3/31/2026 - Jack Dorsey’s Plan to Replace Corporate Hierarchy with AI Intelligence Layers

Transcript

1. 0:00campy. Huh?
2. 0:21Meow. We will be happy.
3. 1:12What is
4. 1:39Best lead singer ever. You're damn right. That champ Shannon, he's a hell of a hell of a dog. Load the car and write the note.
5. 1:52Tell Grab your bags and grab your coat. Tell the ones that need to know we are headed north.
6. 2:11One foot in and one foot back. Well, it don't pay to live like that. So, I cut the ties and jump the tracks.
7. 2:28For never for never to return. Brooklyn, Brooklyn, take me in.
8. 2:42Are you where the shape I'm in? My hands they shake, my head it spins. Oh, Brooklyn, Brooklyn, take me in.
9. 2:58When at first I learned to speak, I used all my words to fight with him and her and you and me.
10. 3:14Yeah, it was just a waste of time. It was such a waste of time.
11. 3:29That woman, she's got eyes that shine like a pair of stolen polished eyes. She has to dance. I satisfied.
12. 3:45I'll see you in the morning time. Brooklyn, Brooklyn, take me in. Are you aware the shape I'm in?
13. 4:03My hands they shake my head. It spins. Oh, Brooklyn, Brooklyn, take me head. Three words that became hard to say.
14. 4:20I am loving you more than I am today. Look at the things I do.
15. 4:38Brooklyn, Brooklyn, take me in. Are you wear the shape I'm in? My hands they shake. My head it spins.
16. 4:53Oh, Brooklyn, Brooklyn, Brooklyn, take me in. Oh, there's a lovely song from a
17. 5:07brothers. Hello, a brothers. What's shaking? What's going down? Wolfman Jack.
18. 5:30Tik Tok has characterized this as performing live karaoke with a pop tribute. Welcome to live karaoke night here at
19. 5:45the AI learning lab. It's dog karaoke.
20. 6:26Genie in a bottle is the rising star of the night. Congratulations in a westerly direct show.
21. 6:41This car is my train. I've been driving. I've been wondering what it is I'm running from again. Feel like an 80y old man.
22. 6:59Holding on to 29. And uphead on that horizon is a California line.
23. 7:17Double hit up trucks car scaring a wide load. Refab house cut in half. Cute little front door and two windows. My love
24. 7:29ain't sure whether to cry or laugh. You see, I broke a home up myself once I stumbled to that dog. I read a note about the dogs li
25. 7:50anymore. Well, I've had enough. All this freedom on my road. Sing along, Chappie. No, it's good with decision.
26. 8:04At least that what I've been told. Hey, you know something? I realized this
27. 8:19sort of halfass janky ass dog singing people singing same 17 songs thing. Um, when all the AI learning labs out there
28. 8:34are just AI avatars doing perfect [ __ ] here's the latest news of the day in artificial intelligence.
29. 8:41Everything's faster. Everyone's making more money and more people are picking blueberries. When they're all doing that, just perfect.
30. 8:51I'll just be here being a human being. You're like, "Oh yeah, that one's real." Listen to the dog and the whatever is that singing.
31. 9:13You're the authentic one. Ha ha. My name is Shannon. I'm the authentic one. Is his guitar out of tune? I don't think
32. 9:28it's ever been in tune. Hey, if you're on YouTube, do we have anyone on YouTube? YouTube or Twitter or
33. 9:43LinkedIn? Do make a comment for me. Like, let me shake make sure my comments are working. I know there's nothing to
34. 9:49really comment on yet. It's like, uh, you know, it's night time, right? You don't need to wear sunglasses.
35. 9:58>> Champion Kyle. Keeping it real. Authenticity is the the big thing of the day. You and I here all alone.
36. 10:16Sunday morning at home. Skye's blue and a coffee is strong. It's true. And I open my eyes to a dream realized
37. 10:33in front of me. And I had got a clue what in the world is happening to me. Think I think I'm happy.
38. 10:49Like first day summer vacation happy. Got to get a little rest and relaxation. Happy like the choir on Sunday morning singing true.
39. 11:31Yeah. Yeah.
40. 11:53That's a seven and garunkle song. Um.
41. 12:11Uh. Uh. Oh, there's no comments coming in. Anyone on YouTube? Is anyone on YouTube?
42. 12:25Can someone from Tik Tok hop over to YouTube and post a comment? >> There's one. We got someone from Silver Fox. Oh,
43. 12:31>> hang on. Let me reload. Oh, wait. I don't want to reload. Hang on. I need to go away from it and come back. There we go. Hey, chief.
44. 12:43Okay, commenting. Okay, we've commented. Comments are working now. I thought something was weird. Something was
45. 12:48weird. It was my It was my window. I had a window refresh issue. Yes, Mr. So, >> it wasn't weird. It was a regular Thank you. Thank you.
46. 13:04Oh, yes. Yes. Yes. Yes. It is the 31st of March. The quarter is complete.
47. 13:55Whatever time I see you now, get that looking. Every time I see your mouth, I hear that smile. The early misty morning light heard the
48. 14:12engine turning. The old foot outside. I was thinking about So the new Sunno 5.5 is supposed to do good at cloning your
49. 14:27voice. So, I was thinking of doing that. I don't know if I want to do it live because I don't know, performance
50. 14:34anxiety. You know, what if I can't get it musically up and I just hose it and make a model on my shitty singing voice.
51. 14:49That would be not fun, wouldn't it? Scott Parker, hey there. Can't wait to tell you how participating early on just
52. 14:59getting ready to change my life. Oh yeah. Scott Shioone Parker is a longtime watcher and a rising star and you're live.
53. 15:10If you're watching Kyle and Chappie here Chappie here in the AI learning lab, what are you doing with your life that's more worthwhile?
54. 15:21Well, here's the thing. If you if you come here expecting to get anything from me other than maybe, you
55. 15:28know, a smarmy ass joke once in a while, um, that ain't what this is about. What this is about is being in the conversation
56. 15:39with the people that come here >> and understanding that um, there's a lot of stuff to figure out. We're We're
57. 15:54This next quarter is going to be pretty intense. The first quarter was pretty [ __ ] intense, right? The first three
58. 16:00three months of the year, we got We got Opus 4.6 in December, which I know is not this year. Shut up. I I I know how to do months,
59. 16:10but we got Opus 4.6 in in December, where all the programmers like, "Ah, [ __ ] All right. It's good enough now to do my job.
60. 16:21God damn it. I used to like coding. Now I'm just babysitting agents. This sucks. So that happened in December. Open Claw
61. 16:30comes out in January and it's like, "Hey, you want your own little baby AI that's going to hang out with you and do shit?"
