11/26/2025 - A Deep Dive into Notebook LM's AI-Powered Infographics, Videos, and Slideshows

Transcript

1. 0:14Yeah. Ow.
2. 0:53I love you. Oh,
3. 1:23yeah.
4. 1:42Freedom came our way that night just like a jet plane in and out of sight. I was hauling ass at a million miles and
5. 1:59now I won't here. >> When they came into the station,
6. 2:13they said I was bad beyond repair, but I got no qualms with my situation. Say here I am.
7. 2:32So say Sheree Sheree Sheree won't you dare to say Sheree Sheree Sheree won't you dare to say Sheree Sheree Sheree.
8. 2:49Uhhuh. Yeah. Leave a message and your number. Please. Take a time to want to satisfy me.
9. 3:04Take all these old fantasies and send them care of me.
10. 3:33Come on.
11. 4:02There's been something, baby, I've been trying to say for an age, and it seems I don't know How with the past and the future now
12. 4:17surrounding me surren
13. 4:39of the rest I would have quit. met you long ago, but I couldn't do that. Oh, tell me now. Women and wine never went too well.
14. 4:56Make a man crazy, make him hold his hell.
15. 5:13with R. Good evening, good peeps. What is happening? Happy Thanksgiving Eve. I'm guessing tonight might be a lightly
16. 5:28attended affair. People are either traveling or with their families pretending like they still like them.
17. 5:34Listen, it is possible that there is someone out there that likes their family. That is absolutely a possibility. I haven't met them yet.
18. 6:17Heat. Heat.
19. 6:53He's This place I can rest my hobble head to gather my thoughts in sweet silence.
20. 7:10And this is the place where the feelings aren't dead. Running over exposure to violence. And is this a place I can slowly face?
21. 7:28The only one I truly can know. These are tears from a long time ago. Got these tears from a long time ago.
22. 7:42I need to cry 30 years or so. These are tears from a long time. Go.
23. 8:08What do you think, Champy? You a good boy? You a good doggy? Or are you fat? Are you fat? Are you big
24. 8:15fat fatty? You big fat fatty. You You a lot of dog. Is that a lot of doggy? It's a lot of doggy for that frame, isn't it?
25. 8:27Yes. The AI learning lab where fat shaming dogs is acceptable. You can you can you can take the Gen Xer
26. 8:40out of the 70s, but you can't take the 70s out of the Gen Xer. You know, I was I was watching the AI
27. 8:54learning lab the other evening. It was it was when you know when it was it was the night before Thanksgiving because I
28. 8:59remember how triggered I was. I got so triggered. You know, we're there. We're cooking all the food. We've got the
29. 9:05stuffing. And they call it stuffing for a reason. You know, we've spatchcocked the cir turkey, which, you know, every
30. 9:09time I say that, I feel so naughty. Anyway, we we did that to the turkey and we were making all the stuffing and we
31. 9:16were making the gravy and I don't know if you know this, but gravy is just full of fat. So, I'm already a little self-conscious.
32. 9:23And then and then this gentleman, I think he's the professor of the of the the college, the learning lab college. I
33. 9:30don't know what they call it. He he starts referring to his dog as as a fat a big fat fatty. I was so triggered. I
34. 9:38felt I I thought, "Oh my, is he judging me? Is he I I don't even know how to where to place them." I thought I came
35. 9:44here to learn about artificial intelligence and all of a sudden I'm questioning my frame. You understand
36. 9:50what I'm saying? It is not appropriate on any level. Any level whatsoever. What did that dog do to deserve that? How do
37. 9:57you think that dog feels? Self-esteem. Can't be good. Can't be good. And sure, maybe he's had a few too many chicken snacks. Chicken jerky
38. 10:07snacks. Dogs love those. Okay. And he likes cheese. It's a cheesel loving dog, apparently. It's not his fault.
39. 10:37Oh, buler. Buler. Buler. Oh, man. What are we gonna do tonight? What do you want to do, people? We could just talk. Does anybody have
40. 10:49any questions about artificial intelligence? Um, there's there's so much to explore. Um, Gemini was supposed to have gotten
41. 11:02interactive images. I didn't get them yet, but they look pretty cool. So you can just kind of dynamically generate um educational modules
42. 11:23dragonfly alchemy. What's happening? Good to see you. Sea slug of doom in the house. What do we got here? Usman,
43. 11:32Titan, Todd, ion on China, Steo, Tobias. Oh, that was my button for my shirt. That's kind of cool.
44. 12:09What entertains what what entertains the ADHD mind will always be a mystery. A button vibrating on a guitar string.
45. 12:22There's four hours of your life gone. Oh my god. If you're not neurosicy, you don't know what you're missing. If
46. 12:34you've got a functioning preffrontal cortex, [ __ ] must just seem normal to you. Oh, man.
47. 12:50All right.
48. 13:12Um Nate B. Jones, who if you don't follow him on Tik Tok, you should. He's super smart, wicked smart guy, really,
49. 13:20really consistent um content creator. about AI and he was talking today about um Gemini 4 not Gemini 4.5 Claude Opus 4.5
50. 13:34is the clear winner between Gemini 3 and Chat GPT 5.1 um which I don't he he had a very specific
51. 13:49measure he was doing it was some long context thing that had contradictory information in the middle of it. And so
52. 13:55it was a very specific test. Um, but the uh if you were here last night, we were vibe coding up some video game clones
53. 14:04from the 70s and and Claude Opus 4.5 was was pretty pretty remarkable. It did it did some pretty remarkable stuff. It
54. 14:12made some crap. Um, but I was I was not giving it good prompts. I was giving it really shitty lazy prompts which one of
55. 14:18the ways you test these models is give them lazy prompts and see if they do any good because if you if you write
56. 14:24incredible prompts you can get shitty models to do good things but that's not how most people use these
57. 14:30um for large token update. Yeah. What what he was talking about Mary Mary Mary oops is uh within within this he had multiple
58. 14:42sources in this document or in the inputs and in the middle of the sources there was there were two two pieces of
59. 14:52information that were contradictory and Gemini 3 and Open AI basically just missed that and just basically just
60. 15:02chose one and pres presented that as fact and opus 4.5 was the only one that actually recognized the discrepancy and
61. 15:10called it out. Um so he said it was a very it was a very clear winner on that particular thing. So so just so you know
62. 15:18there's there's what's called a U-shaped um response curve with with large context windows. So a context window is
63. 15:27basically what's the working memory? How much information can you have in a conversation before it starts forgetting things? Right?
64. 15:36So, Claude Opus 4.5 has a million token context window. So, it's a very large context window. It's got a very big
65. 15:43memory. But what happens in large context windows very often is they'll be very accurate about capturing the
66. 15:50beginning of the document, then they sort of tail off in the middle of the document. they miss a bunch of [ __ ] and
67. 15:56then the U-shape is they're good at the end of the document. Um, so that's one of the downsides of large large context
68. 16:04windows. So I think he was testing for that that dip um in in large language model performance and it and it sounds like um Claude 4.5
69. 16:16Claude Opus 4.5 was was pretty good at that. So, um, that said, you know, these models are getting the
70. 16:29the models are getting quite good. I I heard rumor today that OpenAI's got a new version of their image generator
71. 16:36image generator coming out that's supposed to be that's supposed to rival um, Nano Banana Pro. So, and then I also
72. 16:45saw today that Perplexity has now added image generation uh to their to their thing where you can make you can upload
73. 16:52a picture of yourself and and then put yourself in different im like everyone's doing everything. Everyone's copying everything. I it is
74. 17:01my contention that every major AI company is trying to make the everything app and everyone wants to be AOL.
75. 17:10If you don't if you don't remember ALOLL before the internet, America Online, it was America's little uh news center and
76. 17:19chat chat hole where where people could connect and form little clubs and groups before the internet existed. And it was
77. 17:28this very safe walled garden. You had to dial into it. You paid by the minute and once you were in, you couldn't really
78. 17:35get out. And then this rebel internet like the dark force came was the open internet and very slowly people started
79. 17:44bleeding out of open AI to the open internet. Uh and then you know eventually AOL just went away. But I
80. 17:50kind of feel like I kind of feel like all the AI companies are trying to create the single experience that will
81. 17:57give you everything you need. It'll give you social things and it'll give you shopping things and it'll give you
82. 18:03content things and it'll give you entertainment things and it'll give you Right. So, I don't know. I don't know.
83. 18:13The good news for us is that if everyone is trying to make apps that do everything, someone's going to succeed
84. 18:19and then once they do, everyone's going to copy them. So, we're we're all going to have really cool really cool
85. 18:25computers that just do [ __ ] for us, right? So, that's what's coming. And then what does it do? That puts it
86. 18:31back on us to figure out what we want to do. So if you if How many folks do we have here? We have 23. So it's mostly
87. 18:38people that know. But if you don't know, um you should join the AI salon, the salon.ai. Um and within that, we've got this area
88. 18:48called the AI Salon Mastermind. And within that, we've created a thing called the AI salon mastermind practice.
