I actually used a, um, an AI assisted code editor, code editor called Cursor and I paid for that. That's uh, $20 a month. Everything. $20 a month. Right. But I use ChatGPT for the planning phase. Uh, uh, and then I chatted with it a bunch throughout the actual build of the web app, namely to save money because basically you pay per interaction that you have with the AI that is integrated into Cursor. So I was just asking a lot of the questions to ChatGPT because I'm like, I'm already paying for you. I'm not gonna run up my tab on this side. Um, um, so that was dope. So, you know, in many ways we're gonna nerd out for a second. In many ways, I think you'd want, ah, something your model to be m. More deterministic when writing code. Right? Deterministic, meaning one input always the same output. Right. But we know these language models are probabilistic.
[00:19:11] One input, many different outputs just based on probabilities. Um, but when you don't write code at all, that's me, I'm not a coder. Uh, being able to use AI to write that code and to check it, that's one hell of a start. And I'm fully, I'm fully, I'm all in, fully on board with it. So first web app shipped, what comes next is making more of them. And I, this is definitely the route that I'd like to go down.
[00:19:39] Um, ultimately I think that there is opportunity to be had, namely, um, based on my interest and just what I see out there, there's opportunity to be had in basically like creating bespoke web apps for people. It's like, hey, you have this problem that you want solved in a specific way and you wish that there was a, a soft piece of software that did that. Okay, let's, let's build it. So, um, that is my, my focus moving forward. And I've been in the computer. I'm really, really excited about this and I, I'm really enjoying this stuff. So we shall see where it takes me. Um, I will of course share the journey and the things that I learn. I'll bring it into here. Um, like I said earlier in the show, you know, I, I call, I name this, this podcast ChatGPT.
[00:20:28] Curious. But I went with Chat GPT because it's a household name and largely for like, kind of searchability. But my, my goal and my plan has always been to talk about AI in general and all the different LLMs and, and all the different stuff that's out there. So as I get more well versed with these other things, I will definitely bring that into the show. Um, it's definitely a nerd out thing and there's some specificity with it, but I know that there's at least a handful the, in the audience that, that might be interested. So that time is coming and if you want to reach out before then and just nerd out or ask questions, I'm, I'm more than happy to chat. I, I'm just literally I'm eating, breathing, drinking this stuff. Um, and I'm very, very excited for, for what the next year year holds. Um, if you are interested in the pill tracker, I'm not going to put it on the in the show notes. Um, um, because I don't just like, want it to be out there. Mainly because, like, I don't really. I'm still like understanding the uses of it in terms of like, does it cost things? Like, so it's free to host it and the, the scaffolding, the framework that I use to build it, also free. Next JS is free. Um, but the hosting and the database are free. But I don't know if it's like, hey, if you get above a certain amount of usage, it goes up. I don't really know. I need to find that out. So I'm not like trying to put it on the web. Uh, but I'm really proud of it. And if you're like, hey, I actually would love that and that'd be helpful for me. Reach out, I'm happy to send it to you. Um, yes, you have to log in because it's the only way that you can like maintain your own information and be able to sign in from the, from your computer and also from your phone and like be able to see all the information. But, um, I'm really proud of how it came out. So if you're interested in that, by all means, hit me up. If you're interested in something else and being like, hey, could you make this thing?
[00:22:15] Hit me up right? I'm in the like free time of things. I just need reps. I got to figure stuff out. Um, it won't always be free. We know this. So that's, that's my goal is to turn this into a business. But these initial things, I just got to get reps and practice and learn. So if you have an idea, if you're wondering if you know something can be built, reach out and if it can, I'mma do it. And if I can't, then I can't. Um, but that we'll end it there. That right there. That is, is all for today. Uh, hopefully you found the episode helpful and, and I didn't get too, you know, too derailed with my, my excitement at the end. Um, but I'm, I'm in, I'm all in on this stuff. So if you did enjoy the episode, consider sharing it with someone who you think would enjoy it. Because like I said, I want everyone to know about Google Notebook. Like, I think it is that awesome. And I want everyone to be able to learn things and understand the things. Like this is. This is what my vision, at least a big part of my vision for AI is and how I, I see it being used and being so helpful and why I champion it. So share with the people. Don't forget I also have that companion newsletter, the Curious Companion. It drops every Thursday and it's basically the podcast episode in text format. So if you know that you prefer to read or you just want a written record of things, join the newsletter FM you can head to chatgptcurious.com forward/newsletter or check out the link in the show notes. As always, endlessly, endlessly, one more time. Endlessly appreciative for every single one of you.
[00:23:51] Until we chat again next Thursday, stay curious.