62. 16:39Everyone lost their [ __ ] mind. The the sales of Mac minis went through the roof. Um, you know, Seed Dance 2.0 came out, the
63. 16:55the Hollywood thing. Oh, did you see that? Hollywood was like, "We're going to sue you for copy." Hey, could we get
64. 17:01access to your little model there? We'll we'll give you $2 million. Oh, you'll take it? Okay, let's go. That's awesome.
65. 17:14[ __ ] [ __ ] Hollywood. Everyone's up in arms still. They're like, "Oh, we could get a movie, too. Joy Party's here." Joy Pertie's like,
66. 17:24"Yeah, I'm making movies. Anybody need an AI filmmaker?" Hollywood's like, "Uh, yeah, we do. We got the seed dance thing
67. 17:32here. Do you could you turn that into a Hollywood blockbuster? We're not going to tell Russell Crowe if that's good with you."
68. 17:42No, we've already got his likeness. We've got that from some other movies. We're just put it Joy, if you could just
69. 17:48We're going to put you in the back room. There's a What's it You There's still some brooms in it, but if you could work
70. 17:57back there and just come come in the back door with it. Don't drive a fancy car. Okay. In the basement, Joy Party. We We got We
71. 18:08We keep people like you in the basement. We're calling it Hollywood 2.0. I'll definitely come back soon. Got a
72. 18:19couple of things in soul procurement I need to testify. I'll testify soon. Okay, good. Beautiful. There was some really interesting news
73. 18:30today. We'll get to I'm going to I'm going to just blast through X tonight because there's a bunch of a bunch of
74. 18:35really interesting stuff happened today.
75. 19:26The boxer. Sitting in sitting in railway station now taking a ticket to my destination.
76. 19:53Hey, how do I know if this channel's real? Play the guitar. Okay, it's real.
77. 20:11Chimpy, you want to sing little A minor?
78. 20:33There was a single card David played and it plays the Lord. You don't really care for music, do you?
79. 20:51goes like this. The fourth fifth round of fall lives king. Hallelujah.
80. 21:07Hallelujah. Hallelujah. Hallelujah.
81. 21:22Hello. Yeah. All right, let's get this show on the road. Enough of that singing them sappy
82. 21:37songs. They're making me all sad. You put me to sleep. Wake up, people. We got some AI to do. Good lord. All right,
83. 21:52friends pap. Hello friends. Just got in from the AI documentary. It came out Thursday. Highly recommend. Very nice.
84. 22:00Here we get education, community, group learning, great conversation, relationship building, creativity. Exactly. That's why you come here.
85. 22:10I am just the I am just the carrier of the community. That's it. I show up. I show up. You show up. That's what we do.
86. 22:21Lori Kay is a longtime watcher and a rising star of the AI learning lab. Congratulations, Lori. You're a rising
87. 22:28star. Do I get something? No. I know, champ. It's exhausting. I know. I'm tired of it, too.
88. 22:45You done? What do you want? Cheese. I said cheese. He wagged his tail. Invites to AI learning lab. If you see a
89. 22:56guy playing guitar singing with a dog, yeah, just scroll Tik Tok until you see a guy singing with a dog. All right. Um, let's get let's get
90. 23:11going. Let's get going. Let's get this show on the road. I tell you what. I tell you what. I TELL YOU WHAT. OH YEAH. MHM.
91. 23:22I tell you what. I tell you what. We'll put it. We're going to share a tab in case there's some sort of audio we
92. 23:29want to listen to. All right. So, we're on the Twitter. It's technically called X. Uh Elon Musk. Uh Elon Musk purchased Twitter uh for
93. 23:42$42 billion, of course, in a famous transaction that he didn't want to actually complete, but uh uh it was a
94. 23:48was a a failed negotiation. Anyway, $42 billion. It's now worth about 20 $18 billion, something in that neighborhood.
95. 23:57It's so he parlayed the 42 billion into the 18 billion. That's it was a parlay. It was what they call a reverse parlay,
96. 24:03a negative parlay. And uh and then but but then what he what he wanted to do was uh he didn't like the bird. Um I I
97. 24:13think the bird reminded him of his childhood. I don't I don't know what it was. And he liked the letter X uh
98. 24:17because of airplanes and so he named it X. And so that but it used to be called Twitter. Uh what's going on over here? I don't have enough
99. 24:33black bar on the bottom for the Twitter people. Hang on people, I'll be with you. Just everybody calm down. Everybody calm down.
100. 24:46Fantastic. There you go. There you go, Twitter people. That's a little more better. You can see it.
101. 25:00Okay. So, one of the things I wanted to show you, was it this one? Yes. So, apparently,
102. 25:16um, Claude Codes source code was leaked yesterday. Listen to this timeline. So this is shrudy uh on on uh on X. Anthropic leaked
103. 25:31512,000 lines of claude code source code yesterday. What happened in the next next 12 hours is absolutely wild.
104. 25:414:00 a.m. Enthropic pushes an update to npm. Inside the package, their entire codebase, a 60 megabyte debugging file
105. 25:52accidentally bundled in. 23 minutes later, researcher uh Chow ofen Shu spots it, downloads the zip, posts it on X. Within six hours, 3
106. 26:06million views. So by 10:00 a.m. there's three million views of the uh of the downloaded source code of Cloud Code. By
107. 26:15the time Anthropics team woke up, the code was forked 41,000 times across GitHub. Anthropics started firing DMCA
108. 26:24takedowns. Too late. A Korean developer developer named Sigrid Jyn woke up with his phone exploding. He's Claude Code's
109. 26:33biggest power user. Wall Street Journal reported he burned through 25 billion tokens last year. He read the leaked
110. 26:40code. He wrote he he rewrote the entire thing in Python in eight hours. His repo hit 30,000 stars, faster than any GitHub
111. 26:49project in history. Then he rewrote it again in Rust. That version now has 49,000 stars. Someone mirrored it to a
112. 26:57decentralized platform with one message will never be taken down. The code is permanent. Enthropic cannot get it back.
113. 27:09Here's the part I can't stop thinking about. Anthropic built in something called undercover mode. Its only job,
114. 27:15prevent Claude from accidentally leaking internal secrets. They shipped an entire anti-ak system in their own product,
115. 27:23then leaked their own source code in a map map file. Uh oh, someone's unemployed. And in fact, the the guy that leaked it posted
116. 27:34his uh I no longer work at Anthropic. Uh here, Kevin Notton Jr., Kevin Notton Jr. wrote, "I was fired from Anthropic today.
