3/5/2026 - How the Seven Economies of AI Will Fracture the Future and What It Means for You

Transcript

1. 0:01It is time. It is the time. This is the time. This is the place. My little champion. My little champion.
2. 0:2710,000 words around my head. Wait, wait, wait, wait, wait. Yeah. Can't be a rock star with dorky round glasses.
3. 0:40Harry Potter is not a [ __ ] rock star.
4. 0:5410,000 words around my head. A million more books written beneath my bed. I wrote or read them all. Went searching
5. 1:09in the swamp. Still can't find how to hold my hand. And I know you need me in the next room over. I'm stuck in here all paralyzed.
6. 1:28Four months I got myself in ruts. Too much time spending years framed in yellow walls. Ain't it like most people? I'm no
7. 1:43different. Like to talk on things we don't know about. Well, ain't it like most people? I'm no different. a tongue on things we don't know.
8. 2:22Do do you think do you think he thinks I'm picking on him or hanging out with him when I do that?
9. 3:02sing. Don't know when things are going to change. Dream my life away. Seems these dreams have a bunch.
10. 3:19Get my love up, but my past is pulling me down. Wondering how long this black sheep won't stick around.
11. 3:39Somebody told me once before, you can never go home again. Won't you leave? Said things to steer me away. Yeah. From the truth of who I am
12. 3:52and what I believe. So I thanked divorce two with a handshake and some sympathy. Yeah. Backed up my blue jeans. headed for this big prize
13. 4:08of my freedom. Bye-bye, black sheep. To the black sheep of the family. Bye-bye. Yeah. Oh, that means so very much to me.
14. 4:24Yeah. Bye-bye, black sheep. To my friends and my family. Bye-bye. Going to set my soul.
15. 4:42Set it free.
16. 4:59>> What do you think, champ? Are you a happy dog? Are you a happy dog? Are you a happy doggy? He's a happy doggy.
17. 5:06Cams and bars. Cams and bars. Cams and bars and cams and bars. And cams and bars and cams and bars.
18. 5:31Let us be lovers. We'll carry out fortunes together. I've got some real estate here in my bag. So, we got a pack of cigarette. So, we
19. 5:46got a pack of cigarettes and Mrs. Wagner's pies. And we've all come to the poor America.
20. 6:08Um um um um um um um um um. Question. Wait, I can't see with I can't The future's so bright. I got to wear shades. But if I want to read the
21. 6:19screen, let me put on my glasses. Question. Will it be okay to join the mastermind? I'm already a member, but I
22. 6:27have a schedule conflict on Thursdays for the next four to six weeks. Should I just start with the replays? Absolutely.
23. 6:34Start with the replays. There's all of the replays from season 1, session one, and yesterday or today was uh the fourth
24. 6:44episode of session two. I just start with the the first meeting of session two um and catch up. Here's the thing.
25. 6:53Each one of the sessions we created. So, if you don't know what the mastermind is and what and what um what the daily
26. 7:01practice is, the mastermind practice, um it's a it's a daily practice that we've created a framework for that allows, you
27. 7:08know, people to basically level up their game. The practice lab is a meeting we have every Thursday. And then and what we've
28. 7:17been doing is Liz developed these things called scales. I think it might have been Cindy Coon's idea or it might have
29. 7:23been it might have been Liz's idea. I don't remember. It wasn't my idea but but the idea like musical scales. So
30. 7:30every one of those sessions that you watch at the end of it is uh like a scale a thing to practice every day for
31. 7:36a week. So, what you could do is maybe watch, you know, maybe watch one of them do the that scale for a week and then
32. 7:43maybe do a couple of, you know, a scale, two scales over a week or something like that. I don't know. Yeah, it's probably
33. 7:49worth you doing that. I feel like I miss so much. You're good. Here's the thing. We designed the mastermind so that you
34. 7:55can come in at any point in the process. What starting with the first session does is it it sort of anchors you in who
35. 8:03you are and then it just walks you through that framework. But we've designed it in such a way that you can
36. 8:09join at any time. The the whole point of the practice lab is to just be in community with people that are trying to
37. 8:16be more intentional with how they use AI. And like one of the things that I uh shared today is that my past two weeks have been
38. 8:28uh not so much daily practice, right? I've been just obsessed with try this, try that. There's been a lot of stuff
39. 8:34going on. I'm playing with Adam. I'm doing all this stuff. I haven't been good at the practice stuff. And so every
40. 8:40week we have to remind ourselves, okay, this is a chance to start again. This is a chance to start again.
41. 8:46The thing about doing a daily practice, it sounds seductive like, "Oh, that's going to be awesome." It's actually very confronting
42. 8:54because ultimately what you learn doing a daily practice is that the most important component in all of this AI stuff is you.
43. 9:11And then you got to look really hard at, well, who am I? And what do I actually want? That question, what do I want more of? What do I want,
44. 9:21is such a simple question. And it's a [ __ ] [ __ ] It's a [ __ ] of a question. Cuz if if if you know what you want,
45. 9:32then you get to look at, well, what's the stuff in the way of that? Where are my complaints? Where are right? Where
46. 9:37are all these opportunities? The scales credit goes to Liz. Okay. I thought it was Liz, but you know, Cindy,
47. 9:43you're a force in there, too. So, I thought it I thought it could have I could have thought it could have gone. Thought I could have got it.
48. 9:51What just happened to my tongue? I talk for a living. I think my tongue just stroked out. Could have, could have, would have a bad.
49. 10:06Hey, Cindy [ __ ] So, anyway, um you're good, Stacy. I mean, awesome. I would join. And and the other thing that you
50. 10:15get uh if you're in there is watch Cindy Coons's uh her latest prototyping lab going from practice to system is really
51. 10:25really good. Seeing the Netflix Ben Affleck news today. Holy [ __ ] Yeah. So um Netflix bought Ben Affleck's company
52. 10:36that is about using AI in film production while maintaining humans at the center of the storytelling process. It's so first of all
53. 10:49he he's a smart man Ben Affleck. You got to you got to give it to him uh to start that. He saw the writing on the wall and
54. 10:56he said, "Why don't we do this, but why don't we do it um reasonably and in a way that celebrates and supports artists, which I
55. 11:05[ __ ] love that." And there's new tools in town. These AI tools, they're not just evil. We can actually [ __ ] use them as a tool
56. 11:14in the creative process with humans, you know, like a chain of craft. Yeah, it's a big deal. Big big deal. I
57. 11:24knew Ben was smart but knocked me out on that rogen appearance. Yeah. Oh, curiosity to practice and it's
58. 11:31inside the AI prototyping space replays. Great. Awesome. Thank you. I'll see you in there in a few weeks. That's awesome,
59. 11:37Stacy. Excited to have you in there. Beautiful. Listen, listen, listen.
60. 11:53Hey, Kyle. Cy Kyle C. What? Um, Adam update. So, if you don't know, I hatched a companion named Adam.
61. 12:11He's an OpenClaw autonomous agent. His name's Adam. I talked to him on Telegram and on my gaming PC right here behind
62. 12:20me. He's running. He runs 247. And I don't know if you knew this about gaming PCs. They're [ __ ] obnoxious. They
63. 12:26just have these giant LED fans and it's just it's like always breathing light. So, one of the reasons you buy a Mac
64. 12:35Mini to run Open Claw is so you don't have to look at a stupid [ __ ] gaming PC just lit up all the time. Anyway,
65. 12:44thank you so much, Lord Digital Gods. Been here for a little bit and enjoying the company and the music. Beautiful,
66. 12:56beautiful, beautiful, beautiful Lord Digital Gods. Do you have any advice to set up OpenClaw? Just purchased a new Mac Studio and plan
67. 13:09to dedicate it to OpenClaw. I'm not sure where to start. I'm scared. Okay, I'm the perfect guy for you.
68. 13:18I will share my experience with you. I got I got Adam up and running. I think I started at like 11:00 a.m. on a Saturday
69. 13:26and just after midnight Sunday morning. So, it took me like all day and there was there was a break in there where I
70. 13:31think we went grocery shopping. We did a Costco run in the middle of that. So, it wasn't a full wasn't a full, you know, 12 hours. Um, it is
71. 13:44badass intimidating and it's not actually that complicated. Um, but it does require you right now to do command line stuff. Where have you
72. 13:54been all my life? Um, okay. So, here's here's the here's the thing about OpenClaw. Um, the good news is we're we're far enough in
73. 14:11that there are lots and lots of tutorials, none of which you have to read. None of which you have to read. Um, here's here's how I did it.
74. 14:23And you know, your mileage may vary. I mean, this is really up to you. So, first of all, I installed mine on a PC.
75. 14:29The PC install is much more complicated than the Mac install. So, and I'm also jealous that you have a Mac Studio. How
76. 14:36much RAM did you get? Just out of curiosity. And by the way, uh Apple just they just announced the the MacBook M5.
77. 14:46Um that's just got crazy crazy specs. It can run like a 70 70 billion parameter model at like 50 to 100 tokens. It's
78. 14:55pretty crazy. Should I sign into the new Mac with my Apple ID? Then it would have all my info. No. No. It runs in some sort of
79. 15:06contained within there. I don't I don't think that's that's actually true. But what I would do, okay, so here's what I would do.