89. 18:55And it's a framework for designing a daily practice for yourself around AI. How do you use it? How do you think
90. 19:02about it? How do you learn it? How do you apply it? How do you have it make a difference in your life? Where you're
91. 19:09not putting the AI front and center, you're putting yourself front and center. Because here's here's how I like to think about it. If indeed
92. 19:18these apps are just going to do everything we need them to do. Ah there's a there is a there's the uh URL online. Although that
93. 19:29did blow up our vertical Oh no it Oh, I see what it did to our vertical thing. That's fine. The vertical thing is just
94. 19:34horizontal right now. Um, but if all these apps are indeed going to do all this stuff for us, so then what do you do? What do you do?
95. 19:49Where do you start? Which one do you learn? Why do you learn it? Which one's best? That's all secondary to what you're trying to accomplish.
96. 20:01And a lot of people, if you've been in a job for 20 years, especially if you've been in a job where you haven't really had to use critical
97. 20:09thinking, you're just sort of doing the work and you sort of punch in and p punch out and you haven't been asked to
98. 20:18exercise critical thinking and your taste and your ideas. You know, in some companies, having ideas is a culturally
99. 20:27dangerous thing. Kak, did anyone see the extent to which Facebook is changing terms and conditions on December 16th? I
100. 20:35have not. Let's go uh let's go ask Chat GPT to research that, shall we? We don't have to we don't have to read
101. 20:44no stinking articles. Um, please research um the extensive face
102. 20:59book changes to T and C. Um going into effect December 16th
103. 21:16tab. Yes, there you go. Here's a breakdown. What's official? What what Meta says is changing.
104. 21:31Data policy now renamed privacy policy. That's [ __ ] hilarious. Meta or Google talking about privacy policy is one of the funniest things
105. 21:47I've ever heard, but okay, fine. They're going to call it privacy policy now. Uh, in terms of service have been rewritten
106. 21:53and redesigned. The update includes more transparency about what type of data it collects, including data from people who
107. 22:01don't even have Facebook or Instagram accounts. Um, the updated policy gives more detail on what kinds of Oh, yeah. I
108. 22:07remember hearing about this. I do remember hearing about this. Um, yeah, this is sort of [ __ ] up and weird.
109. 22:13Yeah, I'll give you that. Yeah. Mhm. Sure. You're not wrong. You're not wrong to question the evil Zuck. The evil Zuck and his evil ways.
110. 22:27Meta is clarifying how location information may be collected, whether via device settings, IP address, or
111. 22:33other means. The update also clarifies how long Meta retains your information. Forever. Um, and under what circumstances they may
112. 22:42disable or terminate accounts, inactivity, security, concerns, violations. As of December 16th, one of the biggest changes, Meta will begin
113. 22:50using users interactions with its AI features, voice or text interactions with all Meta AI to personalize content
114. 22:57ads, recommendations across its platforms. Okay. So, anything you do, making pictures of uh of Yeah. just dancers from the 1920s was
115. 23:13now going to impact how what kind of ads you get served up. How did it know I like the roaring 20s? Meta claims this update does not change
116. 23:24how private direct messages between users are handled. I.e. normal 1:1 DMs are not being opened up for AI training
117. 23:34or ad use under this change yet. what it means in practice. If you chat with Meta AI about say hiking, that interaction
118. 23:44may be used to show more hiking related content, group ads, blah blah blah. I mean, this is what they always do. I
119. 23:52I I don't know. I I mean, this makes sense to me, but I mean, their their data policies have sucked for decades because Meta
120. 24:02will process AI chat data for personalization. Even non-account holders may be subject to some data collection. But how? How?
121. 24:17Even some non-account holders. You have to have an account to use Meta AI, don't you? Anyway, whatever. Okay.
122. 24:29Tab. Did I switch tabs? I I did switch tabs, right? We're good. Yeah. Okay. Then it adds more clarity around how third parties
123. 24:45how data is being shared or aggregated, how long it's being stored for users. There is no super opt out if you use
124. 24:51meta. The only way to avoid this kind of AI data usage is not to use AI features at all. Okay. There's been some confusion and
125. 24:59misinformation circ circulating. Meta will read all your DMs once the update happens. Official word word the claim is
126. 25:08false. The update does not change how private DMs are handled. What's uncertain or controversial? Some
127. 25:16third party writeups claim the changes grant Meta the right to use private. Okay, so that was the same thing.
128. 25:24Because the policyy's broader now includes data from people outside of Facebook, there's increased potential for shadow
129. 25:32tracking. Okay, listen. I understand the data privacy people being all upset about this. I don't disagree with that. Um,
130. 25:48this ship has sailed. This ship has sailed. I mean, it's not just Meta. It's every company out there is tracking
131. 25:58everything and they're all sharing data. There are data brokerages that just like all of our data. Everyone knows everything about you
132. 26:08already. I don't think this is a big deal. I mean, no, this is a big deal. Like, we should understand what it is.
133. 26:16We should understand it. But like basically what they're saying is we've got these cool new AI things and if you
134. 26:22talk into them, we're going to use the [ __ ] you talk about to serve up more ads to you. Okay, that's what they do.
135. 26:29Anyway, if you're on Facebook, I haven't been on Facebook in like two years now. Um, but anyway, that's my uh that's my hot take. My hot
136. 26:46take is, oh, Meta's being shitty with data. Shocker. Oh my god. I just don't care anymore as
137. 27:01long as they don't take my passport at the airport. I know theers. Is Facebook going to listen to my every
138. 27:11word? I'll have to fight my Fire TV for that.
139. 27:49Yeah, notebook LM. I guess we should go look at Notebook LM. Um, okay. I'm I'm I'm pre-thinking about
140. 28:06Thanksgiving weekend homework and
141. 28:25Jed X and Boomer artist only do Facebook and maybe Insta if they're really progressive. Uh yeah, I know. I mean, Facebook is what it is.
142. 28:43Um let's go to let's go to Notebook LM. What what I'm thinking about
143. 28:58let's do the arrestedellian logic motion and substance. So all right if you're new to notebook LM maybe we should make a new one. Yeah
144. 29:09let's make a new one. We'll make a new notebook. Let's do that. We're going to create a new notebook. Okay.
145. 29:19Discover sources. That's what I want to do. Search the web for new sources. Let's Let's look up.
146. 29:33Okay, here's what we're going to do. I'm going to find um find me articles and sources of um
147. 29:49the myriad ways that companies like Meta and Google capture and
148. 30:05exploit our personal data. Be sure to include
149. 30:20data brokerages that fill in the gaps that any given company might have on you.
150. 30:37This is going to be terrify terrifying have on you. Okay. So, let's let it let's let Google let's let go Google go
151. 30:46find all the sources that tell us how evil Google is. Let's let's see if if this is an actual search or if they're
152. 30:52going to censor this. Uh, no one's supposed to know that. Uh, don't don't give them that search return. Yeah. No,
153. 30:59throw that one in the uh that's in the special cases. Yeah. Throw that in the the special special attention uh
154. 31:07category. That would be terrific. That'd be terrific. Yeah. We're not going to those those partic what we've discovered
155. 31:14is the the websites that claim that we know everything about everyone. Those we've determined to be fraudulent. So,
156. 31:22we're not going to serve those up in the interest of protecting the safety of our customers who we value. We value them. I
157. 31:30they have a lot of value. I mean, we value their their uh safety a lot. D.
158. 31:48Uh, the unknown video man. Why you so unknown, Mr. Video Man? All right. What's happening here? Why is this taking so long?
159. 32:07All right. So what we're going to do, so what I'm doing if if uh to explain the interface here, the left hand column is
160. 32:15what's called your sources, right? So you can upload YouTube videos, you can upload songs, you can upload PDFs, you
161. 32:22can put websites in there. You can connect this to your Google Drive. So you can just aim Notebook LM at a Google
162. 32:30Drive. I think it's got a max capacity of 300 documents and like a ridiculous size like 500,000 tokens each or something like that. Uh
163. 32:41because I'm trying to hide from the data brokers, they're not paying me for my data. Exactly. Exactly. We live in a world now where
164. 32:52your data is worth something. So if you're hiding from it, you know, you're not you're not going to get your cut.
165. 33:01Um, it's the funniest thing about the the all the panic about AI getting all your data. I think here's the real the real
166. 33:11challenge or the real the real risk of all of our data being out there is that before generative AI, if you wanted to if you wanted to to
167. 33:25search through all of someone's data, you needed to be a kick-ass data scientist. And now you don't. You can just go learn [ __ ]
168. 33:37right? Source camp. what's happening. Um, and so it's just like just like AI is democratizing knowledge for everyone
169. 33:49else, it's also democratizing the ability to exploit data in in incredibly efficient ways. Um, so yeah, that starts
170. 33:57to become a problem, but but like people are acting like their data is not already out there. It's already out
171. 34:03there. You already gave it away. Do you use Google? Yes. Do you have a Facebook account? Yes. Okay. Done.