117. 27:46I was the engineer responsible for shipping the latest dev cloud code npm package. Wanting to improve the
118. 27:52debugging experience for the team, I decided to include some source maps in the release. This resulted in the entire
119. 27:59internal codebase being publicly exposed including thousands of files and every agent command all system prompts the
120. 28:07complete query engine undercover mode bypass permissions mode and our internal telemetry configuration. So so anthropic so anthropic is more
121. 28:18open source than open AAI right now. I take full responsibility. I genuinely believe the safeguards Claude code had
122. 28:24built in for me would be adequate and it was a serious mis miscalculation on my part. My actions have unintentionally
123. 28:31open sourced major parts of Claude's architecture well ahead of schedule although well ahead of schedule means
124. 28:38that they were planned for for being open sourced. It sounds like I apologize to the team and Claude. And then if you
125. 28:44go to his um his thing, it says recently exanthropic looking for work. Poor dude. Oopsie. Exactly. Why would
126. 28:56you share that with the world? Not great for future hiring opportunities. Well, I don't think he I don't think he intended
127. 29:02for it to get out. He But I don't I don't know. I mean, clearly he clearly it was a miscalculation. And here's the
128. 29:09deal, man. Developers are [ __ ] brutal. If you if you make one mistake, they will find you out, you know. Crazy. It's crazy. I'll tell
129. 29:22you. It's crazy. I'll tell you. It's crazy. All right, let's see. A simpler tech. If you could figure out HTML, you could play
130. 29:38the href. was the soul in ovation that changed the world. AI
131. 29:54is like a thousand hrefs a day.
132. 30:10Oh, you can't see this. Share this tab instead. Okay. Oh, here's another one. Okay. All right. This was the other one I wanted to talk
133. 30:21about. Um, so we y'all need a homebased data center equivalent to Anthropics where we can have our own cloud code at home. Well,
134. 30:33you're not I mean the the thing about Claude Code's source being uh put out there is that people are going to be
135. 30:40able to take open source models and make them really good because Claude Code was the thing that was it was the wrapper
136. 30:47around their models that was the good part and that part's now been leaked. It's amazing. Okay.
137. 31:01So why use anthropic now? Well, that's what we'll see. We'll see what the I mean, I just have a feeling there's
138. 31:08going to be a whole bunch of Claude code clones now that are based on Claude code. And then what what will be cool is
139. 31:14there will be some developers that make it better. So this will lead to innovation, but crazy. Okay, Jack
140. 31:20Dorsey. Let's read the Jack Jack Dorsey post. It's long. You want to go in with me? I'm not going to read that whole [ __ ]
141. 31:29thing, but we're going to read we're going to read a lot of it. So, Jack Dorsey, if you don't remember, he um
142. 31:40he founded Twitter. He founded the Twitter back when it had a little birdie and it was blue. The only thing that's
143. 31:46blue that remains is the check mark. Jack Dorsey founded Twitter and his most recent company is called Block. I don't
144. 31:57even know what Block does, but Block has 10,000 employees. No, Block had 10,000 employees. And about two weeks ago, Jack Dorsey fired
145. 32:104,000 of his 10,000 employees and said, "Here's the thing. This is not because we're hurting for money. We are
146. 32:17profitable and we are growing. We could have kept you on. We're choosing not to because our employees armed with intelligent AI
147. 32:28working in smaller flatter teams are changing how we start and run companies. That was one of the lines within
148. 32:37within his document firing those 4,000 people laying them off. And so today he put out this thing. I think it was
149. 32:45today. Was it today? Jack, where's the date? There's no date here. article article 10 hours ago at Sequoia we see the speed we see that
150. 32:59speed is the best predictor of startup success most companies are focused on AI AI as a productivity enhancer now I want
151. 33:07to show you something I want to show you something this oh you can't see it I got to change how I'm sharing everybody calm down
152. 33:17everybody calm down. All right. Okay, Kyle, we'll calm down. We're calm now. Okay. All right. Oh, yeah. The seven
153. 33:26economies. You talked about that before. Okay. Look, here's here's Jack Dury. What does he say? At Sequoia, we see speed as the best
154. 33:38predictor of startup success. Most companies are focused on AI as a productivity enhancer. See that right there in the in the in the seven
155. 33:49economies. That's a economy. Number three is the efficiency economy. AI is cool. It makes our business more efficient.
156. 33:57I'm going to take AI and I'm going to use it to take what we currently do and make it more efficient. Okay, that's his his second sentence.
157. 34:07Few are focused on the potential of how AI of AI to change how we work together. few, right? So, the transition economy are companies
158. 34:20that do see that possibility of AI to to innovate things, but they're basically going to keep things as they are.
159. 34:28But he says, "Block is showing is showing what it looks like to fundamentally rethink organization design, ultimately harnessing AI to
160. 34:37increase speed as a compounding competitive advantage. Block is showing what it looks like to fundamentally
161. 34:43rethink organ organization design. That's economy five, the aggressor economy. Burn the org chart. Let's reinvent ourselves. So this is his
162. 34:54manifesto on doing that. 2,000 years before the first corporate org chart, the Roman army solved a problem a very large organization still
163. 35:03faces. How do you coordinate thousands of people across vast distances with limited communication? Their answer was a nested hierarchy
164. 35:11within a const a consistent span of control at every level. The smallest unit was the uh conternium conternium
165. 35:22eight soldiers who shared a tent equipment and a mule. Remember when I talked about groups of six to eight
166. 35:29people, cohorts within the AI salon? I talked about we should we should find ways to create persistent groups of six
167. 35:38to eight people collaborating and working together. We should experiment with that in here. Maybe we'll kick that
168. 35:44off. So in about well so so here's the thing. April 30th marks the 3r year anniversary of AI Learning Lab going
169. 35:52live on Tik Tok. And for the past year it's been live on Tik Tok and YouTube. We're going to be shifting this inside
170. 36:00the AI salon because I want to get back to the AI salon being this this really vibrant collaborative place. Maybe we
171. 36:08play around with this structure in our community. There's something here. Anyway, eight soldiers who shared a tent,
172. 36:15equipment, a mule led by a danis. I don't know what a danis is. 10 uh contraburnia formed a century of 80 men
173. 36:25under a centurion. Six centuries made a cohort. 10 cohorts made a legion of roughly 5,000. At each layer, a named
174. 36:32commander held defined authority, aggregated information from below, and relayed decisions from above. The
175. 36:39structure 8 to 80 to 480 to 5,000 was an information routing protocol built on a simple human limitation. A leader can
176. 36:49effectively manage somewhere between three and eight people. Romans discovered this through centuries of warfare. Even today, the US Army's
177. 36:57hierarchical chain follows a similar pattern, what we now call span of control, and it remains a governing
178. 37:04constraint of every large organization on Earth. The next big change came from Prussia. After Napoleon's army destroyed
179. 37:11the Prussian forces at the battle of Jenna in 1806, a group of reformers led by Scarnhorst and Jezanau Jizenu Jizanu
180. 37:22Nissa Nissan now I don't know rebuilt military around an uncomfortable truth. You cannot depend on individual genius
181. 37:30at the top. You need a system. He created the general staff, a dedicated class of trained officers who job whose
182. 37:38job was not to fight but to plan operation operations process information, coordinate across units. Scarnhorst uh intended these staff
183. 37:47officers to support incompetent generals providing the talents that might otherwise be wanting among leaders and
184. 37:54commanders. This was middle management before the term existed. professionals whose purpose was to route information,
185. 38:01precomputee decisions, maintain alignment across complex organizations. The military also formalized the distin
186. 38:08distinction between line and staff functions. Line advances the core mission. Staff provides specialized
187. 38:16support. Every corporation still uses this vocabulary today. Military hierarchy. Anyone saying anything? So, we all need a homebased
188. 38:26data set. Okay. The sort of thing that prevents elite capture many people were afraid of. All right. Roman Empire.