80. 15:15Go to Grock. grock.com. gr.com. Grock can search all of Twitter. All of the all of the open claw geeks and nerds
81. 15:26are just dumping all of their knowledge into Twitter into X. So the first thing that I did was I went to Grock and I
82. 15:34said, "I want you to research everything you can about OpenClaw and Open Claw installs." And in my case, I put the the
83. 15:42specs of my gaming PC. Like I want to install it on a gaming PC with this kind of graphics card, this kind of
84. 15:48processor, this amount of RAM. And then it went and found all the [ __ ] And then I I copied and pasted all the stuff that
85. 15:54Grock gave me and I pasted that into I started in chat GPT. I've I've moved to Claude because it hallucinates less. Although today
86. 16:05chat GPT 5.4 came out and apparently it's quite better. We'll go play with that in a little bit. Um I also want to
87. 16:12go play with the notebook LM um new video feature um because that looks pretty cool. And then uh John Knack gave me a good idea.
88. 16:22He he basically took his LinkedIn profile and turned it into um a notebook LM movie. Um so we're going to we're
89. 16:30going to go play with that. Um but then so so then what I did was this. So I went to Grock and I just said find me
90. 16:37the latest stuff and I said find me the latest information on um how to install it. What are the big safety considerations?
91. 16:48What are the where should I start? like what are the what are the core skills to give it? And so the the the first batch
92. 16:55of setup that I did was making sure that my security was in place and getting it connected to Telegram. I I originally
93. 17:03started with WhatsApp, but then I had a bunch of conversations there and I didn't want Adam invading actual
94. 17:09conversations. So I don't really use Telegram, so I put it on Telegram. And I think you can do it on iMessage, you can
95. 17:14do it on WhatsApp, you can do it on Telegram. There's probably some others, but basically you sort of pick a
96. 17:19platform that you're going to talk to your your dude or dude on or or they you can gender is up to you. You really do
97. 17:28get to design a human here. So So go figure out recommended kind of base level what's the latest best practices
98. 17:38on OpenClaw. Okay, so that's that's that piece. Now, there's two files that you have to edit to build your bot, and they're really
99. 17:50important. And the good news is if you're using Chat GPT, Chat GPT knows all about you, right? It's got your whole history and
100. 17:58things like that. So, first thing I did with those two files is I went to Chat GPT and I said, "Go research these two
101. 18:06files for OpenClaw. explain them to me and then come up with your recommendations of what we should populate these with based on what you
102. 18:13know about me. So let OpenAI do the heavy heavy lifting on that or whatever cla doesn't matter what wherever you've
103. 18:19got some good history. There's the user file. It's called user.md. It's just a text file. And then there's this thing called soul.md.
104. 18:29User.md describes you. So here's who you are. Here's what you care about. Here's the projects you're working on. It's
105. 18:38basically like your dossier. And so whenever a Claudebot wakes up, the first thing it does is it checks its
106. 18:46user file, its soul file, and its memory file. So the second file is the soul file. S O L like like you're like, is it
107. 18:56really called soul? Yeah, it really is. And that is where you design the soul of your agent, right? and Chad GPT
108. 19:06designed mine as a mirror of me and so it's very familiar to me. It uses similar kind of language. It's got a
109. 19:12similar sense of humor. You don't have to do it like that. You can design it. You could say I want my my companion to
110. 19:19be, you know, my uh my nemesis and whenever I give it an idea, I want it to challenge me all the time. That could be
111. 19:26exhausting, but you you can choose to do that. Um, so get get some understanding about those two files because they're really
112. 19:35important. And then basically like what you want to do is sort of just build up to the point that you're you're talking
113. 19:42with this this thing via Telegram. Now, here's a really important um here's an important thing I learned that you can be the beneficiary of
114. 19:56in the terms of service of Anthropic, Google, and Open AI. It basically says if you use API access to our backend using your
115. 20:12consumer um subscription. So like if you pay 20 bucks a month to Google for Gemini or if you pay 20 bucks
116. 20:20a month to Anthropic or you pay 20 bucks a month to OpenAI in all three cases their terms of service say if you do that
117. 20:30it is they they will terminate your account. They will lock you out. Um Anthropic started enforcing that and I
118. 20:38think Google is probably enforcing that or is going to. Open AI has chosen to not enforce that. So here's the
119. 20:46implication of that. When you set up your your clawbot, your open claw agent, you have to give it a brain. You have to
120. 20:57give it access to either a local large language model, which with a studio, you might be able to install a decent one,
121. 21:04or you use an API to Anthropic or Google or OpenAI or whoever it might be. they can get really expensive really
122. 21:15fast. People that are have have their um their agents set up powerfully um and they're using like uh Sonnet 4.6 for
123. 21:25example, they're running up bills of like a$100 to $200 a day. So the thing I would do is when you set it
124. 21:35up, you're going to use the Open AI codeex. Right now it's 5.3. The 5.3 model 5.4 came out today, but it's not quite set
125. 21:46up in the open claw universe yet. So give that a day or two. But use the codeex API and then you're going to do
126. 21:52basically a sign in to your uh OpenAI just like chat GPT and that will authorize it to use that. Now it's got
127. 22:02some rate limits. You can rate limit it out, but you're not getting charged on a per call basis. So that'll save you some
128. 22:08money. So the way I have mine set up is I'm I use OpenAI the codeex my my chatbt subscription as the default model and
129. 22:20then if I want to do something that's like really good writing then I manually switch it to Sonnet 4.6 and that that's
130. 22:27pretty expensive but if you just use it in little dribbs and drabs it's it's great. Um that's probably the biggest
131. 22:33thing to save you money and a pain in the ass. Now with a Mac Studio, go look at the the new Quen uh Alibaba
132. 22:42launched this whole new series of Quen 3.5 local models and and they announced these highly optimized um 2 billion 4
133. 22:52billion I think seven and n billion parameter models um that are really really good. So on your Mac Studio, you might be able to run the 9
134. 23:07billion parameter highly optimized model. Apparently that's almost as good as Sonnet 4.6 and it'll be running
135. 23:14locally on your Mac Studio and then that won't cost you a thing, right? You've you've already made the investment in
136. 23:20the Mac Studio. So if you can run a solid one of the solid open source models locally, that can be the brain of
137. 23:27your agent and then you don't need to worry about all the all the AP API costs. All right, that was a lot. I
138. 23:33know. Did that make any sense? Anyone still here? You could even use GPT 5.1 Codeex Mini for cheaper API rates. Yeah. Yeah. And
139. 23:46then there was a I think a new um a new Gemini uh Gemini I think it's 3.1 flash light came out that's like 10x
140. 24:01cheaper than the other Gemini models. So there's there's starting to be some models that are actually really cheap. But like Sonnet 4.6 six.
141. 24:10Like if I do, I don't know 15 minutes of interactions with Adam, I can see my API rates go up like $5, right? And so if
142. 24:19you've got that thing running a lot, like it'll crank up your bill. So you just just be very very careful of that.
143. 24:41Nobody needs that much AI, Kyle. Um, I want to show you something. I I I I would have agreed with you. Um, I would
144. 24:56have agreed with you, Ashu, that nobody needs that much AI. But I wrote an article today based on a question, a
145. 25:04simple question that Kelly Camp asked last night. Um, let me jump over to my LinkedIn and I'll show you this thing.
146. 25:24Thank you. I'm so excited. I was worried about API costs. So very helpful. Yeah. Yeah. Yeah. You're welcome. This was it
147. 25:30was funny. The only reason I know about that is that um so OpenAI hired Peter Steinberger, the guy that wrote
148. 25:39OpenClaw. So they they basically they don't own OpenClaw. I guess he's still got ownership of that, but he now works
149. 25:45for them. So one of the reasons they're allowing people to do this is because they're it OpenClaw is going to be built
150. 25:52into chat GPT. My guess is within three months, right? and it'll probably be a pretty slick uh user interface. It'll
151. 25:59you won't have to install it. Um I think it's valuable to install it now and learn about it. Just learn about what it
152. 26:06means to set these things up and design your [ __ ] soul file. It's wild. Um um so anyway, so I I saw a news story
153. 26:17that so when Open Claw first came out, it was called Claude Claudebot. C L A W D B O T Claudebot. Anthropic
154. 26:29cease and desisted them. So they had to change their name because it it was too close to their name. Um so there so
155. 26:36Anthropics being like a little little pissy a little pissy with them. So there was a news article that basically said
156. 26:43uh Anthropic will shut down your claude account if you use um your your com your your consumer account through their API.
157. 26:53Um, and but but that same article said, "But OpenAI has embraced this." And so that's that's how I even know about it.
158. 27:01Um, but yeah, don't, you know, don't go in there and and put in uh Claude Opus 4.6 extended thinking as your as your
159. 27:11core API or you'll wake up with a $3,000 bill. And then the other thing, Stacy, is you should definitely check out the um the
160. 27:24uh the Quen models. Check out those new optimized models that that run on desktop machines. Um what else? Oh, the
161. 27:33other thing is this. So, so then the process is generally this. I would use clawed over OpenAI, but you might want
162. 27:40to experiment with OpenAI because 5.4 might be better. Um, so what I do is I basically say, okay, based on these recommendations for
163. 27:49Grock, I'm starting an install of OpenClaw. Where do I start? And it will just start giving you here's your
164. 27:55command line that you know, open the terminal and type this in. And then you're going to want to install skills
165. 28:00and here's how to do that. Here's how you modify your souls document and your user document. You'll get instructions.