172. 34:18How long have you had those accounts? Since 20 minutes. Yeah, exactly. Why is this taking so long? Why is this taking so long? Um, let me
173. 34:32go I'm going to flip over to just do a Google search. Let's do um how it just finished. Did it? Did it really? Oh, yeah. Okay.
174. 34:49Fast research. So, we're going to import those. Okay. Great. All right. So, so this is cool. So, junk junk inferences by data brokers. Data
175. 35:02brokers call for transparency. FTC takes action against data brokers. Google hoards more personal data than you know.
176. 35:09Here's my meta and data. Okay. >> When you're done reading the tab, can you switch to >> Oh, that's really funny. That's
177. 35:16hilarious. Um, producer Brandon coming in coming in hot with the zingers. Coming in hot. Hey, hey there, Mr. Big Shot.
178. 35:26May maybe uh instead of picking on Facebook, you could stop acting like someone who still uses it. These [ __ ] millennials. These
179. 35:39millennials, they'll get you every time. All right. I assume you're a millennial. Is that right, Brandon? I think so. Um,
180. 35:49let me see. Elder Millennial. Okay, good. Um, okay. So, the way this works down the lefthand side, you've got your
181. 35:56sources. And in this case, we had Google go out and find a bunch of sources for us. So there's three PDFs and a bunch of
182. 36:02websites. We don't know what's in those websites. We don't really care because we've got Notebook LM. So this little
183. 36:07center thing here, this is now a synopsis of all of these articles. So the way Notebook LM works is you just
184. 36:15dump [ __ ] into it. Unstructured data, structured data, websites, just anything you want, you dump in there. It's it's
185. 36:25multimodal, so it can see videos, it can hear audio. It's pretty amazing. And then it now it just knows. It just
186. 36:31knows all this stuff is in here. Okay. So, data brokers, surveillance, capitalism, and digital risk. Let me
187. 36:38make this a little bigger because my eyes are old. They're old and tired. Good day, Mr. Shannon. Good day. The collected sources provide
188. 36:50comprehensive critique of the opaque multi-billion dollar data broker industry emphasizing the vast scale of
189. 36:57consumer data collection. This is going to [ __ ] freak me out. We like I knew we should all be paranoid, but now I'm
190. 37:04going to understand how paranoid we should be. And this is this is a good thing, right? Okay. When major tech companies like Google collect
191. 37:12wait while major company when while major tech companies like Google collect the most individual data points third
192. 37:20party brokers like Axiom maintain records on billions worldwide often synthesizing raw data into potentially
193. 37:29discriminatory or inaccurate junk inference. Right? So, this is the classic if they bought beer and diapers,
194. 37:38they were something. I don't know, a redneck. I I don't know what it was. There's some famous There's some famous
195. 37:45uh grocery grocery correlation. Um, these bad predictions can lead to practical harm such as elevated
196. 37:52insurance pricing, uh, discriminatory ad targeting. Yet, yet some data notably precise geoloccation tracking is
197. 38:01dangerously accurate, prompting recent enforcement actions by the FTC. The FTC and privacy advocates stress for Yeah.
198. 38:09Because basically every company that's willing to pay for it knows everything about you every minute of the day
199. 38:16including where you are right so this massive widespread broadcasting of personal information through ad networks
200. 38:24such as RTV real-time bidding further underscores the urgent need for heightened regulatory blah blah blah.
201. 38:29Okay. So so this basically just says what we asked for. We asked for go get go get a bunch of articles that tell us
202. 38:35about these things which we did. So now over here on the right there there's there's all sorts of things you can do
203. 38:41in the center section. I can just start prompting it. So I can say um give me a bulleted um comparison between
204. 38:56uh Facebook and Google's uh data collection uh practices.
205. 39:11And so I assume somewhere in these articles someone will have talked about that processing the material, checking your
206. 39:19files. Okay. And look, so so we're we're now instant experts in data privacy. Okay. Oh, look at that. Oh, that's what
207. 39:30was there before. Okay. Scope and scale hoards more personal data for Google than any other tech company.
208. 39:40has Facebook has reportedly reigned in its data collection accumulating only about 14 data points about its users. Yeah.
209. 39:52Number of data points. Google 39, Facebook 14. Um, types of data collected, contact details, photographs, documents. Yeah. A
210. 40:02content type photographs. So, all of your photographs, right? YouTube. YouTube comments, keywords, video views,
211. 40:12details about incoming and outgoing calls, Facebook extremely personal data, records of the people users talk to, the
212. 40:22groups they belong to, the comments that they leave, their private messages, primary business role. So anyway, okay.
213. 40:30So, so we now have a thing where we can just talk to our data, which is cool. But what's what's really cool is over
214. 40:36here on the right um there's all sorts of um pre-built ways to understand this data. Um the first one that they did was the
215. 40:50audio overviews. You've probably heard those. Uh so I'll just go ahead and and hit one. I'm not going to modify it in
216. 40:56any way. I'll just generate one. This will generate a podcast about these these things. But I can also make a video. And we'll
217. 41:05go in and we'll we'll we'll uh personalize this one. So, we can do an explainer video or a brief. Let's do a
218. 41:11brief, a bite-sized video. And then you've got different styles down here. So, we'll do retro. Oh, we'll do anime.
219. 41:20Anime looks good. And then what should this focus on? Um, make me a um video video that scares
220. 41:35scares the pants off people about how they have already given away their data.
221. 41:51and worrying about AI data security now is like closing the barn doors
222. 42:10barn doors after the horses escape. All right. So, we're going to now we're going to make a video so we can we can
223. 42:20guide that in a certain way. One of the other things it does is a mind map. These are really cool. So, we're going
224. 42:25to generate a mind map. Um, generate a report based on the sources. Well, you know what? We'll just do this. A briefing doc, a blog post.
225. 42:37That's kind of cool. Oh, wait. And it looks like it's generating specific formats. policy briefing, a risk
226. 42:43assessment, a concept breakdown, an explanatory article. Let's do let's do a policy briefing. Um, draft a comprehensive. Yeah, that's
227. 42:58fine. Let's just we'll do a policy briefing for the report. So, then we can do flashcards. We can do a quiz. And
228. 43:04then the two new things are an infographic and a slideshow. And both the infographic and the slideshow take
229. 43:13advantage of nano banana, the new I wouldn't even call it an image engine. It's a it's a visual reasoning engine. Um,
230. 43:25so we're going to make an infographic. So let's let's edit this and see what our options are. We can do a standard
231. 43:30one, a detailed one. We're going to do detailed. We're going to do it landscape. Guide the style, color, focus. We're
232. 43:37going to do um um let's make the style an iso metric um
233. 43:51Lego construction. Nah. Do I want that? That's kind of boring. Let's see. Let's do Oh, we'll do we'll do um
234. 44:05I want this to look like an adult um comic in the style of
235. 44:20Darkhorse comics of the 90s. I forget. Uh, I think Venom was one of them, right? Um, but whatever. We'll do
236. 44:34we'll do a dark comic one. So, that's going to be our infographic. And then we're going to do a slideshow.
237. 44:40And apparently these are pretty good, too. So, a detailed deck or presenter slides. So, we want presenter slides.
238. 44:48Um, and then here we'll use we'll do um use Lego metaphors. throughout and make
239. 45:03um explainer graphics um that leverage custom
240. 45:18Lego builds build Lego models. that bring the ideas to life.
241. 45:32Okay. All right. So, now we've got six different outputs that we're working on. So, here's the mind map.
242. 45:46So, there's data brokers. If you haven't seen these mind maps, these things are pretty sweet. Um let me if I go no okay
243. 45:57the nature and data nature of data and quality issues products and services data collection methods and scale. So we
244. 46:04click on that and then now within that there's collection technologies and industry scope but but oh no there's a
245. 46:14bunch of other things aren't there? Oh no yeah let's see we're inside data collection methods. So what it's doing
246. 46:20now is it's writing us an article about this node of the mind map. So every time you click on a node of the
247. 46:28mind map, it gives you an overview with citations from your from your data, right? So you roll over these citations
248. 46:38and it shows you the actual data that's underneath it. So this is from the the website data brokers, a call for transparency.
249. 46:50I love mind maps. They're neurospicy. Have you listened to some of these AI covers that bend genres?
250. 46:58You You mean the the musical ones? Yeah. The ones that are on TikTok all all over where they're like, you know, here's
251. 47:05here's gangster rap as as 1950s kuners. Yeah, those are pretty cool.
252. 47:19Okay, let me close this. So, what else do we have here? Now, we've got the secret. Is this This is our audio.