189. 38:41That small group dynamic sounds interesting, Kyle, doesn't it, Cam? So, so the basic idea was thinking about
190. 38:49this is this is inspired by um in in content evolution. I run this group called Collab and it's it's evolved into
191. 38:57being about eight of us that show up on a regular basis, seven or eight of us. And like we've been working together now
192. 39:04for three years and like we know each other. We know each other's businesses are, but we don't we don't collaborate
193. 39:09in our businesses necessarily, but when we come together, we can talk about stuff and get stuff done and pay
194. 39:15attention to the news and and prototype stuff if we need to. And I just thought, wouldn't that be a fascinating? And we
195. 39:22we kind of have that informally in the salon right now. We've got little groups of of, you know, irregulars and things
196. 39:29like that that that pile up and and travel around in packs. What if we formalized that as as a as a kind of
197. 39:36unit that you could sign up for a cohort and and just meet with them on a regular basis? I don't know. I think there might
198. 39:43be something there. Anyway, military hierarchy entered the business world through the American railroads in the
199. 39:491840s and50s. The US Army lent um West Point trained engineers to private railroad companies. These officers
200. 39:56brought military organizational thinking with them. Staff and line hierarchies, divisional structure, bureaucratic
201. 40:03systems of reporting and control. All of it was developed by the military before the re railroads adopted it. In the 1850s,
202. 40:10um, Daniel McCllum of the New York and Eerie Railroad created the world's first organizational chart. Let's go find it.
203. 40:18Daniel McCllum, world's first organizational chart. I'm sure this exists somewhere. Images. There it is. Look at that.
204. 40:38What's happening? Oh, funny. I set a timer on my watch. It was buzzing. Dr. J, there's something there. Okay.
205. 40:52With the with the small groups, right? Isn't there something cool there? Introducing the organization chart. Holy [ __ ] look at this thing.
206. 41:01Can I copy image? Holy [ __ ] this is cool.
207. 41:17Oh, you can't read it. It's too small. That's too bad. New York Eerie Railroad representing a plan of organization. Wow. That's crazy.
208. 41:32There's the first org chart, huh? Who knew the informal management? Wait, let's
209. 41:47see. Uh, the informal management styles that worked for smaller railroads were failing. Train collisions were killing
210. 41:54people. Macllum's chart formalized the same hierarchical logic. Romans had used layers of authority, defined reporting
211. 42:01lines, structured information flow. It became the blueprint for the modern corporation. YouTube comment. Um, I
212. 42:09worked for one for 10 years. That looks like a railroad organization chart. All right. That's wild. BNSF railroad.
213. 42:16That's that's wild, dude. Amazing. Um, okay. Let's see. Frederick Taylor 1856 to 1915 often called the father of
214. 42:26scientific management optimized what had become what what happened within that hierarchy. Taylor broke work into
215. 42:33specialized tasks. So this begins okay so 18 so this is probably he was born in 1856 so probably around 1900
216. 42:44Taylor breaks work into specialized tasks tasks right end of the 1800s early 1900s specialized tasks
217. 43:00we train for skills you're going to do your tasks tailorism here we Go. This produced functional pyramid organization, a structure structure
218. 43:10optimized for efficiency. Oh, wait. Taylor broke work into specialized tasks, assigned them to trained experts, and managed through
219. 43:20measurement rather than intuition. Right. So, KPIs. What are your KPIs, people? Key performance indicators. Huh? This
220. 43:31produced a functional pyramid organization, a structure optimized for efficiency within the information routing system that the military had
221. 43:38pioneered and the railroads had commercialized. The first rail stress test of functional hierarchy came during
222. 43:46World War II. The Manhattan project required physicists, chemists, engineers, metallurgists, and military
223. 43:53officers to work across disciplinary boundaries toward a single objective under extreme secrecy and time pressure.
224. 44:00Robert Oppenheimer organized Los Alamos into functional divisions, but insisted on open collaboration across them.
225. 44:08Resisting the military's instinct to compartmentalize, and NASA compartmentalizes. When the implosion problem became
226. 44:16critical in 1944, he reorganized the lab around it, created crossf functional teams unlike anything in corporate
227. 44:23America at the time. It worked, but it was a wartime ex exception led by a singular figure. this kind of question.
228. 44:32Wait, the question the post-war business world faced was whether that kind of crossf functional coordination could be made routine.
229. 44:41With the growth in globalization of companies after World War II, the scale limitations of functional design became
230. 44:49acute. In 1959, McKenzie's Gilbert Cle and Alfred Dipio published creating a world enterprise in the Harvard Business
231. 44:57Review, providing an intellectual framework for a matrix organization that combined functional specialties with divisional units.
232. 45:07Fascinating wild stuff because it's all about to blow up. It's all about to blow up, which is what this article is about. Under the
233. 45:18leadership of Marvin Bower, McKenzie helped companies like Shell and GE implement these principles, balancing
234. 45:24central standards with local agility. This became the professional or modern corporation that propelled the post-war global economy.
235. 45:34Over time, other frameworks uh came in to address complexity, rigidity, bureaucracy. The McKenzie 7S framework
236. 45:42developed in the 70s by Tom Peters and Peter Robert Waterman distinguish the hard S's strategy structure systems and
237. 45:51the soft S's shared values skills staff and style. The core idea was structural elements alone were insufficient.
238. 45:59Organizational efficiencies required alignment across cultural traits and human factors blah blah blah. Okay. Something coming from Brandon.
239. 46:15The reading something learning lab. Oh, the railroad learning lab. Okay. In more recent decades, technology
240. 46:26companies have experimented aggressively with organizational structure. Spotify popularized crossf functional squads
241. 46:33with short sprint cycles. Zappos attempted hocrisy eliminating ma management t titles entirely. Um we faced this when we did agency.com
242. 46:44we tried all sorts of different structures. Where we ended up in the end was crossf functional 25 to 30 person
243. 46:52teams that had basically were like little mini agencies. So we had um we had like a VP that was we had VPs for
244. 47:02head of the disciplines and then within a team you would have project management, account management, creative
245. 47:09um technologists and site builders. So front end and back end basically and they would all sit in you know
246. 47:16physically colllocated units um with groups of four or five sitting together. So and it it ended up working really
247. 47:24really well. Champ, what do you want? You want cheese? You want mama? What do you want? All right, hold, please.