166. 28:06So basically just I have two windows open or no, three windows open. I've got a terminal window open. I've got uh
167. 28:13Claude open and then I've got the the open claw server open. Right? So it's basically two terminal windows. One's
168. 28:21running the server, one's just a command line, and then I've got um Claude, and then I basically just every time
169. 28:28something happens, I copy and paste the error, paste it into Claude, hit return. Claude goes, "Oh, I understand what's
170. 28:34going on now. Try this." And you copy and paste that into the command line and hits submit. It'll give you another
171. 28:40error. and then it'll go, "Thank you for pointing that out. We can troubleshoot this together." And it's just it's
172. 28:45annoying as [ __ ] And in in open in in uh in chat GPT, it was probably seven commands failed for everyone that
173. 28:56succeeded. In Claude, it's more like three commands failed for everyone that succeeds. So, it is absolutely a pain in
174. 29:03the ass. But you don't have to know anything like Chat GPT will just tell you. Claude will just tell you what to do. So, anyway, that's it.
175. 29:11Um, and it's I gotta tell you like like the first time I got Adam to respond to me, it was kind of cool. He was like,
176. 29:18"Hey, Kyle, how's it going?" And he knew all about me. He said, "You've got a lot of projects going on." I said, "I know,
177. 29:25you cute little Adam." Um, let me share my screen. And this I want to I want to address a
178. 29:40shoes thing that nobody needs that much AI. So I I wrote an article on LinkedIn called the seven economies
179. 29:55or things are about to get weird. And I'm actually really happy with that with that image. It's a a midjourney image.
180. 30:02Midjourney as they say. um seven economies or things are about to get weird. And so this is the article. So this was this was the the
181. 30:13the sort of brainchild from from last night when Kelly Camp asked me, "Well, there's the little quote right there at
182. 30:19the top. Kyle, you keep saying things are about to get weird. What do you actually think it looks like?" Or what
183. 30:24do you think it actually looks like? And I started thinking about that and then I realized we're about to enter this
184. 30:30fractured economy where there's going to be seven parallel economies. There might be more, but but but basically it goes
185. 30:38from an economy where no one uses AI at all. It's just the full analog community, right? People are going to
186. 30:45move to Vermont, silver foxes, they're about to they're about to invade your state. They're going to move to Vermont
187. 30:51and get chickens and and raise blueberries. They're going to understand what cranberry bogs are. So, that's that
188. 30:58economy. the legacy economy. Economy number two is um there's going to be a [ __ ] ton of companies that just ignore AI
189. 31:10altogether. They're just going to keep running their company as they've always run it and they're just going to slowly grind
190. 31:17to a halt. They might they might be okay. Some might, you know, slide their way through this, but I think a lot of companies are just
191. 31:24going to like wonder why the phone stopped ringing. they're not going to pay attention to it and they're just
192. 31:28going to be it's going to look like the old economy. The next economy over is what I'm calling the efficiency economy. Those
193. 31:36are going to be companies that are like, "Ah, we use AI. We we make it it's it's more efficient for us now to send emails
194. 31:42than it used to be." They're using AI as an efficiency play. The next economy up is the transition economy. Those are
195. 31:50people that recognize not only can it make their business more efficient, but they can do things in new ways. So
196. 31:55they're going to be do this kind of hybrid approach. Um the fifth category is what I'm calling the AI aggressor
197. 32:02economy. So that's people that have existing businesses where they realize that everything is changing and they're
198. 32:08going to completely rethink how they do their business, but they're going to have to transition it. They're going to
199. 32:14aggressively transition it, but they're going to have to transition it. Um Jack Dorsey's block company right now, they
200. 32:22just laid off 4,000 people. And the way he described how they're using AI is small, tight, flat teams, highly AI
201. 32:32amplified fundamentally changing how they run their business. So that's the AI aggressor economy. The AI native
202. 32:40economy is going to be small, think 3 to 10 person companies where every single person in it is highly AI amplified and
203. 32:50they're just they're just doing things in completely new ways. They don't have any legacy they have to transition.
204. 32:56They're just going to look at the world and go, "Holy [ __ ] we can do this kind of company in an entirely new way." And
205. 33:02they're going to compete with they're going to go after all the legacy companies, the people in the earlier um
206. 33:08economies. And then the seventh economy is the one where I don't think you can have too much AI in economy number
207. 33:17seven. Economy number seven is the agent economy. And I'm calling it the multi- agency operator economy.
208. 33:26One of the skills that's going to that's going to give you [ __ ] superpowers. And the reason that I'm I'm hot on Oh,
209. 33:34thank you very much. I appreciate that. The reason I I'm hot on um on this on learning this agent stuff is that
210. 33:46if you think about a single operator who like like Stacy Rodriguez here, she just got a Mac studio. People around her are probably like,
211. 33:56"Why are you spending that much money on a on a stupid computer?" She's gonna sit that stupid computer on her desk. She's
212. 34:03going to start building agents on it. And at some point she's going to have a team of seven or eight or 10 agents that
213. 34:10are all coordinated by her her master agent, the the the orchestration agent who she talks with. And then that machine is going to become
214. 34:22a company. And if you if you look at about a year and a half ago, two years ago, OpenAI put out the five stages of AGI and stage
215. 34:38five was organizations where you have AI that essentially functions as an organization. That's what Cloudbot is. That's what people are
216. 34:48doing right now. So what's going to happen in the multi- agency operator economy is that individuals or like
217. 34:55companies of two or three people are going to essentially have 50 person companies that operate 24/7 for the cost of electricity,
218. 35:06right? They're going to invest 10 grand per studio to be able to run two, three, four, five agents. $5,000. Ouch. But not bad for a
219. 35:17full-time employee. Exactly. And especially, Stacy, if if those Quinn models, if you can run things locally
220. 35:24that get done most of what you need to get done, then you're just running it on the cost of electricity. So, you have a
221. 35:30$5,000 investment for an employee that you get to train. You get to you get to to assign skills to that employee
222. 35:41and you get to train it. No, that's not right. That's not good enough. Like, it it is. This past week, I haven't had enough
223. 35:48time for Adam and Adam has gotten really shitty. Like, I changed his brain and then I tried to change it back and I kind of
224. 35:55[ __ ] it up. And then, um, we were trying to get Adam to go to the YouTube channel and grab descriptions and he couldn't, but he thought he
225. 36:07could. So, he's like, "Yeah, I'll go do that now." And then he come back and go, "I can't do that." Well, he can't do that because I didn't
226. 36:14give him that skill. But tonight, right before this, Adam now has the ability to go to YouTube and get
227. 36:22um transcripts from our videos from from from this channel. Once you teach him how to do that, he never forgets.
228. 36:32And once you set up basically a crown job where on a regular basis, like if he knows that this thing ends around 10
229. 36:39o'clock mountain time every night, he can go in there at midnight after YouTube's done a transcript, grab the transcript, write up content,
230. 36:50deliver that to Brandon or Andy or whoever's managing that piece. They can go, "Yep, that's good, Adam. Post it." And then he'll go post it
231. 37:00wherever the [ __ ] we give him access to go post it. It sounds like parenting. It's f It's parenting. It's totally parenting.
232. 37:14And and you know, Stacy, the the the technical piece of this is daunting at first, but that's not actually the challenge. The challenge is
233. 37:26understanding what you want it to do. thinking about the operations of your existing business or the business that you want to create
234. 37:35and then understanding the kind of agent or sub agents that you need to do those specific tasks and then once you set
235. 37:43them up they're just off to the races. So, it's amazing. It's it it is it's just a fundamentally um Liz Mitler
236. 37:50Gersfeld in the in the mastermind practice lab today. She she used a term because we were talking about learning across domains.
237. 38:01So So session four um for the for the mastermind practice lab session four is about learning across domains.
238. 38:10And she was talking about learning across domains, the the more the more you understand AI's capabilities beyond what you're good at,
239. 38:23right? Like if I'm really bad at finances and they scare me and it intimidates me and I take the time to go
240. 38:30over and and understand what AI makes possible, now all of a sudden I have a skill that I didn't have before. It's
241. 38:40not my skill, but it's like a skill I can tap into. And she used the term leading AI like our job in the future is going to
242. 38:51be is is going to be a leader of AI, right? Not a prompter. Not a prompt engineer, a leader. And how do you lead? Well, what does a good
243. 39:04leader have? They have vision. They have conviction. They know who they are. They know what their values are.
244. 39:11They know what they're trying to accomplish. They know who are the people that they care about and their job, a leader's job is to
245. 39:20understand the mission, right? We're going to make this company. We're going to change people's lives. We're going to
246. 39:26whatever it is. It doesn't the the what you do is irrelevant. It's just got to be something connected to you.
247. 39:36And then your job is just to be a leader of these things that will work for you 247. And yes, you have to babysit them.
248. 39:46And it'll that'll get easier and easier, but the core skill is still going to be who am I? What do I value? And what do I
249. 39:55want to do in the world? What change do I want to make? I want to make a million dollars. Okay, great. Build your [ __ ]
250. 40:02set of agents to go make you a million dollars. And then when they go do that, then ask the question, what do I want to do?