253. 47:32Welcome to the deep dive. Today, we're going to try and pull back the curtain on an industry that is uh well, it's
254. 47:39powerful, it's complex, and it's designed to be completely invisible. We're talking about a data broker economy,
255. 47:45>> right? These are the companies whose entire business is to quietly collect, package, and then sell your personal life.
256. 47:52>> And when you say personal life, you mean everything. Every time you shop, use an app, browse a website.
257. 47:58>> Every click, every purchase, every location, ping from your phone, it's all being gathered up, aggregated, and sold.
258. 48:04They're building a synthetic identity of you that might be more detailed than >> The other thing you can do with these
259. 48:10audio podcasts is you can make them interactive. So, they're going to talk here and then I can raise my hand and they'll they'll
260. 48:22let me ask them questions. Welcome to the deep dive. Today, we're going to try and pull back the curtain
261. 48:28on an industry that is uh well, it's powerful, it's complex, and it's designed to be completely invisible.
262. 48:36We're talking about the data broker economy, >> right? These are the companies whose entire business is to quietly collect,
263. 48:41package, and then sell your personal life. >> And when you say personal life, >> oh, hey, our listener wants to join in. What's up?
264. 48:48>> Yeah. Hey, wondering like isn't AI going to take all of our data?
265. 49:02Take all of our data. Restart audio. Welcome to the deep dive. Today we're going to try and >> Oh, hey there. What's up?
266. 49:12>> Hey, isn't AI gonna Well, anyway, there's something about my the way my screen
267. 49:27shared. >> Welcome to the deep dive. Today, we're going to try and >> Yeah, we we Vicki, they should lose
268. 49:31that. Welcome to the deep dive. They could come up with like three other openings, right? Pete M, what's happening? Welcome,
269. 49:40welcome, welcome. Does it call out Pal Palunteer or is it only about Facebook and Google practices? I I only mentioned
270. 49:48Facebook and and Google practices, but we can ask. Let's let's just ask um is Palunteer mentioned here at all?
271. 50:05Yeah, Palunteer just cut a deal with the with the current administration to listen to everything and provide reports
272. 50:12on us all. So, so yeah. So, yeah. All right. Uh, based solely on the sources, Palunteer is not mentioned. So, that
273. 50:20would be something good to actually Let's go, let's go do fast research. Okay, fast research. Search the web. Um, find uh articles
274. 50:35on the on Palunteer. How do you spell it? Yeah, I think I got it. Palunteer.
275. 50:52Um, find articles in Palunteer. Um, be sure to get ones that mention the new deal with the government
276. 51:08about um making US citizen profiles. We're we're just this side of uh you know, social credit scoring
277. 51:22here in the States. Except we won't call it that. Okay. Um, by the way, Pate, I don't know if you've been here recently, but I've been I've
278. 51:32been given lots of love to Google. Um, Gemini 3 and uh Nano Banana Pro are insane. And I've also been talking
279. 51:44up the fact that that was all trained on on TPUs and I've been uh been mentioning your work. So, welcome. I haven't seen
280. 51:54you in a bit, so it's good to see you. Um, okay. So, we've got the audio thing. Is this the video? No, that's the mind
281. 52:00map. Okay. So, here's the video. Here's the policy briefing. Is this the
282. 52:15Is this the infographic? I think this might be the infographic. Yes. Look at this. This is so cool.
283. 52:29Wow. Okay. Wait. Oh [ __ ] What did I do? I hit the back button. Are we still good? Let's see. Let's download this thing.
284. 52:44Okay, we just got it. This is very cool. Um, I'm gonna change my sharing so you can see what I'm gonna do here in a
285. 52:52second. I'm going to share all screen.
286. 53:08This is it, right? Yeah. Okay. All right, look at this thing. This is bonkers, man. This is flipping bonkers.
287. 53:26So So just to be clear, if you're new here, here's what we're looking at. We went to um we went to Notebook LM. We didn't have any anything
288. 53:42because Google's Google. One of the things you can do within Notebook LM is you can search for [ __ ] So, we were
289. 53:53talking earlier, someone asked a question about data privacy, and I'm like, oh, you know, they asked about Facebook and their their upcoming
290. 54:01changes that are about them using your interactions with their AI tools to fe feed their algorithm to serve you ads.
291. 54:10And so, I just kind of went on a rant about, well, they've like our data is already gone. They've already got
292. 54:14everything. So, then I was talking about data brokerages. So, we went to Notebook LM and just Googled had Google Google find us
293. 54:23articles and PDFs on data brokerage services. And so, it took I don't know five minutes or so, went and found a bunch of
294. 54:33sources and then now we've been interacting with it. And one of the new features in Notebook LM is an infographic. So, I said make me an
295. 54:41infographic about all the data that you found in the style of Darkhorse comics from the '9s. Right? So like venom and
296. 54:48and those sort of things. The shadow industry. Who are data brokers? A hidden multi-billion dollar industry that buys
297. 54:55and sells your personal data. They collect, package, and sell data without your knowledge. BTOC people search. B
298. 55:02TOC people search white pages, complex web data source from brokerage services. So anyway, so they're I guess
299. 55:10they're they're basically just scraping all of the data that they can get. Then too, the data machine, how they get
300. 55:17your information, harvesting public records, commercial sources, public info. Yeah. So, so Google and and Meta
301. 55:25and and other companies like that, they're only capturing particular data points from their services. What these
302. 55:32brokers do is they then take all that and compare that to all the other data in your life where your name might be associated. So then
303. 55:43they track your footprint. Real-time bidding is a massive data breach. Um the profile what they know and sell.
304. 55:50This is terrifying statistic. 178 trillion data broadcasts annually in the US and Europe. Your activity and
305. 55:59location is exposed 747 times daily. Hey AI nerds. What's happening to Shu? We're totally nerding out on the new um
306. 56:11infographic feature in um in Notebook LM. Um junk inferences, inaccurate goies. I don't know what that is. That's bad bad text.
307. 56:24Sensitive segments, discriminator discriminatory categories like expectant parent, diabetes interest, um urban
308. 56:32scramble, and royal everlasting. I don't I think those are just fuckups in words. All right. Um the damage the risk of
309. 56:41unchecked for surveillance, discrimination, predatory targeting, cremation services, government surveillance,
310. 56:50um physical safety, the fightback regulation, US laws, weak patchwork, right? Because it makes sense because
311. 56:57all the companies collecting the data want to keep collecting it because they can make so much money on it. Um and so
312. 57:04they therefore they pay politicians for that. Okay. Regulatory taking action PTC settlements. You can try to opt out a constant baffic
313. 57:15battle. I assume that means contact brokers individual. So anyway, and then there's a little notebook LM thing in
314. 57:22the lower right hand corner. But how cool is this that we can go from not knowing anything to having a decently
315. 57:33right? This is so fascinating. I'm getting ooze a eyes wide and then mind blown. That's good. That's how you should be. Okay,
316. 57:43let's go back to Notebook LM and let me share because we're going to watch the video because I told I want I
317. 57:52told it I wanted it to make a video that would scare the crap out of us. So, let's see how it did. Oh my god, people. Oh my god.
318. 58:06Oh my god. There we go. Fantastic. Okay, so that's the video. It's generating the slide deck. It's still going on the
319. 58:16slide deck. We heard the audio overview a little bit. We did the mind map. Oh, here's the policy briefing. Let's look at that while the
320. 58:27while the uh the deck is being created. The policy briefing. Um, introduction, the invisible and unregulated data economy,
321. 58:37the data broker ecosystem, collection and commodification, and then there's some sources in there. Oops, black bar. Sorry about that.
322. 58:54Um, the products of personal data. So, this is a whole education component. Very cool. Risk mitigation,
323. 59:07an assessment of key consumer risks and societal harms. So listen, for those of you So for those of you
324. 59:18that are part of the AI salon, you know that we, you know, we will occasionally take stances on if we think there's
325. 59:24shitty AI bills out there, we'll take stances on that. One of the things that we encourage is what what we call generous leadership in
326. 59:32the AI salon and some of that leadership looks like connecting with other people and teaching other people. Some of that
327. 59:37leadership might be advocating for the things that you believe in and that could be AI related or not doesn't
328. 59:44really matter. Um a tool like this, the fact that you can create a policy briefing on a bunch of uh a bunch of data sources. So, if
329. 59:54you're passionate about a particular area and you've got all sorts of interesting data and interesting articles and interesting points, you
330. 1:00:01could put them into something like Notebook LM and have it write you, you know, a policy briefing um and and
331. 1:00:09really help accelerate some of the things you want to do. So, that's pretty pretty [ __ ] slick. Okay, still making the slide deck. Let's
332. 1:00:16go watch the video and then we'll look at the look at the slide deck. So, play go full screen. You can all see that
333. 1:00:30>> your data, it was sold a long time ago. Let's get into what that really means for you. >> So, is AI coming for your data? It's the
334. 1:00:38big question on everyone's mind, right? But here's the thing. The real threat, it isn't new at all. It's been around for years.