248. 48:00Ow, my knees hurt. Okay. Valve operated with a flat structure nor no formal hierarchy. Each of these experiments revealed that something
249. 48:09about the limitations of traditional hierarchy, but none solved the underlying problem. Spotify moved back
250. 48:15toward conventional management as it scaled. Zappos saw significant attrition. Valve's model proved difficult to scale
251. 48:24beyond a few hundred people. As organizations grew into thousands, they revert to hierarchical coordination.
252. 48:29Okay, so now we're moving on. The constraint is the same one the Romans faced in the Marine Corps rediscovered
253. 48:37in World War II. Narrowing span of control means adding layers of command, but more layers means slower information
254. 48:44flow. 2,000 years of organizational innovation has been an attempt to work around this trade-off without breaking
255. 48:50it. Okay, here's the setup. This is the setup. 2,000 years of organizational innovation has been an attempt to work around this
256. 49:02trade-off. The trade-off being as you narrow the span in a large organization, you increase layers and you slow down communication.
257. 49:12At block, we're questioning the underlying assumption that organizations have to be hierarchically organized with
258. 49:19humans as the coordinating mechanism. Instead, we intend to replace what the hierarchy does. Most companies using AI
259. 49:28today are giving everyone a co-pilot, which makes the existing structure work slightly better without changing it. Economy three,
260. 49:40the efficiency economy. We're going to take what we currently do and make it slightly better. This is what Jack's talking about. He's in
261. 49:47economy five. He's he's talking about being in economy five. He's telling us how to be an economy 5. Okay.
262. 49:58Most companies today are using AI are giving everyone a co-pilot, which makes the existing structure work slightly
263. 50:03better without changing it. We're after something different. A company built as an intelligence or a mini AGI. We're not
264. 50:12the first one to move beyond traditional hierarchy. Her hires renden hay model platform organizations datadriven management. These are real
265. 50:22attempts at the same problem. What they lacked was a technology capability of actually performing the coordination
266. 50:29functions. So basically what he's going to say is he's going to turn loose small pods of teams and let the AI organize it
267. 50:37and track it and collect speed up make instant the communication challenge. This is wild. Um what they lacked was a technology
268. 50:50capable of actually performing the coordination functions that hierarchy exists to provide. AI is that technology.
269. 50:59For the first time, a system can maintain a continuously updated model as an entire of an entire business and use
270. 51:08it to coordinate work in ways that previously required humans relaying information through layers of management.
271. 51:17There's going to be no middle management in his his organization. And there's going to be probably a seuite and then pods
272. 51:27and AI will just manage the pods. Go do this [ __ ] I need this [ __ ] done to for what these guys want. You all go do my [ __ ]
273. 51:36It's amazing. Um, for this to work, a company needs two things. a kind of world model of its own operations
274. 51:45and a customer signal rich a customer signal rich enough to make that model useful. Block is remote first.
275. 51:53Everything we do creates artifacts, decisions, discussions, code, designs, plans, problems, and progress all exist
276. 52:01as recorded actions. It's the raw material for a company world model. In a traditional company, a manager's job is
277. 52:08to know what's happening across their team and relay that context up and down the chain. In a remote first company where work is
278. 52:17already machine readable, AI can build and maintain that picture continuously. What's being built, what's being
279. 52:24blocked, where the resources are allocated, what's working and what isn't. That's the information the hierarchy used to carry. The company
280. 52:32world model carries it instead. This is [ __ ] brilliant. AI is the tech. Bingo. Yeah, exactly. This is brilliant.
281. 52:45Uh but the capability of the system is only as good as the quality of the customer signal. Okay. So the first thing he's saying is
282. 52:54if all of your business knowledge is locked in middle management's heads, you're [ __ ] If you're a remote company,
283. 53:10you've already got it digitally. So the AI can deal deal with the digital stuff. It can it can read the story. But now
284. 53:16he's saying, "But the capability of the system is only as good as the quality of the customer signal feeding it." And
285. 53:22money is the most honest signal in the world. People lie on surveys. They ignore ads. They abandon carts. But when
286. 53:29they spend, save uh send, borrow, or repay. That's the truth. Every transaction is a fact about someone's
287. 53:37life. Block sees both sides of millions of these transactions every day. a buyer through Cash App and a seller through
288. 53:44Square plus the operational data from running the merchants's business that gives the customer world model something
289. 53:51rare. A per customer per merchant understanding of the financial reality built from honest signal that compounds.
290. 53:59The richer the signal, the better the model. The better the model, the more the transactions. The more the
291. 54:04transactions, the richer the signal. Jack Dorsey's [ __ ] smart dude. So cool. Together, the world model and the
292. 54:16customer world, the the company world model and the customer world model form the foundation for a different kind of
293. 54:22company. Instead of product teams building predetermined road maps, you build four things. This is so cool. First, capabilities. The atomic
294. 54:32financial primitives, payments, lending, card issuing, banking, buy now pay later, payroll, and so on. These are not
295. 54:40products. They're building blocks that are hard to acquire and maintain and they have network effects and regul
296. 54:46regulatory permission. They have no no UIs of their own. They have reliability, compliance and performance target
297. 54:52targets. Second, a world model. This has two sides. The company world model is how the company understands itself in
298. 55:02its own operations, performance and priorities. replacing the information that used to flow through layers of
299. 55:08management. The customer world model is the per per customer per merchant per market representation built from prop
300. 55:17proprietary transaction data. It starts with the raw transaction data today and evolves towards a a full causal and
301. 55:25predictive model over time. It's amazing. So basically, you've got the company world model that understands the
302. 55:32AI understands here's all the [ __ ] I can do. I've got these financial infrastructure that I can deal with
303. 55:41money. I've got these skill pods that do [ __ ] And then I've got customer transactions down to the transaction
304. 55:51level and up to the merchant level that I can track and say, "Oh, that thing we did over here made those
305. 56:00numbers go higher over there. Do more of that. In fact, let's take three of these pods. You three pods. You're not doing
306. 56:06this anymore. Now you're doing that." And go do go do. It's crazy. Third, an intelligence layer. This is
307. 56:17what comp composes capabilities into solutions for specific customers at specific moments and delivers them proactively.
308. 56:26A restaurant's cash flow is tightening ahead of seasonal dip. The model has seen before. The intelligence layer
309. 56:32composes short-term loan from the lending capability, adjust the repayment schedule using the payments capability,
310. 56:39surfaces it to the merchant before they even think to look for financing. A cash app's user spending pattern shifts in
311. 56:48the way the model associates with a move to a new city. The intelligence layer composes a new direct deposit setup, a
312. 56:55cash app card with boosted categories for the new neighborhood, a savings goal calibrated to their updated income. No
313. 57:03product manager decided to build either solution. The capabilities existed. The intelligence layer recognized the moment
314. 57:09and composed them. So this is he's essentially talking about a company that's building custom dynamic solutions per customer. [ __ ] insane.