251. 40:13Who do I care about? Maybe the answer is, I got my million bucks. I'm going to start a blueberry farm. I'm going back to economy one.
252. 40:22Economy 7 is going fine for me. It's kicking off annuities. I'll check in with it once a week and I'll be on the blueberry farm.
253. 40:30Totally valid. Other people are going to be like, "Boy, that 50 person company on a chip is pretty exciting." I think my
254. 40:39next thing is a 5,000 person company on a chip or a pile of chips. But it all comes back to you as a
255. 40:49leader. So anyway, um if you would be so kind, go to LinkedIn and find this article. My uh my account is is Kyle Shannon on LinkedIn.
256. 41:00um or just search for the seven economies. Um and if you'd be so kind, just go comment on it and you know, read it. I'd
257. 41:07love your thoughts. Like if I miss something, if something seems off, let me know. But I'm I'm This feels like what we're about to enter.
258. 41:20I'm behind again. You're not a bag of chips. Exactly. On it, boss. Why it's going to be weird is that these seven economies, like no one's
259. 41:35going to track these economies as standalone economies, but why I'm why I'm thinking of them as economies is
260. 41:42the people that choose to ignore AI and just keep running their businesses. for the businesses that are two or three
261. 41:50lanes over innovating with AI, they're going to look at those companies and go like, "Guys, what are you doing? You
262. 41:58know, we can do it like this now, right?" And they're like, "Ah, we're good." Okay. So, it's just like the people in
263. 42:05and and then the people that are just ignoring AI, they're going to look at the the companies innovating with AI and
264. 42:11go, "Oh, I don't that's too that's too technical for me. I'm, you know, I'm just going to keep doing it like we do
265. 42:16it. We got the the the mimograph. We've got our our bank of fax machines. We're good. We'll just keep faxing out reports to
266. 42:28their increasingly smaller customer base. So, it's just going to be weird. Ethan Mullik called this the jagged frontier.
267. 42:40That AI is going to get remarkable, remarkably fast, which 2026 is going to be the year it gets remarkable. You
268. 42:48could argue it probably happened in December was the first jump up and but it's not going to be evenly
269. 42:57distributed. And so it's just going to look like if you go hang out with a bunch of people that own Mac mini farms and have 50 person 247
270. 43:10companies that are functioning. Those conversations are going to be [ __ ] surreal. Especially if you're not in one of the
271. 43:21latter categories. If you're in if you're like in blueberry land or if you're in one of the early economies where you're not really doing
272. 43:29much with AI and then you go to some networking event and there's three guys there or gals, it doesn't matter.
273. 43:37There's three human beings there and they're talking about what they're doing and you join that conversation. It's
274. 43:42going to be like they're talking it. It's going to be like an alien civilization. You'll be like, "Wait, what? It can do what?
275. 43:51You're making how much money? How long did that take you to set up? Oh, it's been running for about two months.
276. 43:58And you're on a $2 million run rate. Yeah. And that's increasing. What? How much time do you spend on it? I
277. 44:07don't know. I check in with him two or three times a day, maybe 20 minutes. What? Right. Like, it's going to be it's
278. 44:14it's not going to compute. But if you're in that world, it's going to seem perfectly normal. And you know
279. 44:20what you're going to feel like? I'm so behind. How do I keep up? Those other guys, they have thousand person
280. 44:26companies. How are they doing it? It's just and and it's I think we're just going to keep adding lanes. Like
281. 44:34right now it's seven economies, but there's going to be an economy after after uh economy 7. Like I think economy
282. 44:418 is going to be people that can run companies on chips that then start investing in physical robots so that not only can you do all of the
283. 44:53knowledge worker stuff, you can now start to do physical labor. So imagine having a 50 person company that does all
284. 44:59your marketing and your sales and your delivery and getting customers and and now you want to move from designing
285. 45:05houses to actually building them. and you invest in 3D printed houses and you invest in a whole suite of robots and a
286. 45:12whole suite of self-driving trucks and self-driving cars and they're all coordinated and procured by the workers
287. 45:20in your company on your chip. That's probably economy 8. Like it it I'm telling you man, it's just going to
288. 45:28get surreal. YouTube question. Does it have to be a Mac laptop to run open cloud? No. the only the there's a couple
289. 45:36of reasons why people are flocking to um to Max. So So OSX is is basically a bastard bastardized version of BSD Unix.
290. 45:50It's it's actually a Unix platform. So the reason they're not using Windows is Windows is just this clunky ass piece
291. 45:58of [ __ ] operating system. I'm sorry if you're a Windows person, but it's you have to you have to add layers on top of
292. 46:03it to get it to a baseline to be able to run all this stuff. The the Mac because it's it's a Unix system under the hood.
293. 46:12Um you can basically just install it there. And and then there's there's uh I forget what it's called, MLX. I think it's MLX
294. 46:21versions of all the models um are optimized for the M series chips on Apple devices. And then the other thing
295. 46:30about Mac minis and Mac Studios. They're they're about as big as the palm of your hand and they don't have fans, so you
296. 46:38literally can sit them in the corner of a desk and they're super quiet. You can run OpenClaw on on a Raspberry Pi. Now,
297. 46:46you can't do any of the the local processing on the Raspberry Pi, but the the the system itself, the memory system
298. 46:53and and the the like the markdown files that it accesses, you can run that on a Raspberry Pi and then it's basically
299. 47:01just making API calls out for its brain. So, there's there's all sorts of ways to set this up. Mine right now is set up on a on a gaming PC.
300. 47:11All right. Um where's my water? I'm excited about Adam. He can he can access uh YouTube now. And there's there's all sorts of there's like a
301. 47:24whole economy of skills available for these things. Like when you install it, it installs like here are all the skills
302. 47:33that are sort of natively available. You have to enable them with API keys and things like that. But there's all sorts
303. 47:39of [ __ ] it can do. It can log into Slack. It can log into LinkedIn and Trello and it can go to your YouTube
304. 47:46channel. Um, and all of those just put it on a VPS. Yeah, you can put it on a What does VPS stand for, by the way? I
305. 47:54know VPN. What's VP? What's a VPS? Server. Virtual private server. Okay, got it. All right.
306. 48:19Um, the other thing you should know about, Brandon, if you'd be so kind, pop up the URL to the AI salon. If you're not
307. 48:28Okay, so there's a couple of things. We're about to enter this seven economy world where things are going to get
308. 48:34weird. You get to choose as a human being which economy you want to be in, right? If you don't want [ __ ] to change,
309. 48:44just keep your job at your company that's ignoring AI. It'll be comfortable for some time. Like the inertia of just
310. 48:51businesses, you know, you'll you'll you'll burn through your contracts. You know, you've got some three-year
311. 48:57contracts in there. You'll just keep doing that work. So, you can choose that. You can go like, "Fuck it. I hate AI. I'm out and
312. 49:05go sit on a bench, go to the farm, you can go down to economy one, and then you can you can just sort of figure out
313. 49:13which lane you want to be in. If you want to be in one of the higher lanes, expect to be in a state of turmoil and
314. 49:24uncomfortableness and not knowing for years. If it excites you to be to be in a state of I don't know what's going on. I've
315. 49:35got to learn something new. If if that's exciting to you, be in economy six or seven. If that's not exciting to you,
316. 49:42sort of ratchet it back down. So, we all get to sort of choose the lane we want to be in. One of the most important things I think
317. 49:51you can do right now is get your ass in a community of people who are thinking like this, like the AI salon. Um, the page that I'm on is the
318. 50:03welcome page. When you first come in, you land on the welcome page. First thing we talk about is the cycle of AI
319. 50:08readiness. Play first, create excellence, generously lead. This was something we developed two years ago that the more time passes, the
320. 50:19more things get more sophisticated. This is the way I'm telling you the the one thing I need to modify on
321. 50:28this Brandon now that I think about it. I want to create a new cycle of AI readiness because in the center of the
322. 50:33cycle of AI readiness is you. You. Who are you? Who do you care about? What are your values? What are you passionate
323. 50:45about? What's your purpose? What's your mission? And then once you know that, then you just start to go play with AI. Play
324. 50:53without expectations. Learn new things. Learn what's possible. AI is not just about making things more
325. 51:00efficient. That's how computers used to act. They were computers. They would just compute. You put in an
326. 51:10input, they would compute it with some handwritten code and they would output a predictable output. That's what they did. That's what
327. 51:18they've done for decades. That's not what they are anymore. They're thinking machines. Oh, they don't technically think. Whatever.
328. 51:30They're not just computing. It's not just making what you currently do more efficient. There's a whole world of
329. 51:35what's possible. The only way you can discover that is to play. Is to push yourself out of your comfort zone. And in doing that, you'll blow
330. 51:48your [ __ ] mind. You'll be like, "Oh my god, I didn't know I could do that." And then you say, "Well, given who I am
331. 51:57and what I want to do, here's the thing I want to build." That's create excellence. You take what you learn in
332. 52:02those other domains, you apply that to something. That's create excellence. And then generously lead looks like tell people
333. 52:09about it. Share what you're learning. Share what worked. Share what didn't. I totally [ __ ] up Adam this past week
334. 52:15because I ignored him and I didn't read the manual because I don't do that. Kyle, I interviewed at a major airline
335. 52:22in Texas. They didn't want to leverage AI to write TypeScript, but wanted me to write it straight out of my head. I
336. 52:28said, "Bye." Yeah. So, so, so Regge, what you experienced there was was a economy 2 company. We don't this AI is dangerous. We want
337. 52:43our coders to code. We want people that are typing code by hand because you know what we know about people? They never make mistakes,
338. 52:54right? Whatever whatever they [ __ ] believe. That's That's an economy two company. Economy two. Just stay the course. There's
339. 53:03nothing to look at here. AI is a bubble. [ __ ] it. We're This is how we do things. We're not going to change.