335. 1:00:47>> Uh-oh. What happened? Oh, I resized the window. It didn't like that. Is AI coming for your data? It's the big
336. 1:01:00question on everyone's mind, right? But here's the thing. The real threat, it isn't new at all. It's been around for years.
337. 1:01:10Way back in 2014, an FTC report blew the lid off this huge hidden industry. Meet the data brokers. Their whole business
338. 1:01:17is collecting and selling your personal information. It starts with stuff that seems harmless. You know, your age, job,
339. 1:01:24what you buy. But how they see you is way more intense. We're talking shockingly specific labels. As one official put it, they build a story
340. 1:01:33about your past, present, and even your future. And this whole system, it's called surveillance capitalism. Your
341. 1:01:39life is literally their product. And man, is it profitable. Just nine of these companies rad in hundreds of
342. 1:01:46millions back in 2012. So, how do they know so much? Easy. They connect what you do online with your real life. It's
343. 1:01:53called onboarding. It's how your grocery run turns into those targeted ads you see later. They'll tell you it's for
344. 1:01:59stuff like fraud prevention. But the real world risks, they're actually terrifying. And this isn't some whatif
345. 1:02:05scenario. Just this year, they got caught selling super sensitive location data. Think about it. Trips to the
346. 1:02:11doctor, your place of worship, even domestic abuse shelters. So that biker label for ads, it could easily become a
347. 1:02:18high-risisk label for your insurance. Which leads to the big takeaway. Worrying about AI now, that's closing
348. 1:02:24the barn door after the horses are gone. Look, the data is already out there. It's being sold every day. So the real question is what now?
349. 1:02:34H All right, that was pretty slick. Um, very good. There's a comment in here about um deep research is something I
350. 1:02:40use all the time. So, in in notebook LM, how do I get out of here? There we go. In notebook LM. Um, why am I just on this page? Hang on a
351. 1:02:54sec. Oh, there we go. Okay. Uh,
352. 1:03:09okay. over here on the left hand side on on the the data sources you can what's going on why am why am I not able
353. 1:03:20to search here import Import or delete results try deep research okay so there's a there's a pulld down menu here let me.
354. 1:03:40Okay, there's a pull down menu right underneath the little search bar here where you can do fast research or deep
355. 1:03:48research. So, if you put it in deep research mode, it's going to go out and do, you know, really significant
356. 1:03:54research. So you can populate your notebook LM with with really really good research. Um, all right. So the last thing we're
357. 1:04:03waiting for is generating the slide deck. So that's still going. So anyway, any questions, thoughts on notebook LM.
358. 1:04:12And I'll I'll keep an eye on this. I'll come back to this. Slide decks take the longest in my experience. Yeah, they're taking
359. 1:04:19forever. The company best known for the Neielson ratings is generally considered a data broker. Yep.
360. 1:04:28I mean, you know, we'd be naive to think that they they haven't been doing this, but I guess that 2014 finding was was a big deal
361. 1:04:40that it really exposed that whole industry. Um but but my thoughts here whether you you're doing this inside notebook LM or
362. 1:04:53you're doing this at AI Studio um for Google or you're doing this in the Gemini app. The Nano Banana image generation tool
363. 1:05:06is a visual reasoning engine and all I know is that it's more than an image generator. I don't know. I came up with the term a
364. 1:05:17visual reasoning engine. Well, I don't know if I came up with I didn't come up with that term, but that's what I'm
365. 1:05:22calling nano banana that that because it's truly multimodal. When you ask it to make an image with nano banana, very
366. 1:05:30often what you're getting is large language model reasoning steps happening before it generates the image. And so
367. 1:05:38what it means is you can have it do things. Um Ethan Malik showed this. I think it was Ethan Malik. Um he took a scan of a homework test
368. 1:05:49um question that was like a math question and it and he said to Nano Banana, "Take this this homework page
369. 1:05:57and and answer the question like solve the math problem with handwritten like pen marks." and it filled in the homework page with pen marks
370. 1:06:12because it ain't just making an image. It's doing all this reasoning before it makes the image. And then I assume on
371. 1:06:19the back end it's giving this it this remarkably long very very specific um instruction set to make the image
372. 1:06:28that's correct for the for the math problem. So the reason I keep talking about this and the reason I think this is important is that
373. 1:06:41this is such a new capability that there's going to be people that get really good at how to leverage that power in some interesting way.
374. 1:06:54Are you going to be one of those people? You could be if you're like, well, I don't know anything about images or I don't know
375. 1:07:02anything about data or I don't know anything that that doesn't matter. What you know about is what you know
376. 1:07:09about. So, you might be really good at a particular subject matter um an area of expertise where you're going to use that
377. 1:07:16tool in a way that someone else wouldn't. Black bar. Sorry about that. Right. And then the other thing that that you
378. 1:07:31know last night was a fun night. We went and we played with all these. Do I have any of them up? What's this one? Yes.
379. 1:07:45We went and we made a bunch of video games using Claude Opus 4.5. And so here's this was this was a oneshot Asteroids game. And if you're
380. 1:08:00like, what's oneshot? Oneshot means in a single prompt that was basically like I want you to create a clone of the classic Asteroids
381. 1:08:10game. I want it to have the right physics and look and feel and I want it to I want it to be aesthetically,
382. 1:08:16you know, like I'm playing the game in the 70s. And so it created this thing like notice notice it's got the ring the
383. 1:08:23phosphor burn of a CRT ring and and sitting in here is a burned in the word asteroids is burned into the monitor.
384. 1:08:32And if I start playing this thing, if you're old, you remember paying
385. 1:08:47quarters for this. And that should give you anxiety like it just did me. [ __ ] [ __ ] UFOs. every [ __ ] time
386. 1:09:00and be like, "Fuck, [ __ ] fuck." I think that's I think this is the game where I learned to swear. God damn it.
387. 1:09:13So, anyway, this was created with I didn't have to go back and forth 20 times to get this to work. It just
388. 1:09:22made it. I bet you can't beat Pong. I can't cuz the pong the pong one's too fast and I don't have a paddle.
389. 1:09:30But if I did have a paddle, I could just vibe code paddle controls, right? Um, so I would argue at this point with
390. 1:09:42Claude 4.5, Gemini, Gemini 3.1, and ChatGBT 5.1, they've all got vibe coding modes, right? Um,
391. 1:09:57Gemini in particular, if you go to aistudio.com or no, if you go to uh Vicki taught us this, if you go to ai.dev,
392. 1:10:06that'll take you into the the Google's vibe coding tool that uses um Gemini 3. And if you know nothing about coding, that doesn't matter.
393. 1:10:22It doesn't matter. You're going to be able to make an app. If you have an idea for an app, you're going to be able to make an app that's
394. 1:10:31some version of functional right instantly. Now, is that app ready to ship? Is it a scalable, bulletproof,
395. 1:10:40highly secure app with authentication and e-commerce built in? No. And you know, could it be? Probably. But you're
396. 1:10:49going to need to know some expertise to get it there. But learning to do this now, learning to say, "Oh, just just
397. 1:10:56knowing that even if you're not a coder, you can talk an app into existence. You can talk an infographic into existence.
398. 1:11:04You can do explainer concept creations kind of instantly. For most people, these are going to be very, very new skills.
399. 1:11:20And some of the capabilities here have never been capabilities in human history before. The fact that you can just go
400. 1:11:27say, "Huh, it would be nice to know about how the Egyptians got clean drinking water and send deep research
401. 1:11:36off to solve all that." And within, I don't know, half an hour have a film about it, a presentation about it, an infographic that you could
402. 1:11:46put on LinkedIn and say, "I've been doing some research, and here's what I put together." People will pay you to do that [ __ ]
403. 1:11:54but you got to know that it's possible. And you've got to get your spin on like like what are creative ways that
404. 1:12:01you could put together infographics that people have never done before? I don't know. That all comes back to you. Like what's
405. 1:12:11what is it that you care about? What is it that you're good at? What is it where are your values? You know, when producer Brandon created
406. 1:12:19the the custom GPT for people that might be losing their SNAP benefits in his values are empathy for people. And so he coded that
407. 1:12:32in such a way that not only did it give people information, it gave them some compassion and some empathy.
408. 1:12:40Right? So functionally what it did there there were probably a lot of people that made custom GPTs that functionally gave
409. 1:12:48similar information but his did it in a very Brandon kind of way. That's the thing ultimately that's going
410. 1:12:59to make things rise above the noise. If everyone can do everything then all of the crap that's just created
411. 1:13:07without thought is going to be the new bottom. Right? So there just going to and no one will pay attention to
412. 1:13:12anything that gets made but there will be some things that rise above that noise and those are the things you want
413. 1:13:20to be associated with. What's going to rise above the noise is stuff that is personal to you because you've got a
414. 1:13:27unique view in the world and you're going to use these tools in a way that no one else will if they're not thinking
415. 1:13:32about it. If they're not filtering it through their values. So, I'm saying all this to say that I think over this weekend and you know
416. 1:13:42coming up to Festivus, we've got we've got AI festivist coming up uh December 26th and 27th. So, it's it's
417. 1:13:512 days of 24 hours of programming. So, 9:00 a.m. to 9:00 p.m. on Friday the 26th and 9:00 am to 900 pm Pacific on Saturday the 27th.