315. 57:21If you don't think the world's about to [ __ ] change, think about all of the systems that are hardcoded right now to do
316. 57:32everything he just talked about where you either have to build those systems or you interface with other players that
317. 57:39take a a cut of every transaction. He's talking about dynamically building those things per customer. It's insane.
318. 57:48Fourth, interfaces, hardware and software. Square cash app, Afterpay, Title, Bit Key, Proto. These are all
319. 57:54delivery surfaces which the intelligence layer delivers composed solutions. They are important, but they are not where
320. 58:00the value is created. The value is in the model and the intelligence. When the intelligence layer tries to compose a
321. 58:06solution and can't because the capability doesn't exist, that failure signal is a future road map. The traditional roadmap where product
322. 58:16managers hypothesize about to what to build next in any company's ultim is is any company's ultimate limiting factor.
323. 58:24In this model, customer reality generates the backlog directly. Customer wanted something, we couldn't deliver it. Dynamically design that
324. 58:34feature, deliver it to the customer. That's [ __ ] crazy. If this is what the company builds then the question becomes what do people do?
325. 58:48Good. The org structure follows from this and it inverts the traditional picture. In a conventional c company the
326. 58:56intelligence is spread throughout the people and the hierarchy routes it in this model. The intelligence lives in
327. 59:03the system. The people are on the edge. The edge is where the action is. The edge is where the intelligence makes
328. 59:09contact with reality. People reach into places the model can't yet go. They sense things the model can't perceive. Intuition, opinionated
329. 59:19direction, cultural context, trust dynamics, the feeling in a room. So the intelligence, the AI intelligence
330. 59:28is sitting at the center figuring all this [ __ ] out and then it's know it knows it has this interface of people
331. 59:34out into the world and with the other people in the organization. Lord digital gods. The seven economies include my pillow makes me sleepy.
332. 59:49This is really cool. They sense and if I if I'm boring you, sorry about that. That's okay. I mean, sometimes it's boring here.
333. 59:59Um, they make calls the model shouldn't make on its own, especially ethical decisions, novel situations, and high
334. 1:00:04stakes moments where the cost of being wrong is existential. A world model can't touch the world. It's just a
335. 1:00:11database. But the edge doesn't need layers of management to coordinate it. The world model gives every person at
336. 1:00:17the edge the context they need to act without waiting for information to travel up and down the chain of command.
337. 1:00:24This is [ __ ] brilliant. This is I I I had a feeling this was coming from what Jack wrote when he laid off 4,000
338. 1:00:32people, which is why I created this economy five, the aggressor economy. I basically made this surmising this is
339. 1:00:40what Jack was about to do. And he did it two weeks later after he laid off 4,000 people. So he obviously had this in in the in the can.
340. 1:00:51Um, in practice, this means we normalize down to three roles. Oh, here we go. You want to know your job of the future?
341. 1:01:02Hello, my friends. Hello, Mimi in the making. Okay. Three roles. Three human roles.
342. 1:01:19Three. Three individual contributors who build and operate capabilities. The model, the intelligence layer, the interfaces. They
343. 1:01:29are deep specialists and experts in a specific layer of our system. The world model provides the context that a
344. 1:01:35manager used to provide. So individual contributors can make decisions about their layer without waiting to be told what to do.
345. 1:01:43DRI directly responsible individuals who own specific crosscutting problems or opportunities and customer outcomes. A
346. 1:01:52DRRI might own the problem of merchant churn in a specific segment for 90 days with the full authority to pull
347. 1:02:00resources from the world model team, the lending capability team, the interface team as needed. Dr. DRRI may persist on
348. 1:02:08certain problems or move elsewhere to solve new ones. And then finally, player coaches who combine building with
349. 1:02:16developing people. This is brilliant. This is brilliant. So people do [ __ ] People figure out what the customer needs and make sure
350. 1:02:26that those resources are are are well allocated. And then player coaches that make sure people are developing.
351. 1:02:33They replace the traditional manager whose primary job was information routing. The player coach still writes
352. 1:02:39code. A player coach still writes code or builds model or designs interfaces. They also invest in the growth of people
353. 1:02:45around them. They don't spend their days in status meetings, alignment sessions, and priority negotiations. The world
354. 1:02:52model handles alignment. The DRRi DRRI structure handles strategy and priority. The player coach handles craft and people. It's really good.
355. 1:03:05There is no need for a permanent middle management layer. Everything else the whole whole old hierarchy did the system
356. 1:03:14coordinates or oh everything else the old hierarchy did the system coordinates and everyone is empowered with a role
357. 1:03:22that's much closer to the work and the customer. Block is in the early stages of this transition. It will be a
358. 1:03:28difficult one. Yeah, this is this is brutal. This is brutal and parts of it will likely break before they work. We're writing about it now
359. 1:03:37because we believe every company will eventually need to confront the same question we did. What does your company
360. 1:03:43understand that is genuinely hard to understand? And is it that understanding? Is that understanding getting deeper every day? If the answer
361. 1:03:51is nothing, AI is just a cost optimization story. You cut your headcount, improve the margins for a few
362. 1:03:57quarters, and eventually get absorbed by something smarter. If the answer is deep, AI doesn't augment your company.
363. 1:04:04It reveals what your company actually is. Block's answer is the economic graph. Millions of merchants and customers, both sides of every
364. 1:04:12transaction, financial financial behavior observed in real time. That understanding compounds every second of the system, every second the
365. 1:04:21system operates. We believe the pattern behind this, a company organized as an intelligence rather than a hierarchy,
366. 1:04:28is significant enough it will reshape how companies of all kinds operate over the coming years. Block is far enough
367. 1:04:35along to show the ideas more than theory. Though we welcome debate and feedback, we believe the pattern behind this, a company
368. 1:04:47organized as intelligence rather than a hierarchy, is significant enough that it will reshape how companies of all kinds
369. 1:04:55operate over the coming years. That's all we need to read. There's the punch line. Oh, that was it. Companies
370. 1:05:04move faster or slow based on information flow. Hierarchy and middle management impede information flow. Yes. Go watch office space.
371. 1:05:17Yeah. Yeah. Peter. Yeah. Hey, listen. We're going to need you to go on ahead and come on in on Saturday. Yeah. Well,
372. 1:05:24did you did you get the memo about the TPS reports? Right. Hierarchy and middle management impede information flow for 2,000 years from
373. 1:05:35the Roman Ku Bernium to today's global enterprises. We have no real alternative. Eight soldiers sharing a tent needed decanus. 80 men need a
374. 1:05:46centurion. 5,000 needed a legate. The question was never whether you needed layers. The question was whether humans
375. 1:05:53were the only option for what those layers do. They aren't anymore. Block is building what comes next. Called this
376. 1:06:00two weeks ago. Made a chart about it. Here you go. He's talking about economy three. He's building economy five.