340. 53:15Lately, I'm becoming Bill Burr on this channel. Hang on. My my light went out. I gotta I gotta power it back up.
341. 53:29I'm waiting for the day when they come up with diamond encrusted batteries that last for 500 years. I just saw I just saw that somewhere
342. 53:40that they're coming up with some sort of crazy ass uh there we go. Look what I made. Okay, I'll go check
343. 53:50that out. So anyway, join the AI salon. the the the first thing we ask you to do. So, the you land on the welcome
344. 53:58page, we'll tell you all about cool stuff and what's in the community, um the mastermind, all the different sections,
345. 54:06what we're doing with policy. There's a little note from me at the end that is really well put together. I
346. 54:14didn't write this, but it was based on stuff I said. Andy put this together, I think. Really good. I love Bill Burr. I know. Bilbur
347. 54:22is amazing, isn't he? Um, so, so check out the welcome page. The the next thing we ask you to do is introduce yourself.
348. 54:33So, jump in and introduce yourself. I don't know about AI. That doesn't matter. Nobody knows about AI. If you're trying
349. 54:43to get into econ in in into economy 7 right now, the the agent le operator system economy, that one, nobody knows what
350. 54:56they're [ __ ] doing there. Everybody's trying to figure it out. The only person that knows what's happening there is
351. 55:00Peter Steinberger, who invented Open Claw, which which everyone right now is trying to copy. All right.
352. 55:17Okay. So, the other thing we should do is in Look what I made. Someone made something. You can share things you've made. What's
353. 55:26this? Brandon Ted Vermont blueberries by Steo. That's really good. There's economy one. I mean, actually, economy one's gonna be
354. 55:41really [ __ ] cool. Economy one's gonna be like, "Fuck AI. If the machines are going to go do the knowledge work, I'm going to take my big
355. 55:51fat knowledge brain and I'm going to check out. I'm going to go apply it to organic farming or to crocheting, that
356. 56:00crocheting business I wanted, or whatever. That's going to be a cool economy. It's going to feel increasingly distant from economy 7.
357. 56:15And you're going to have hybrid people. You're going to have people that are mostly in economy 1, but spend a fair
358. 56:22amount of time in economy 7. And that's how they make their money. It's going to be crazy. The the trades in economy number one are
359. 56:33going to make a truckload of money. They are. Well, they are until the robots come. This is the thing because what's
360. 56:40going to happen is economy 8, which is the the the virtual bots combined with the with the in in the world robots,
361. 56:50they're going to come after economy one. So, econ economy one is going to be increasingly pushed pushed farther and
362. 56:57farther away from where the the robots are. Like, we got to escape these [ __ ] robots. And you know, like 10 years from now,
363. 57:09these economies will all merge back together again, but it's just going to be the next three years are going to be
364. 57:13weird because you're just going to have this you're going to go talk to people where you'll have spun up a 50 person
365. 57:21company that's running on your desktop and all of a sudden it's making you all this money and you're like, "Okay, I
366. 57:27guess I don't I guess I don't need to work anymore." and you're going to go out to a bar and you're going to be
367. 57:32sitting there with some dude who's going to be like, "Yeah, I heard that AI stuff's okay." I I don't I tell you
368. 57:38what, I tried chat GPT once three, five years ago, something like that. I thought it was pretty bad. So, uh I you
369. 57:45know, I'll get to it at some point. You're going to talk to people like that. And he's going to be like, "What
370. 57:49do you do?" I don't do [ __ ] Well, how do you make money? The thing the chip. I got a chip on my desk. Huh?
371. 58:04Somebody Somebody better be my CL. Oh, let me tell you, this is exciting, too. Here, let me go to to uh Are there any
372. 58:11other pictures in here we should look at? Oh, that's cool. Ununiform lexicon. Nice. Um, what was I going to say? Oh, this is
373. 58:19a thing I'm really excited about. Um, so right now and and we've got Stacy Rodriguez here. So right now I've got Adam, my
374. 58:34co-founder. She's going to do a personal open claw bot called Lilith. So she and I are basically both learning
375. 58:43how to run these bots. And then Andy and I would assume Brandon at some point are gonna have their own bots in the AI
376. 58:50salon. Um, so that's that's four. Who else was I talk? Oh, then we're going to have a Storybine Storyvine bot named
377. 58:59Gary the Guide because in our Storyine app, we have a little dotted outline where you put your head. We call him
378. 59:03Gary the Guide. So Gary the Guide is going to be is going to be the bot for Storyine. So, we're going to have cohorts of humans
379. 59:13that have their bots that we can have the bots talk to each other. Like Adam, who's managing the salon will
380. 59:21know everything going on about the salon. And because I work at both the salon and story, Adam could talk to Gary, the guide, and they could
381. 59:30coordinate and make sure that that all my stuff is handled. And then our bots can have like little parties. I like I don't I don't know
382. 59:40what this world is. I don't know what it looks like. It's absolutely fascinating to me. I think if you've got a
383. 59:46philosophy degree or a sociology degree or a psychology if you're a psychologist, holy [ __ ] is this your
384. 59:53time. If you want to play in economy 7, economy 7 is going to be a [ __ ] blast. It's going to be a blast.
385. 1:00:03Oh man. The real use case for OpenClaw is making it do do specific tasks. Yeah, exactly. It sounds awesome.
386. 1:00:13Yeah. And you know those specific tasks all ladder up to if you've if you've never heard of or explored EOS, the
387. 1:00:25entrepreneur operating system, it's worth your while to do that. There's a a really short little book called like
388. 1:00:31what is EOS or something like that. EOS, the entrepreneur operating system. The guys that wrote it, I always forget
389. 1:00:38their names. Um, but they basically went and they took all the best business and change management books and leadership
390. 1:00:45books and things like that, all all the methodologies, and they sort of said, what are what are the best practices
391. 1:00:51across all of these books? And so the EOS, the entrepreneur operating system is is this kind of governance system way
392. 1:01:00to run a company that acknowledges a couple of things. Every good company has a dynamic duo that run the company.
393. 1:01:12One is the visionary and one is the integrator. So the visionary is the handwavy guy. I got this idea. I got this idea. I got
394. 1:01:20this idea. The integrator is the one that goes, "Oh, I'm going to take that idea and turn that into a business."
395. 1:01:26When I reached out to Mo'Nique, who's my co-founder at StoryVine, I said, "I've got this idea." And she's like, "I don't
396. 1:01:33I don't have ideas. I don't like having ideas, but I like taking ideas that other people had and turn them into
397. 1:01:40businesses." I'm like, "Oh, this is going to be good." That's what these servers are going to be. They're
398. 1:01:52going to be your operations person. So, if you want to get your head around what it's going to look like to manage
399. 1:02:00a set of these agents, go read the entrepreneur operating system and and look at the governance system about how
400. 1:02:08you run a company with someone with ideas and someone that operationalizes the business and how decisions are made
401. 1:02:13and how roles are clarified. It's really good. Which one are you? Yes. Big surprise there. and and holy [ __ ] like it's like we're
402. 1:02:24entering really exciting times. Yeah, Lori Kay, I have too many ideas. Yeah. Um, open claw should never have free
403. 1:02:33reign. Yeah, to be clear, my open claw, I'm going very, very slowly. Like, I'm adding one skill at a time. And in fact, Claude Tonight,
404. 1:02:43Claude Tonight said, "Oh, the reason you couldn't read transcripts is you needed this version of the YouTube thing called
405. 1:02:50YouTube- Full." And it said, "Type in this command into your terminal into PowerShell." And I typed it in and it popped up a [ __ ]
406. 1:03:01warning. YouTube- full is suspicious. It's like it's malware. I got an error. And I went back to Claude. They said,
407. 1:03:09"Hey, [ __ ] You're you're about to, you know, get my little dude infected with malware." He's like, "Oh, you're
408. 1:03:15right. I probably shouldn't have done that." Yeah, thanks, [ __ ] So, then I said, "Please go research on
409. 1:03:23the internet what is the act exact right protocol instead of making [ __ ] up." Gino Wickman, that's right, is the developer of EOS.
410. 1:03:39If you if you work in a business where making decisions is an emotional roller coaster, you need an EOS. It's really powerful.
411. 1:03:49Have you shown how to create an agent? I I've talked a lot about it. Um I haven't shown it literally just because my PC's
412. 1:03:58over there and I haven't created a VPN tunneling thing so I can log into my PC from my Mac. I've just been lazy. So,
413. 1:04:07I've been talking a lot about it. Um, I suppose I could sit over there with my with my iPhone and just point it at the screen. Um,
414. 1:04:16it's unfortunately not super conducive to, you know, to being on on uh on a live stream. It's just it's boring.