418. 1:14:05Every hour is a different presentation of an AI practitioner just volunteering their time to teach you this. It's free
419. 1:14:12to everyone. So go to aifestivist.com right now. Scroll to the bottom. Sign up. Tell your friends. If you're
420. 1:14:22with your family, if Uncle Jim is still being a blowhard and he's a little drunk and he's telling you that AI is going to
421. 1:14:33take the jobs and the robots are going to kill us, just have some empathy for him and sign him up, too. Bring bring Uncle Jim
422. 1:14:42to Festivus. Um but but you know in this next month like this weekend if you've got some time and cycles and you want to
423. 1:14:50think about AI I would start thinking about these new capabilities visual reasoning. What does that mean?
424. 1:14:58What does it make possible? What does it make possible for you? Imagine taking here. Here's a fun one. Oh, I got a good one. Hang on.
425. 1:15:12We live in a time now. Let's say you're home with your family and Grandma Stella is uh she's not doing great, but she's still
426. 1:15:24she's still mentally there, but she's not doing great, but she's got really good stories to tell. Um, take out your phone, go to the voice
427. 1:15:37recorder and put it on and just interview her. Just sit down with her. Just put the phone down. You don't even have to tell
428. 1:15:45her you're recording. If you're in Colorado, you don't legally have to tell her, but you can tell you're recording
429. 1:15:50or not. just put the phone down and just have a conversation with her and have her start telling you stories of when
430. 1:15:55she was a little girl and how she met her husband and what your parents were like and just have her talk.
431. 1:16:02It used to be so so there was a time not that long ago and I I built a whole company around it. My current company
432. 1:16:09StoryVine is built around the fact that unstructured data and especially unstructured video data is really hard
433. 1:16:17to parse, right? That's no longer the case. AI can just ingest anything and understand it. So what that means is you could just sit
434. 1:16:28down for an hour and interview Grandma Stella, interview Uncle Jim, interview your parents, interview your siblings,
435. 1:16:36just get them all to start telling stories. Tell stories. Tell stories. Tell stories. And then voice recorder. I don't know if
436. 1:16:43you know this, but when you record a voice on the on the iPhone, and I assume Android's got the same sort of thing, it
437. 1:16:50it's instantly transcribed. So, you have a transcription immediately. So, you could take all the transcriptions or the
438. 1:16:56raw voice files, quite frankly, and throw them all into Notebook LM. So, by by Saturday evening, you could
439. 1:17:04have all the recordings of all the family members, including the kids. [ __ ] it, do the kids, too. throw all those transcripts into
440. 1:17:13notebook LM and then go do like an infographic of the family history. Why not? Or go turn it into an application where
441. 1:17:26you can choose a family member and then see all the different stories that they that they told, right? You could do and and you could
442. 1:17:35even because of Nano Banana, you could create a your family last name um children's book creator that creates
443. 1:17:46children's books out of your family stories that you all just dumped into Notebook LM and it organized them for you.
444. 1:17:56Cindy Coon's in the house. She So, if you don't know Cindy [ __ ] she's a futurist. you got, you know, she's a futurist. And
445. 1:18:09what she's realized is that there's there's a new kind of future there. There was the future. This is really [ __ ] up. There was the
446. 1:18:19future that was coming and now now we're living in a future. And so anyone if you're showing up to this show regularly, you're actually
447. 1:18:29living in a different future than most people on the planet, including old futurists, right, who are sitting sitting on this side of
448. 1:18:39the AI wall looking at the future. And Cindy is over here on the other side of it in the middle of this AI stuff going,
449. 1:18:49I've got to completely rethink my mindset for what it means to be a futurist. It's like the thing I was just talking
450. 1:18:57about, like by Sunday night, you could have a movie, characters of your family, all of your stories documented. Um, turn it into a video
451. 1:19:08game, turn it into an infographic that you send to everyone. You don't have to talk to them about AI. Just say, "Hey, you know, when I
452. 1:19:18recorded you over the past couple of days, here's what that turned into." And do your little slideshow. It'll melt their [ __ ] faces.
453. 1:19:28How'd you do that? Well, you know, all you make fun of me with AI. I did it with that. Oh, that's kind of cool. Yeah. Uncle Jim,
454. 1:19:40why don't you have another scotch there, you big lush? So, we're going back to the future. Yes. Yes. Source camp. We're all Marty now.
455. 1:19:58That's pretty good. Let me go see if uh if our slideshow is done. Yes, it is. Wait, is it? Yeah. All right. Share this tab instead.
456. 1:20:12Is this it? Okay. Building a U that isn't you. So, so I
457. 1:20:27asked this to make a um a a presentation style slide deck that used Legos as the uh as the as the visual whatever.
458. 1:20:44Um that's a pretty damn good title slide. inside the chaotic chaotic world of data brokers. Kavuno, I get to meet
459. 1:20:52Ethan Mllik in real life next month. Oh, that's often my boss is bringing him in to talk to our company on AI. That is incredible.
460. 1:21:02Um, do you need Do you need a plus one, Chris? Invite me. I would love to be there. That's really cool. Congrats. That's
461. 1:21:12that's super exciting. Um, okay. How do I do this? There we go. The internet thinks I'm I'm a Southeast Asian single mother of two.
462. 1:21:23A major data broker profiled Ariel Garcia, former chief privacy officer that way, despite her being child-free,
463. 1:21:30married, and of an ethnicity which has no connection to Asia. This is junk inference made real. Um, who are the builders on the factory
464. 1:21:41floor? first party broke builders, the official Lego sets, tech giants like Google and Meta get billions of data
465. 1:21:49points bricks from you and then the third party brokers, the miscellaneous brick sellers. This is actually really
466. 1:21:57good. Companies like Axiom and Experian have no direct relationship with you. They collect, buy and trade loose bricks
467. 1:22:04from thousands of commercial and public sources. Where do all the bricks come from? purchase data, observe data, and public
468. 1:22:12data. And then they go into the data factory. Look how good this is. The informed guess machine. So those two bricks say
469. 1:22:28bought something, visited a travel blog. Okay, so that's your activity. They go into the machine and and out comes a
470. 1:22:36conclusion about who you are. We want to be as accurate as possible, but our inferences, they are all informed guesses.
471. 1:22:47More bricks equals more money. Look at all the Lego sets on the wall. How cool is that? We should get a summary of AI
472. 1:22:55Learning Lab recordings by year and do infographics of what you've covered. That's a super cool idea. We can do
473. 1:23:02that. We've we've got the data. In fact, um, Digital Gods has sent me a spreadsheet with all my transcripts in it.
474. 1:23:14So, yeah, we could certainly do that. If I'm doing Yeah, we could certainly do that. That's that's that's a super fun idea. Um, I've
475. 1:23:29seen the same individual be placed in a likely Republican donor audience segment, but also a likely Democrat donor audience segment.
476. 1:23:40Your profile, their profit, marketing, risk mitigation, people search, a foundation of junk bricks.
477. 1:23:55Advertising audiences are often inaccurate. Oh, you know, you know what's amazing? So, because of these data brokers, so
478. 1:24:03one is they've got your data wrong or they've got all your data. The other one is they're making these inferences that
479. 1:24:09that are not necessarily right. This is saying somewhere in the neighborhood of 50% are inaccurate. We'd analyze how accurate this data was
480. 1:24:20and it was half half wrong all the time. So if it's half wrong all the time, 50% wrong. There's two things happening in the
481. 1:24:30advertising world right now. I talked about this earlier in the week. The there's two things happening in the
482. 1:24:35advertising world. One is the media landscape is so fragmented now. There's no center of gravity for eyeballs,
483. 1:24:42right? When we had three network channels, all the eyeballs were watching those three stations. Now we're at the
484. 1:24:49point where, you know, I think it was this year that people were watching YouTube more than they were watching television
485. 1:24:57or Tik Tok or something like that. There was there was some there was some threshold was crossed that that more
486. 1:25:02video was being consumed on the internet than on TVs basically. Um, so you can't find the people. And then
487. 1:25:13what this is saying is even if you pay these people to find the people, most of the time it's wrong. And and so where we're headed is you're
488. 1:25:23going to need to build trust with your consumers. So what what it's looking like is rather than brands going
489. 1:25:30to advertising agencies to make ads to sit next to content that people enjoy, the brands themselves are just going to
490. 1:25:38make the content themselves. So there's a big shift coming in where the power is in content creation and relationship
491. 1:25:47building. Um, Netflix just is doing a deal right now with is it 600 blogs or or podcasts? Like Netflix is starting to do doing do
492. 1:26:01deals with podcasters because the podcasters are more trusted now than the TV shows. Did you see the LinkedIn comment? I did not. Jim Ross,
493. 1:26:12seriously, knock off the gym [ __ ] Let's go with Uncle Fred or something. Listen, Jim. Listen, Mr. Ross, listen. Mr. Ross,
494. 1:26:22seriously. Uh, seriously, put down the scotch and soda, sir. Sir, sir, please. For the sake of the children, put down
495. 1:26:29the drink. Would you like another turkey leg? What's up, Jim Ross? Hope you're doing well, man. Happy Thanksgiving. Um, one
496. 1:26:40brick that's scarily accurate. Geoloccation data is scarily accurate. The great Lego spill 747 times a day. On
497. 1:26:49average, a person in the US has their online activity location exposed 747 times a day through real-time bidding. A
498. 1:26:57process the Irish Council of Civil Liberties calls the biggest data breach ever recorded. Look at these slides. How good they are.