377. 1:06:11People in economy four are going to be pulled toward economy five, but it's going to be hard. So, they might resist
378. 1:06:17it. And then what's going to happen is economies five and six are going to eat their lunch. My stapler. That's my stap. Yeah. My my
379. 1:06:25stapler. My st That's my stapler. I if I I've done the counting, there are not enough pieces of cake. Get that was
380. 1:06:36the That was the last piece piece of cake. That was the last piece of cake. [ __ ] crazy people.
381. 1:06:50From hierarchy to intelligence. quote I'm going to say this is very important
382. 1:07:08if you are in business read this Jack is building an
383. 1:07:22economy company. In fact, his
384. 1:07:37post about laying off 4,000 employees inspired my thinking on
385. 1:07:53the seven economies. Most companies right now are economies two, three or four.
386. 1:08:19you are likely in an economy 3 company which Jack talks about here.
387. 1:08:39Do not sleep on this. And then I'm gonna add my little picture. I'm gonna add my picture of the seven
388. 1:08:50economies. And we're going to send this out in the world. I'll tell you what. I'll tell you something. There, Sarah. Hey, Sarah. Saturated
389. 1:09:03name contains saturated. There we go. Saturated. There you go. Fantastic.
390. 1:09:52post. All right. All right. All righty. That's fantastic. Yeah, that's pretty good. That's pretty hot. All right. Missed the last five
391. 1:10:07minutes. Is someone making shakuderie board of the seven economies map? I hope so. That'd be cool, wouldn't it? Um,
392. 1:10:19that's kind of wild. All right, let me let me come back to Earth here for a second and check in with you all. I just
393. 1:10:24put DRRi on my resume. There you go. Um, that's a trippy article. In fact, I got to send that to some people.
394. 1:10:41Tik Tok question. What What economy does academia and science fit into? So, Cam, the the the way I think about it is this. the the
395. 1:10:54layers like education, business, um what did you say? Uh science, um politics, those are um horizontal layers and the
396. 1:11:07seven economies are vertical slices through them. So think of the each economy as level of AI adoption. Um,
397. 1:11:20so you're going to have educational institutions that some are probably still in economy one. They're just using books and
398. 1:11:29whiteboards or, you know, blackboards and chalk and [ __ ] like that. And they're not really using technology.
399. 1:11:34You've got some in economy 2 that are using, you know, some technology, but they're just, you know, they're still using fax machines.
400. 1:11:42Then you're going to have economy three educational institutions that are using AI to like improve email
401. 1:11:50communications. Then you're going to have economy 4, which is the transition economy. They'll be doing a little bit
402. 1:11:55of innovation. And then you'll have educational institutions doing five, six, and seven. And same with science. I
403. 1:12:00have a feeling science is just going to sort of lean quickly into five, six, and seven. But a lot of scientific
404. 1:12:07institutions are slow and traditional. So, so again, what's going to happen is what the the reason I'm calling these
405. 1:12:15seven economies, you've got AI accelerating up and to the up and to the right quickly. And then you've got 99% of people on
406. 1:12:27Earth that don't even really know what AI is. They maybe use Chat GPT, the free version of it, like like you know, 16%
407. 1:12:34of them use chat GPT like a Google search. 83% of them have never used AI. And AI is accelerating away from us. So as it
408. 1:12:44accelerates away from us, it's going to split. It's going to just fracture. All these different behaviors, all these
409. 1:12:50different levels of AI adoption are going to start functioning like separate economies. And so the the stack of education,
410. 1:12:58science, government, business, private sector, you know, uh you know, citizenship, citizenry are all going to they're going to be all
411. 1:13:11in this layer in an economy is I how I think it goes. Kim Kek, my college has fully embraced AI. That's amazing. That's great.
412. 1:13:27It's a seven economy cake. That's really cool. Um, let me copy this.
413. 1:14:30All right, I sent that off to some people. Um, post on Oh, post on LinkedIn. I'll share with my railroad account. Okay, I'll do
414. 1:14:44that. Let me go post this on LinkedIn. Actually, it's a good idea. Um, I'll I'll share my screen. Yes, AI readiness project podcast is
415. 1:14:54tomorrow um at 400 PM Mountain time. Oh, I'm sharing. Okay.
416. 1:15:30Okay. So, we're going to go next. We're going to go um this tweet from Jack.
417. 1:15:46Who's that Jack? Let's see. Um X is very important.
418. 1:16:07He's declaring what it means to be an economy five company. referring to most companies who
419. 1:16:25are now economy 2 or three. The TLDDR
420. 1:16:40is that for 2,000 years Large organizations
421. 1:16:54required slow hierarchies to manage the information flow
422. 1:17:12between um working units.
423. 1:17:32He is actively replacing the hierarchy at block. I wonder if block is block
424. 1:17:49a company block labs block company financial services he's actively replacing the hierarchy of block with AI with with intelligence
425. 1:18:09if you are in business. This should be required reading and a flipping wake up call.
426. 1:18:26If you are an employee, figure out which of the three roles
427. 1:18:43individual individual contributor what did he say?
428. 1:19:04Directly responsible individuals. Directly responsible individuals
429. 1:19:22or player, coaches. You think you are qualified
430. 1:19:36for Let's see. Paul Rer.
431. 1:19:51Let's see. Michael, Stellner,
432. 1:20:06Retesh, Patel, Ben Jones, let's see, Andy Scarantino, Beth Commtock.
433. 1:20:21Let's see. Chris Hoyer, what the Thomas Eric Makier IC's Dr.
434. 1:20:37These are tech company terms. Yeah, they're they're tech company terms, but like abstract them up. A player coach is someone who can do but
435. 1:20:48can develop people. A DRRi is basically think of it like an account manager with with deploying and budget capability,
436. 1:20:58right? So, think about someone who can say, "Hey, this customer needs this. I'm going to I'm going to tell the
437. 1:21:04intelligence, here's what I need, and the intelligence is going to figure out how to get me that." And then individual
438. 1:21:10contributors will make sure all the shit's working, right? Uh uh E A E Wait, ABCDE E F. Does anyone else do that when you're
439. 1:21:24trying to figure out where you are in the alphabet? You have to start at A. A B C D F G I say T U V V V V V V V V V V V V V V V V V V
440. 1:21:30V V V V V V V V V V V V V V V V V V V V V V V. Am I the only one? I'm meeting Ben on Thursday.
441. 1:21:45Waiting for Kazio. Let's see at G. Did my sharing just go away? It did, didn't it? God damn it.