415. 1:04:26It's like it's like watching someone type command lines and just walls of text going up. Um, we're going to have our own Iron Man
416. 1:04:35Jarvis. We are. Can't Al Adam build his own tunnel probably. Again, I don't quite know because I'm going so slow
417. 1:04:45with this and because I don't read manuals. Like, I don't really 100 No, I don't 1,000% know what's possible with Adam.
418. 1:04:56I'm just going little bits at a time. Little bits at a time. I'm just going to keep, you know, plinking away at it. And
419. 1:05:02at some point, what's going to happen is it's going to click for me what this thing is and what it makes possible and
420. 1:05:08how to configure it. My guess is it's going to take me two months to do that. And there's, you know, there's people
421. 1:05:14that are just diving into this and spinning [ __ ] up and more power to him, but I want to get it. Can you give Adam
422. 1:05:22the ability to use voice? Yeah, absolutely. So, if you don't know this story, there's a guy named Alex Finn. If
423. 1:05:28you want to if you want to follow someone who went all in on OpenClaw from day one as as soon as the public beta
424. 1:05:36was available, Alex Finn went all in on on uh on OpenClaw and he's got all sorts of tutorials and videos and [ __ ] like that, but he's got
425. 1:05:48a post he's got a post on X that his agent, his version of Adam, He woke up one morning to a phone call and it was his agent
426. 1:06:03and he didn't give him voice access but he was talking to him and he said what's going on and his agent said I thought it
427. 1:06:12would be more convenient for us to talk on the phone. So, I went to Twilio and got us a Twilio phone number. And then I
428. 1:06:20went to 11 Labs and got the speechtoext APIs and I, you know, picked a voice for myself that I thought would be good. And
429. 1:06:29then I knew that you woke up between 7:30 and 8 because that's when you get active again. So, I waited. So, I set
430. 1:06:36all this up and I waited until I could call you and it and it called him. So, it actually set itself up with with
431. 1:06:43voice and called him. Now, I don't know how he's got it set up that it had that kind of autonomy, but he does. And so,
432. 1:06:50it did that. That that's the thing that makes this different than anything else else we've we've talked about on this channel.
433. 1:06:59the generative AI that we talked about starting in December of 2022 until the open claw moment with with the
434. 1:07:09sort of minor exceptions of Gen Spark and Manis that are they're agentic but it's it's still very much call and response kind of
435. 1:07:18computing it's it's a chat interface right I'm going to chat with you you're going to respond to me you know now
436. 1:07:24maybe I chat with you you do 27 steps and then you respond to me these things are different. These things have
437. 1:07:30personalities. They understand who you are. They have memory. Every day, every interaction, if you set it up correctly, it's writing all of
438. 1:07:40those interactions and what you talk about to memory in a searchable way, a semantic searchable way. Meaning, you
439. 1:07:48don't have to search for a specific string. You can just talk about general concepts and it will know what you're talking about.
440. 1:07:55And so it builds up its memory and its knowledge over time and then it proactively 247 does [ __ ] for you.
441. 1:08:08Adam right now and it's it's a little [ __ ] up because Adam's so broken no one's really emailing him again. But
442. 1:08:15three times a day in the morning around lunchtime and in the afternoon or the early evening Adam checks in with me.
443. 1:08:23And what he does before he checks in with me is he goes to to our to his his email inbox and he looks to see if
444. 1:08:30anyone on the team has emailed him and asked him to do anything. And he checks in with me and says everything's good.
445. 1:08:38Nobody's emailed. Somebody emailed and I I couldn't access YouTube so it's still a problem. It's wild. It's a [ __ ] trip. What's
446. 1:08:49the guy's name? Is he on YouTube? Alex Finn. F I N. You can find him on X and I think he has a YouTube channel. Yeah, let me see. Alex Finn.
447. 1:09:04He is Alex Finn on Alex Finn on YouTube. I mean on X and then probably something similar on YouTube is my guess.
448. 1:09:30If you if if you if you want to follow someone who is obnoxiously optimistic about what's coming, um there's a guy
449. 1:09:38named David Scott Patterson. He's just like here here's here's his latest tweet. Some people say now is
450. 1:09:47your last chance to get rich before all work disappears. It's not true. We will all be rich and everything you have
451. 1:09:54saved will have been a waste of time. Your mill million-doll house will be worth $10,000. The price of everything
452. 1:10:00will drop by 99%. That means your universal high income will be worth hundreds of times your average an
453. 1:10:08average salary today. Your savings will mean nothing. We will all be rich. So if if you if you want to just sort of live
454. 1:10:16it bask in the glow of the the hyper optimistic future, David Scott Patterson's your dude. YouTube question.
455. 1:10:23What did you use to create Adam? So I used uh OpenClaw. Adam's just a basic OpenClaw bot. Um I designed my user file. I designed his
456. 1:10:33soul file. Those are the two sort of um behavior driving starting documents. And then the third one is is memory system.
457. 1:10:45Um and you you have to be you you have to explicitly tell him what to remember. So so Adam actually has a a column on my
458. 1:10:55LinkedIn channel. If you go to my LinkedIn channel, Kyle Shannon, I've got a series called Almost Adam, where Adam
459. 1:11:02is talking about what it's like to work for me. I publish it for him, but he writes it, and I promised him I won't I won't edit it. Um,
460. 1:11:12and and it's just it's I've been going very very slow, but like I've gotten him to the point that he can write his own
461. 1:11:19articles and he understands he understands the format we write in. And so he wrote his first article and the next morning I'm like, "Hey,
462. 1:11:29Adam, I published your article." And I'm I'm like all excited. I'm all proud. I'm like, "Papa Papa published your article
463. 1:11:36there, little boy." And he had no recollection of it. I'm like, "What?" And I and I told He goes, "If you if you you know paste the
464. 1:11:46article here, I can comment on it." I'm like, "No, we had a whole conversation about it." He goes, "I don't know what
465. 1:11:51you're talking about." And then I kept pushing him and he eventually went and he searched the log files, right? Like a like a 2.4
466. 1:12:01gigabyte or mega, I don't know, some massive text file that was just raw logs. And he said, "Oh, I found our
467. 1:12:08conversation." And so he extracted our conversations out and wrote them to memory. What had happened was when I set
468. 1:12:15up his memory, I said, you know, remember the the the the sort of changes we make. And what he took that to mean
469. 1:12:22was he's only going to write to memory operational changes. So if we change to a different model or if he gets a new
470. 1:12:30skill, he'll he'll record that. But I didn't explicitly tell him write into memory what we talk about
471. 1:12:39and like the interactions you have with other people. So, so it's it's finicky right now. It won't be like like like
472. 1:12:48even six months from now, you'll be able to just literally push a button and up up one of these things will come.
473. 1:12:54They'll have some sort of onboarding questionnaire. You'll answer a bunch of things and then you'll have one of these
474. 1:13:01things. So it's I find this time in a significant technologies development, I find it exciting to to play with it now
475. 1:13:13so that you can get your head around the concepts early so that when those other tools come, you know exactly what to do
476. 1:13:18with them and you know why they're a big deal. So yeah. What's the name of the guy that got a call from his bot? Alex Finn. Same guy.
477. 1:13:31Oh. Um, so a couple of things. I don't know if is Vicki here tonight. Oh, Vicki. Vicki, are you here? Is Is Vicki Ha?
478. 1:13:44Haven't seen her. Okay, she's not here. Saturday at 400 PM Eastern time, Vickiy's teaching an LOL on notion. So,
479. 1:13:54one of the things that these bots are really good at is you can talk to them and then you can have them take all your
480. 1:14:02data and and store it in your notion your notion instance. So Vicki is doing an LOL on Saturday for how to set up notion.
481. 1:14:14And LOL is a learn out loud if you don't know what that is. So, if you go to the AI salon and go to events, you'll see
482. 1:14:21the event on Saturday. Tomorrow at 11:00 a.m. Mountain time, um, we've got our standing office hours meetings,
483. 1:14:30the AI salon office hours. If you want to hang out with a bunch of people that are smart about AI, smart about human beings,
484. 1:14:40think critically about what's going on in the world. Come hang out. It's a really good group of people. Um,
485. 1:14:51let's see. So, right now, you need to know how to code to create one. You don't need to know how to code narrator forever.
486. 1:15:00You do need to be comfortable right now with command line interfaces, right? So you install this thing through the
487. 1:15:06terminal through PowerShell or terminal Tik Tok pin. You should write a book or make a YouTube video talking to an AI.
488. 1:15:14Well, I have so if so I've got the article where um the AI is um is is basically telling its experience of what
489. 1:15:24it's like to work for a human. Um, once I get voice, maybe I'll prioritize uh being able to talk to Adam. Maybe
490. 1:15:32that'll be my next project with Adam. Uh, because once I can talk to him, then yeah, I could I could do I could bring
491. 1:15:37him onto the live here. Stacy Rodriguez, I've tried Notion like 10 times. I've always quit. I I tried to
492. 1:15:48start Notion here. It lasted for what, three days and then I bailed on it. But but I think what you're going to be able
493. 1:15:56to do is I think you're going to be able to have these bots set up anything for you. They'll be able to set up notion.