499. 1:27:07This is amazing to me. This is amazing. The factory is facing headwinds. Now imagine doing something like this where
500. 1:27:15you actually understood the subject m like if Jim Ross took all of his data. I don't know when you came in Jim but we're um
501. 1:27:24we just did a random search. We went into notebook LM created a new notebook and just had Google do the search on uh
502. 1:27:31data privacy and data brokers. And so it went and found a bunch of sources. So imagine Jim if you take all your sources
503. 1:27:37of data all the stuff you've created your marketing materials your methodologies whatever blah blah blah
504. 1:27:43throw that into notebook LM and then go through all these new things like this one is the slide the slide maker and
505. 1:27:49it's also got the new infographic maker which was absolutely amazing and just spin up different different presentations
506. 1:28:01absolutely amazing but the cookie jar was glued back together. In April 2025, Google backtracked on its plan to phase
507. 1:28:08out third party cookies. Citing a shifting regulatory landscape, the king of firstparty builders protected its
508. 1:28:14business model of the third party builders. The old system remains. Oh well, the missing instruction manual. Transparency access and control
509. 1:28:25meaningful opt out. Yeah, that's See, I don't think any of this is happening. Taking control of your bricks.
510. 1:28:32Limit collection. Check and tighten privacy settings on apps. Yeah, that's not daunting. Every [ __ ] app you have that says,
511. 1:28:42"Do you want to opt out of this thing?" Knowing your location data, do you always opt out of that or do you
512. 1:28:48sometimes hit the wrong thing? Like, did I opt out or opt in? Wait, which one was it? Unbelievable. From surveillance capitalism to digital
513. 1:28:57self-determination. Well, here's here's a here's an opportunity, but here's also something maybe promising about AI. Um,
514. 1:29:08someone using AI could absolutely make a system that in real time goes out and finds all of your data, all of your
515. 1:29:18settings on all of your apps, and you give it some general guidance, and it will go out and just start cleaning that [ __ ] up.
516. 1:29:27Is that your 10-person team? That's pretty good. That's awesome. So, I'm working on a book right now called
517. 1:29:3110erson team. I should have done Lego characters, although I'm glad I didn't because then I would get sued. Um, the
518. 1:29:38goal is not to stop building, but to ensure that each person is the architect of their own digital identity with the
519. 1:29:43power to choose the bricks and design the final model. Yeah, that's cool. All right. Well, that was pretty [ __ ] impressive, I gotta say.
520. 1:29:54Pretty impressive. It took a while. What' that take? 15 minutes, 20 minutes, but damn. Now, are these things editable? If I
521. 1:30:05download this, what's it downloading it as? PDF. So, they're not editable, but let me see if they're if they're uh is it is it data or these
522. 1:30:17all just images? I think they're all images. The button to the left. Wait, I got to go here. Go here. Open it in here.
523. 1:30:31Yeah. No. Oh, no. That's text. So, you could you could get the data out
524. 1:30:46of here. You could re you could redo these if you needed to. All right. Cool. Oh, I'm not sharing this, am I?
525. 1:30:54>> No, but if you go back to the um the slide you were on where you downloaded it. >> Yeah. >> You should be able to click the button
526. 1:31:03next to download. >> Oh, wait. How do I close this? Where? What? Oh, I'm not in.
527. 1:31:17Hang on a sec. Show prompt. >> Show a prompt. Um, so you should you might be able to share it to Google Slides as well. >> Oh no, maybe not.
528. 1:31:34>> But still you could extract the data. >> Yeah. >> Oh yeah. Copy link to slides >> where? >> Oh yeah. Oh yeah, that's just copying
529. 1:31:42the link to the slides. Never mind. I I digress. But hey, since I'm here, can I talk about my book? Yes, please do jump up on stage here.
530. 1:31:50Let's get you. >> So, yeah. Uh, so I'm I'm super excited. Um, the 10,000 minute mindset is this is the
531. 1:32:01only copy that exists right now. Uh, it >> it hits Amazon on Sunday, which is the three-year anniversary of ChatGpt.
532. 1:32:11>> Indeed. And if you have been one of my advanced readers, um you have the opportunity to buy it for 99 cents even
533. 1:32:18though you already have a free copy. The reason for that is Amazon cares more about um verified reviews or verified
534. 1:32:28purchase reviews. So if you So that's the what I've learned. I've learned a ton through chat GBT of researching how
535. 1:32:34to do a book launch. By the way, if you are an avid reader and you don't know about advanced reader copies, there are
536. 1:32:41entire Facebook groups where people just give their books away for free in exchange for reviews on Amazon. Um, so
537. 1:32:47if you ever want to read and you don't you're not near a local library and you want to read something that's brand new
538. 1:32:53and something that somebody wrote, um, look up ARC Arc Readers uh, on Facebook. But this is the the synopsis of the book
539. 1:33:02is from Malcolm Gladwell's 10,000 Hours. uh and it is the assumption that yes to become a master of a trade you need
540. 1:33:12about 10,000 hours to be functionally competent in something you need about a week which is 10,000 minutes so this is
541. 1:33:21just about finding that niche finding that lane and using AI to accelerate and just start moving and learning to shift
542. 1:33:29your relationship with learning into something that is more functional versus theory. >> Love it. >> And uh it is available on Amazon Kindle
543. 1:33:39and then um after it goes live, I'm going to have 11 Labs read it and then I'm going to have a audible version of it. >> Nice. >> Very cool.
544. 1:33:48>> If you still want a free copy, hit me in the salon. Uh I've got the PDF version available. >> All right, producer Brandon, not just
545. 1:33:55the I can do the operation. >> Congrats, dude. That's really >> That's really >> um bef uh before we get out of here, you
546. 1:34:08definitely should tell people about Mastermind and AI Salon presents next week as well as what the heck you're
547. 1:34:15going to be doing in New York and if we're even going to be here next week. >> Okay, good. >> Happy Thanksgiving everyone.
548. 1:34:21>> Thank you Producer Brandon. Um okay, so let me do those in kind of kind of order that that Brandon talked about. mastermind.
549. 1:34:32If you're a member of the AI salon, even if you're not, you should also be a member of the mastermind. So, the
550. 1:34:38mastermind is the subscription area of the salon. Um, through the end of the year, it's 20 bucks a month and you
551. 1:34:46retain that rate moving forward so long as you retain your subscription. Um, starting January 1, that's going up to
552. 1:34:5347 bucks. So, um, time the time is the time is good to get in if you want to get into it. Now, why would you do that?
553. 1:35:01We launched two weeks ago what we're calling the AI salon mastermind practice. And if you're a member of the
554. 1:35:10mastermind, you can join us on in the um mastermind practice lab, which is a weekly meeting on Thursdays at noon Eastern
555. 1:35:20um where everyone in the practice lab is designing their own daily practice centered around AI. And the whole idea
556. 1:35:27is the tools are going to keep changing. And so if you're constantly chasing tools, it's you're going to get a
557. 1:35:35exhausted and b frustrated because you can't keep up. Having a daily practice is really about checking in with
558. 1:35:42yourself and figuring out from an intentionality standpoint, what are you trying to do in the world? Who are you?
559. 1:35:47What are your values? What are you what what are you um trying to accomplish? Are you trying to make money? Are you
560. 1:35:54trying to, you know, uh, improve your community? Are you trying to to do things for your family? There's there's there's no wrong answer.