442. 1:22:14H I J John Patel J Jason
443. 1:22:30K Kelly Anderson. We'll put Kelly. Chef Kelly's on the list. Um, H I J K L Log Logan Kilpatrick. We'll give him one.
444. 1:22:44M. We'll do Marique. N. Oops. All right. I'm bored now.
445. 1:23:00That's good. [ __ ] it. That's enough. We got our thing. All right. There you go. So, if you go to my LinkedIn account right now, it
446. 1:23:12says, "This tweet from Jack on X is really important." Okay, heading out folks. See y'all later in the AI salon. Mary Carol, pace out.
447. 1:23:28Yes. ABC D, etc. Beautiful. Beautiful, beautiful people. Fantastic. All right, it's getting to be that time. It's getting a little late
448. 1:23:38there. Little late there in the evening. Don't even mess with LinkedIn anymore. It's messy. Yeah, I know. I'm I'm
449. 1:23:45actually switching over to Substack. So, do me a favor. Um, go to Substack and my Substack is I am Kyle Shannon.
450. 1:23:56Let me make sure that's right. Yep. I am Kyle Shannon. All right.
451. 1:24:11Go to go to Substack and I am Kyle Shannon. All right. All right. Right. Kyle, check out the irregulars channel. I will do that right now.
452. 1:24:27Swing dig gank swing ginking.
453. 1:24:44Oh, that's so good. Seven economies shuderie board. Oh, that's so good. Oh my god,
454. 1:24:58that's so good. We got to extend the top of it. Or is the top of it there? Oh, it is. It's got a little jankiness in the uh in
455. 1:25:10the in the words. But damn, that's really cool, actually. So good. It's so good. Stability over change. Young cheddar improved margins.
456. 1:25:23Garganola blue chip blueberry compost. That should be economy one. That's really cool. That was really
457. 1:25:33good. Really, really good. That graphic makes me hungry. I know, right? It's really good. All right.
458. 1:25:48Oh, Lord. Okay, people. I'm going to get out of here. Um, so tomorrow, AI readiness project, that's at 4:00. Um, go to the
459. 1:26:00AI salon. If you haven't been there in a while, go there. I want to start getting us back into conversation in between
460. 1:26:08these meetings. I want to start getting things hopping again. Shit's about to get weird. Shit's about to go down.
461. 1:26:18We don't want to be doing this alone. I know it's [ __ ] exhausting. I get it. If anyone gets it, I get it.
462. 1:26:29Will you be on this Friday holiday? Um, I don't know what the holiday is. What? I don't even know what month we're in.
463. 1:26:43It's Good Friday. I will be doing something good Friday. It's also first Friday, so I might be late. So, I'll do something a little in
464. 1:26:52informal on Friday. We might even do an experiment. So, one of the things Brandon and Andy and I worked together
465. 1:26:59today in our L10 meeting to to figure out like if we're going to do AI learning lab inside the community,
466. 1:27:07what's that look like? What's the what's the most elegant way to do that? And so, over the next month, we're probably
467. 1:27:13going to be experimenting. So maybe we'll maybe we'll take Friday and experiment. Um so Friday might be might be fun
468. 1:27:22depending on your definition of fun. Office hours this Friday. Yes. This Friday is a really important one. That
469. 1:27:29Jack Dorsey thing that I just read, I'm not going to read the whole thing on Friday, but we're going to talk about that on Friday.
470. 1:27:35That's a That's a really important It's a It's a really important um thing that J Jack just put out there.
471. 1:27:44It's the first thing that I've seen where someone has actually declared what economy five is going to look like.
472. 1:27:51You've got a going concern. You're about to reinvent it. There's your road map right there. Incredibly well thought out. Like
473. 1:28:02really important. really important. I'm exhausted with trying to figure out clott GPT Pro like beer commercials.
474. 1:28:12Yeah, tastes great, less filling. Um, I don't know what to say. Like the the AI right now, it all still sucks. It it's not going to within
475. 1:28:25six months. We have a different problem when the AI doesn't suck, right? when it tastes great and it's less filling and it's cheap
476. 1:28:38and it'll get you drunk like Bllandons and like absin all at the same time. It's going to be a different problem.
477. 1:28:51We got to get We got to get reconnected in community. We got to get reconnected in community. Okay. Do you think Opus 4.6 sucks too? I
478. 1:29:00don't think so. I think Opus 4.6 is good. It's funny today. Oh, this is this is a wild [ __ ] insane [ __ ] thing. So, so there's this audio
479. 1:29:14synthesis thing called Cartisia and it's really good and they've got this data policy that if we want to use it
480. 1:29:23for our pharma clients, we have to turn on data protection. And if we turn on data protection, we can't log into their
481. 1:29:30playground. We we can log into their playground and we can make an audio file from text we put in, but you can't download it.
482. 1:29:42There's they they removed the download button and they're like, "Well, that goes against you. You you turned off
483. 1:29:47security privacy, so you can't download it on the the account you're logged into." So, I actually have to build an
484. 1:29:55application that uses their API to be able to pull my my [ __ ] file down. So, I went into Claude Code today and I
485. 1:30:01said, you know, here's the [ __ ] they gave me. Go build me a front end where I can input text and select from one of
486. 1:30:09the stupid [ __ ] voices in Cartisia and get my [ __ ] file. So, it it spent like half an hour coding. I don't know
487. 1:30:17if it finished it or not. I I left the office before I got to it. Um, I disagree that AI isn't great yet. It's mathematical. Did you mean
488. 1:30:26mathematical? Um, it is mathematical. You tell him, Kyle. Yeah. You know. All right, I gotta go. My eyes are blurry. I can't read like
489. 1:30:39that, you know. You know, man. Um, come check out the AI readiness project tomorrow. Go to the AI salon. Go post in
490. 1:30:48the community feed. Go post in the community chat. Go post in look what I made. Go post in the regulars. Join the
491. 1:30:54[ __ ] mastermind. We just rolled out today. We had um a workshop that I led on um figuring out what is your oneofone
492. 1:31:05thing that you're best in the world at and like really owning that, discovering it and owning it. It was a really cool
493. 1:31:12workshop today. We brought in people from Andy's Humans 2.0 um group of coaches and we mixed them with the with
494. 1:31:20the AI salon people. We crossed the streams. It was really cool. It was a good event. It It's only happening in the
495. 1:31:28mastermind. All the the um the great repurpose stuff is is going to be really [ __ ] cool. I would get your ass in
496. 1:31:35the in the mastermind. We're experimenting with it over the next six months. Um it's going to be good stuff.
497. 1:31:41It's gonna be really good stuff. All right, cool. Peace out. I think y'all are amazing. Appreciate you hanging out.
498. 1:31:48You have a good night. I'm late. Gareth, that's it. I'm leaving. Bye.