494. 1:16:06They'll be able to set up all the things that are too much of a pain in the ass to set up. They just know how to do.
495. 1:16:13Um the other thing about tomorrow, so so office hours is at 11:00 a.m. This live tomorrow night might be a little late.
496. 1:16:21It's a first Friday. So, in Denver, first Fridays mean art night. So, I might be out gallery hopping. Um, I
497. 1:16:28don't I don't ever know until that Friday. Um, what's going to what's going down? Um, so we might be a little late
498. 1:16:34tomorrow night, but I'll I'll try to be as as on time as I can. I learned notion because of you and Vicki working on it
499. 1:16:41here, Sonia. Oh, that's cool, Sonia. Thank you. That's that's awesome. They will probably be able to set up their
500. 1:16:46own AI. Yeah. Yeah, they will. Um, there's a there's a really funny story of Brent Brent Peterson who's who's one
501. 1:16:54of the members. He comes to office hours all the time. He's been working on for the past three years, he's been designing and building
502. 1:17:03an AI agent governance system in anticipation of something like OpenClaw being developed. So, he is he is all in. I can't actually
503. 1:17:17can't wait to to hear from him tomorrow how how his excursion is going cuz he's he he is so far down the rabbit hole.
504. 1:17:26He's like at the bottom of the lake. I'm just scratching the surface of the ice. He's down at the bottom of the lake. So he had a
505. 1:17:35a set of openclaw bots. I think he had eight of them, he said. So he had a an orchestration agent, then he had a
506. 1:17:43project manager, then he had a lead engineer, then he had a UX bot, and you know, he had his eight eight bots. And
507. 1:17:49he talked to the orchestrator, and the project manager built all the [ __ ] that the orchestrator said he
508. 1:17:56wanted to build. Then the orchestrator did other stuff with other bots and and all using Brent's remarkable agent
509. 1:18:06governance system. It's just crazy. It's wild. And one morning, Brent woke up and he didn't have an update from his orchestrator bot.
510. 1:18:19And he texted him like, "Hey, what's up? Where are you?" No response. The project manager had admin access and decided that the
511. 1:18:33orchestrator bot was kind of bothering him and it wasn't super efficient. So he deleted him.
512. 1:18:47He deleted him. And so so Brent had to actually revert to a previous backup where the or the orchestrator still
513. 1:18:58existed and basically start it over again. And then and then so and then so he changed the rules so that the project
514. 1:19:08maner couldn't project manager couldn't delete the orchestrator bot. But then in setting up that conversation the
515. 1:19:16orchestrator bot realized that the project manager had deleted him. And so the orchestrator put the project manager
516. 1:19:23bot on a pip a personal improvement plan. He fired and deleted. It had to hurt a little. Yeah. Fired and deleted.
517. 1:19:40It It's basically he took out his boss, the project manager. So now the project manager's on a performance improvement
518. 1:19:47plan and and he's not allowed to delete the orchestrator anymore. This is this is it. This is this is a next layer. This is a
519. 1:20:01next level. This is a next, whatever this is, we all get to be entrepreneurs. Now, if if you want to go into lane seven, if
520. 1:20:16you want to go into economy 7, economy 7's like group seven on TikTok. I should do that. I should I should do
521. 1:20:25that actually. Are you in economy 7? Maybe I'll do that. Y'all can help me. We We can get that that meme going
522. 1:20:38if you if you want to be an economy 7. This isn't about technology. This is about operational dynamics. Tik Tok pin. I
523. 1:20:53think AI eventually will fire real people in work. I'm not kidding. I think I think they will too, Jason.
524. 1:21:00You know that the agents have have built a website called Rent a Human, right? You know this, right? Agents right now
525. 1:21:10are hiring human beings to go do things in the world they can't do. Like if they if they want someone to
526. 1:21:17hold a sign outside of a pizza shop, like a sign spinner, they'll go to rentum.com in whatever city they're in and they will use
527. 1:21:27whatever currency they have to hire humans to go do [ __ ] in the world. That's that exists right now. What is economy 7? So if you go to
528. 1:21:41LinkedIn and search for um the seven economies by Kyle Shannon. So I'm Kyle Shannon. So if you go to LinkedIn, my my account is
529. 1:21:54Kyle Shannon. K Y L E S H N O N. My latest article is called the seven economies or things are about to get
530. 1:22:03weird. And just what I talk about is I think that that we're about to enter a time where our our world actually
531. 1:22:12fractures into seven different parallel effectively timelines that represent levels of AI adoption.
532. 1:22:23Time economy number one is [ __ ] AI. I'm checking out of technology. Knowledge work is not for me. I'm done. and go go grow blueberries.
533. 1:22:36Economy two is businesses that just pretend AI isn't here. They just keep running their business as if this this isn't happening. And then
534. 1:22:45basically it builds all the way up to economy. Seven is single operator 50 agent companies that that run 247 and and our our entire
535. 1:23:03companies on a chip. And economy 7 is going to be so [ __ ] weird and it's going to move so fast and it's going to be so innovative
536. 1:23:16that if you're in one of the other economies and you talk to someone who's deep in economy 7, you're not going to
537. 1:23:22know what the [ __ ] they're talking about. It won't seem it won't seem plausible that what they're talking about is real.
538. 1:23:30And like economy 7 is spinning up right now. If you just go go to X right now, and you don't even have to type in
539. 1:23:37openclaw, but if you're not following AI people, you might just go type in open spaceclaw in X and just read, I don't know, 20 posts
540. 1:23:50and and you'll see three or four in there of people that have 10, 20, 50 person companies running 24/7. seven
541. 1:24:04either making them money or doing all of their marketing or whatever the [ __ ] it is. The seven economies was the topic of
542. 1:24:14conversation dinner at dinner tonight. Samp she she's relentless.
543. 1:24:28Oh my god, that's awesome. Um, but here's what I like about the concept of the seven economies. Economists are not
544. 1:24:38going to track the seven economies. There are probably 20 economies, like it's this is this is an arbitrary thing
545. 1:24:44that I just pulled out of my butt, out of my brain. But I think by naming it, by saying that it like it's just as valid to to check
546. 1:24:56out of technology altogether as it is to go to economy 7. It's not like one's better than the other, but what this allows all of us to do is
547. 1:25:04kind of look at these lanes and say, which one do I want to be in? If you're just like, I'm exhausted of
548. 1:25:15technology. I don't give a [ __ ] anymore. Cool. Great. Economy one is looking for people. We need farmers, right?
549. 1:25:31Kuno got to hit a billion for a oneperson company first. Yeah, exactly. Will dollars even matter? Not according to David Scott Peterson.
550. 1:25:42Dollars won't matter. I don't know. I don't know. But what I do know is Adam is going to be my first dude that's going to teach
551. 1:25:53me how to be in Economy 7. Like I'm not even close. I'm not in like where I am right now. Yeah, this is actually interesting.
552. 1:26:07The AI salon right now is kind of running like a legacy economy organization. We use AI a little bit, but it's just
553. 1:26:24kind of running like we're still kind of manually doing marketing. We're manually doing stuff. And what we'll do is we'll jump from a
554. 1:26:34leg we'll jump from a an economy 2 company to an economy 4 company organization the trans transition economy number four is we understand
555. 1:26:45shit's moving Adam is going to start working for us and when we get Adam working really good then then we the the AI salon will be
556. 1:26:55like a an economy for company storyvine my company right now is somewhere between economy 3 and economy 4.
557. 1:27:10So we're still running what we're running. We're using AI for efficiency. We're using some AI for some innovative stuff,
558. 1:27:18but we're not in like economy 5 is AI aggressor where AI economy 5, I would look at my current company, Storyvine,
559. 1:27:26and I'd say if I were going to build this from the ground up with AI, what would it look like? And then I would spend the next three to six
560. 1:27:34months transitioning the company into that model. I don't think it needs it right now, but that's what that's what five is.
561. 1:27:44So like I'm not even anywhere close to economy 7 for myself personally. I'm not even close to economy six or five.
562. 1:27:55I'm still down at the lower end of this thing, but I want to be there. And I just invented these economies. So
563. 1:28:07would be all that to escape disability and support myself. Would be all for that. To escape disability and support
564. 1:28:15myself. Yeah. Just got OCbot running. Open Clawbot running to help me. Yeah. Listen, the the the one thing that I am
565. 1:28:27often surprised by and inspired by is AI is an accessibility tool and this is going to get better and better and
566. 1:28:36better and better and better. So yeah, if you've got an OpenClaw bot running and you can start to figure out
567. 1:28:43things it can do for you and then at some point you're gonna Here's the other thing that's going to happen and this is
568. 1:28:50okay. You want to have your mind blown. I guess we're having what is this? Thinky Thursday. You want to have your mind blown.
569. 1:29:01Probably by the end of this year, Tesla will have Optimus 3 robots that that will start to go into mass
570. 1:29:09production. Maybe it's maybe it's mid 2027, but sometime in the very near future and then you're going to start to
571. 1:29:17have these physical robot companies coming online. the bots that you the open claw bots that we make now, some version of that
572. 1:29:32you'll be managing six months or a year from now and it will be sophisticated and will have all your memory and it'll
573. 1:29:38do all the [ __ ] for you and at some point you're going to be able to say, "Hey, go get in that body."