561. 1:36:05The the there is a wrong answer. The wrong answer is how do I use AI too? If you start with AI, right? But if you say
562. 1:36:14here are the things I want to do in the world, now that you know that you want to do them, you say, okay, what AI tools
563. 1:36:18do I need to be able to do that specific thing that's really relevant to me? That's what having a daily practice is
564. 1:36:24all about. And having a daily practice, one of the things that I'm getting in touch with and doing it is that it's not
565. 1:36:29all just Shangrila like, you know, unicorns and rainbows. Part of a daily practice is sitting with the uncomfortableness
566. 1:36:39of holy [ __ ] it looks like my entire sector is going to go away because of AI. What do I do? and like sitting with
567. 1:36:48that uncomfortableness and like I don't know anything about programming but I'm supposed to vibe code something that's
568. 1:36:57terrifying to me. Yeah. Sit with that terror. Like part of part of navigating this new world for all of us is getting
569. 1:37:07better at adaptability and getting better at curiosity and getting better at at raising your own game and and
570. 1:37:14having taste and and and figuring out what you want to do and what your point of view is. And what I can tell you is that the
571. 1:37:22people that have naturally done that with AI have absolutely transformed their lives. And so the whole point of the daily of
572. 1:37:31of the mastermind practice is we've created a framework that lets you really look across all different aspects of how to
573. 1:37:39design a practice, figure out what's right for you and then and then do that. So that's that. Okay. Next Tuesday is AI Salon Presents. So
574. 1:37:49it's our our monthly meeting. Um, we've got Danny who's an entrepreneur from Denver is going to be speaking and he's
575. 1:37:56going to be talking about um what he's done in AI, what he's done in saving uh some venues here in Denver. Um he bought
576. 1:38:06up he bought for example the Mercury Lounge. It was about to close and and be demolished and he bought it and and
577. 1:38:12saved it. Um, and he's a really fascinating guy. And he owns a space in Denver here called ID345, um, which we're going to make an
578. 1:38:23announcement about, uh, next Tuesday. So, I'm really excited about that. So, the AI salon is going to have a home
579. 1:38:29here in Denver. Um, and we will we will make some some uh some announcements about that and you'll get to meet Danny.
580. 1:38:35So, that's that's going to be really fantastic. Um, okay. Next week I am leaving the day after Thanksgiving for New York City.
581. 1:38:45We're going to hang out in the Bronx with the Mrs. family and then um I'm going to be in New York City. I'm meeting we're doing AI salon um
582. 1:38:57strategizing strategery um on Tuesday. a meeting with Andy Scarantino and we're going to be, you know, designing 2026 basically what
583. 1:39:07we're going to do. And then uh and then the end of the week I am meeting with two producers who are interested in
584. 1:39:17talking about how do we get Sydney off the page and onto the stage. And so fingers crossed that that goes well. So
585. 1:39:24I will keep you posted. But what that likely means is that next week may be chaotic in terms of schedule.
586. 1:39:32So here's my commitment. I'm going to one of the things I'm going to do tomorrow is I'm going to um I'm going to design
587. 1:39:47the kind of content I want to put on Tik Tok. So, I've been I've been not all that active on Tik Tok recently
588. 1:39:56in terms of generating content. I'm I do these things, but I I haven't been generating content. Um I'm going to
589. 1:40:03spend some time looking at all of my projects and figuring out what I want to talk about with those when, you know,
590. 1:40:09over the course of a week. Um and and so I'm going to start I'm going to start making that content next week. So, um,
591. 1:40:17hopefully from the streets of New York City. So, that should be fun. Um, holiday homework. Go play with the new
592. 1:40:25toys in Notebook LM. Go play with Gemini 3 at um in the AI studio or in gemini.google.com. Go play with the new Claude 4.5 um, Opus
593. 1:40:38and make apps. um just just go experiment with these new tools because the new tools are really quite remarkable. They're just they're of a
594. 1:40:48different nature. Um so I would go do that. And then the final thing that I want to say tonight is I've realized that I do not
595. 1:41:06I would like to dramatically improve um my ability to recognize remarkable people that that do a lot of work for me
596. 1:41:17um without recognition or without pay or things like that um the mods that mod on Tik Tok and that mod on um on YouTube
597. 1:41:32regularly on a nightly basis just do it without um me acknowledging them or thanking them. I'll sometimes thank the
598. 1:41:39mods, but I really want to say to all of you, to anyone who has ever or continues to um be a mod in here, welcome people into
599. 1:41:50the community, answer questions, support me, cheerlead me, um deal with trolls on my behalf. Um, I want to say I have I
600. 1:42:03have deep deep appreciation for you and you should know that one of the things I'm working on in my life right now is getting better
601. 1:42:20at accepting gifts in the form of service and compliments and things like that. I'm bad at it. No,
602. 1:42:34I'm not bad at it. I have historically been resistant to it. Um, and so, so part of that is it looks like I take
603. 1:42:45people for granted that do a lot of work here. Um, I don't take you for granted. I really appreciate it. So, I just want
604. 1:42:51to say that. And, and producer Brandon, I'm gonna do something horrible. If you're still there, I'm going to pull
605. 1:42:57you up on stage. Um, if you could turn on your camera if you're if you're there. Um, I just I just want to thank you as well because
606. 1:43:05>> Oh, no. >> Producer Brandon has been doing this job, this thankless job of just telling me black bar, black bar, tabs, tabs,
607. 1:43:13tabs. He's been doing it for more than a year. And and you know, we we said we set out to do this. We're you know,
608. 1:43:20we're going to try to get audience and make money and do things like that. And you know, listen, making money on
609. 1:43:25content is not an easy road. So, so that hasn't happened yet. So, you show up here night after night after night,
610. 1:43:32except for the fact when you take a nap with your kids and your alarm clock doesn't go off. But other than that, um
611. 1:43:38even when you've sick, you you've been here. So, I just want to say to you for just being here for for the time that
612. 1:43:45you have, I just want to acknowledge you and thank you because it it really does mean a lot and I really feel supported. So, so thank you.
613. 1:43:52>> Of course. It's a labor of love. >> Yeah, it's great. and I was watching anyway. So, >> yeah, exactly. That's great. Well, the
614. 1:43:58feelings mutual and so thank you. And so, thank you mods. Thank you everyone. Yes. Uh and and >> I also want to extend the thank you to
615. 1:44:05the mods. uh because for all the times I'm shouting black bar uh and monitoring YouTube and Twitter and LinkedIn and all
616. 1:44:14the other things and making sure that you're sharing the right tab, it's really easy to miss the Tik Tok pins. Uh
617. 1:44:21and so and the Tik Tok comments and stay engaged in that audience. Um so I just wanted to say thank you to the extension
618. 1:44:28by extension. Thank you to the bots. >> Yeah. Yeah. Yeah. Exactly. Beautiful. All right. Well, thank you sir. you know, um so to everyone,
619. 1:44:39even if you're not if you're not a mod, um if you just show up here, and you know, it's funny. Um I spoke I spoke
620. 1:44:48with um with Chef Kelly's on her clean food panel two weeks ago. Um, and then I I was I was in another thing and
621. 1:45:00I was I was doing an interview with Anne Murphy and and one of the women that we were interviewing came on and I hadn't
622. 1:45:05met her before and she said, "Oh my god, it's like I know you." And she's been watching this show for like years,
623. 1:45:13right? And she's she's she's one of the quiet watchers that doesn't, you know, that doesn't pipe up. You know, the
624. 1:45:19irregulars are are are, you know, a rowdy bunch. Um, but I know that there's a lot of people that watch this show
625. 1:45:25regularly that may not speak up so much. Um, so for anyone who's taken the time to even watch this show once, much less
626. 1:45:34night after night after night, I I want to thank you as well. Um, I think one of the biggest things I learned about doing this
627. 1:45:46was the the line came from Cindy [ __ ] I don't know if you're still here, Cindy. This was this was a year in. And she
628. 1:45:51goes, "Kyle, do you know what's happening in the AI in AI learning lab?" I said, "No, what?" She goes, she goes,
629. 1:45:57"You know, it's not about you, right?" She goes, "It's about the community." Like, you create the space, but there's
630. 1:46:05this whole other world happening while you're talking and doing your thing. There's this whole other thing happening. So for all of you that
631. 1:46:12generate that energy and create kind of this vortex of possibility and generosity and empathy and intelligence,
632. 1:46:28um, thank you. Thank you. because, you know, this would be hard to do if it were just, you know, putting it out and and,
633. 1:46:39you know, knowing that it's just an audience that's just sort of walk watching and passing by. It's absolutely
634. 1:46:44not that. It's absolutely a community. So, for all of you, I really appreciate that. So, go off and and uh spend some
635. 1:46:51good time with your family. Um, eat some good food and know that I I deeply deeply deeply appreciate all of you. All
636. 1:46:57right. So, with that, peace out. Um, homework for the weekend is just go play with these tools. Just do that and
637. 1:47:03connect with your family and think about who you are, what you want, what your values are, because ultimately that's
638. 1:47:10more important than any of these tools. And then next week will be a little chaotic, but we'll have some fun. All right.
639. 1:47:19I wish there was a way to capture all of the conversations and relationships here. I know, Source Camp, it's crazy,
640. 1:47:24isn't it? It's really good. Happy Thanksgiving. Yeah. Yeah. Yeah. All right, groovy man. I appreciate y'all. Happy Thanksgiving.