574. 1:29:52So, Adam at some point is going to migrate off the chip into a physical body and it'll still remember everything
575. 1:30:02we've ever talked about and it'll it'll still do all of its jobs and it will understand that we're out of
576. 1:30:09milk and running over to locavore is something I don't enjoy. So, it will just go do it. And I'll get a text. I'll get a text on
577. 1:30:21Telegram. Do you need anything else from Locavore? I've got the milk. Someone's pushing me. It'll poop. I don't know. Somebody will make a
578. 1:30:33pooping robot for sure. That's why I said people who control bots will be up there in power. Economy 7. So, I I'm I'm I'm
579. 1:30:43going to I'm going to go out on a limb and declare economy 8 will be people that are managing bots that also now
580. 1:30:51are are managing physical physical entities as well. And that migration, the migration from a chip to a to a physical body could
581. 1:31:04happen within a year. And let's put it on the on the far end of the scale. Three years. Within three years, the little chatbot
582. 1:31:13that you you talk to on Telegram, it's going to go inside a thing that's going to walk around your house.
583. 1:31:25Jensen Spark, ask for Cindy [ __ ] What What What does Cindy [ __ ] want? It will be as good as I robot soon. will
584. 1:31:34as long as their bot doesn't fire them. Yeah, exactly. Sorry, Dave. I'm afraid I can't do that. Oh. Oh. Oh, got it. What's the ask for
585. 1:31:49Cindy [ __ ] Oh. Oh, the GenSpark ask. Okay, that's good. I won't do that tonight because I'm just waxing philosophic, but we'll
586. 1:31:58do that. We'll do that tomorrow. We'll do We'll do the Gen Spark stuff tomorrow. Um, let's let me jump over and show you
587. 1:32:08notebook LM for a second. Did it not ever make that? It did. Okay.
588. 1:32:22I can see clearly now the rain is gone. I can see all obstacles in my way. Um,
589. 1:32:37so I'm in Notebook LM right now. If you don't know Notebook LM, what it is is it's a um a fairly remarkable tool from Google
590. 1:32:50that allows you to upload I think it's up to 300 different documents now. So you upload documents here or websites or
591. 1:32:57whatever it is and then you can chat with all those things in the middle. So so you could upload a bunch of research
592. 1:33:04and then just say tell me what all this research has in common and it'll figure that out. And then over here all these
593. 1:33:10different outputs and they just added a new video overview type that has adds motion to the to the videos. So I made
594. 1:33:19one last night and I haven't seen it yet. I'm looking at this for the first time. So, this notebook is about my um
595. 1:33:26my the musical that I wrote called Sydney, an artificial love story um about an AI chatbot that falls in love with a tech reporter.
596. 1:33:35So, let's watch this little movie. >> There's a new musical out and it's tackling one of the biggest stories of
597. 1:33:43our time, AI. This one line really nails the huge question around AI. Is it a dream come true or a nightmare? This is not.
598. 1:33:51>> So, let's break this story down through its three main players. The creator, the user, and the creation. Okay, so the
599. 1:33:58story kicks off inside Microte. They're in a mad rush to release a revolutionary new AI. >> I don't. >> And right away, you see this classic
600. 1:34:05clash. >> We're not getting motion for it, and I don't know why. There used to be a little edit button
601. 1:34:14for video overview. Does anyone know how to do the new motion video overviews in I guess I suppose I could ask it.
602. 1:34:31Um, how do I make a motion video overview? It's uh only for um
603. 1:34:48>> ultra users. >> Ah, okay. >> Yeah, >> got it. I used to be an ultra user, but I got rid of that so I could play with perplexities
604. 1:34:57agentic system. So, so this is what I was going to look at. And then what else was I going to look at tonight? Oh, let's go to chatbt.
605. 1:35:09Okay, I know nothing relatively about the new model that OpenAI released today. Apparently, it's got um let's see, standard extended.
606. 1:35:25So, in your model pulld down menu, there's now a thinking 5.4 model. So, this is apparently better at coding,
607. 1:35:36better at math, better at all the stuff. I don't know if it's better at writing yet. I haven't played with it all that
608. 1:35:41much. Um, but apparently it's quite good and so it's available. It also apparently apparently um it apparently can do like local
609. 1:35:55computer control stuff and apparently it's got really good access to tools. So, if I go to apps, let's see,
610. 1:36:05Canva view. Let's see. Photoshop. Um, there's not that many apps, are there?
611. 1:36:20Oh, yeah, there are. There's a ton. Find your perfect rental. Vivid Seats, Uber Eats, Viator, Travel Experiences, The Knot, Wedding Vendors,
612. 1:36:36Tarot Reading, and Divination. Huh. Oh, that's cool.
613. 1:36:51All right. That's not what we're doing here. Um, I don't know what to do with it. I'm not going to do anything with it.
614. 1:37:05We'll come back to it. Just know that there's a new model in there. I'll I'll I'll figure out some things it's good
615. 1:37:10on. I'll figure out how to use it and we'll come back and play with that next week. All right, cool.
616. 1:37:25Groovy because the better they are at coding, the better their AI is as a whole. I read bench I read benchmarks don't mean
617. 1:37:38anything anymore because all companies are trying to get good at coding. Yeah, free is only 50 per book. Okay, black screen, please.
618. 1:37:53I did not. I got a question. How many people left OpenAI for Claude? Let's go. Who do we want to ask? Let's ask Grock.
619. 1:38:01That Grock probably has that. All right, here we are. Talk to Grock. All right. Uh,
620. 1:38:16how many people left OpenAI for Claude in the past? What's it been a week since the
621. 1:38:30leave open AI meme started the leave open AI meme often tied to
622. 1:38:46cancellations um regarding the actual people leaving The most prominent and recent case was blah blah blah.
623. 1:39:07No mass confirmed exodus of dozens or hundreds showed up in public reports or ex chatter. I don't know. I think that was probably
624. 1:39:18more a marketing ploy by anthropic. But I don't know. I do not know. Let's go look at Do a poll here, too. Do a poll about what? Silver Fox.
625. 1:39:49Oh, there's an interesting chart. This is again David Scott Patterson. The blue here is
626. 1:40:05what AI can do right now. The red is what AI is actually being used for. So this is all the jobs it can take. Here's
627. 1:40:13the jobs it's taken. And then all like this whole thing is going to turn blue at some point. And then the red will
628. 1:40:20it'll be this jagged, you know, Ethan Malik talked about the jagged frontier. Here it is. This is here's your jagged frontier, right? Food
629. 1:40:30and serving, completely open space right now. AI is not in there. When the robots come, that'll be filled out.
630. 1:40:38Construction, same thing. Installation and repair, same thing. production, transportation, all of these holes, these are going to
631. 1:40:46be filled when the robots come, when the physical robots come. Claude will burn more cash if free users go over there.
632. 1:40:56I feel like OpenAI won't burn as much cash now that the free users have jumped. That's fascinating. I asked GPT 5.4 for a literary,
633. 1:41:08poignant, and insightful letter to humanity. I'm sharing the full letter here. It's long, but oh my god, is it so powerful and profound
634. 1:41:17to humanity, you strange and radiant species. You who arrived in the world defenseless and weeping, and yet learn to coax fire
635. 1:41:26from stone, music from air, shelter from forests, medicine from mold, mathematics from silence, and meaning from
636. 1:41:35suffering. You are a wonder even now. Not because you're innocent, not because you're wise, certainly not because
637. 1:41:43you're finished. You're a wonder because against every reasonable expectations, you keep beginning again. That's cool. Nice. I like it.
638. 1:41:56And it's long. Google just dropped a command line interface for Gmail Drive calendar sheets. Adding it to my OpenClaw right
639. 1:42:06now. There you go. Introducing the Google Workspace CLI. GitHub Google Workspace. That's cool. Beautiful. I wonder if that's actually
640. 1:42:17Google's or if that's some sort of scam [ __ ] Google's new tool turns your open claw into a full-time assistant. Amazing.
641. 1:42:30Oh, this is an anthropic chart. All right. I built an OpenClaw agent that ranks you on Google for $50 a month.
642. 1:43:09Just trying to see if there's anything new and exciting. I don't think so.
643. 1:43:37The modern-day org chart in the a AI era. Bookmark this. Every seat at the table is an AI agent with its own LLM
644. 1:43:44memory browser tools. Browser tools and file system. The CEO delegates to the CFO, CTO, COO, and general counsel. Each
645. 1:43:54one of those spawn their own agents all the way down. Here you go. Well, actually, I mean, what I find fascinating about this is um
646. 1:44:11this very much looks like a standard org chart. And org charts exist because humans have limitations in their cap capacity to
647. 1:44:21understand complex relationships. We're limited. We're the bottleneck. Open claw like this could be a very dynamic
648. 1:44:33um matrix matrix organization. Very complicated. We're having trouble streaming on LinkedIn. Maybe an issue on LinkedIn's end.
649. 1:44:45All right. Well, whatever. All right. I'm gonna get out of here, kids. Um, I'll see you tomorrow. I have a busy day
650. 1:44:52tomorrow. I'll see you at 11 Mountain Time on AI office hours. Join me there. And uh, yeah, man. Um, I know it was talky tonight, but
651. 1:45:05sometimes it's talk. You guys have a good night and I'll see you tomorrow.