WEBVTT 00:00:00.000 --> 00:00:01.760 You know, I think for most people right now, 00:00:01.940 --> 00:00:05.960 the focus is all on using AI tools. They're finding 00:00:05.960 --> 00:00:10.199 clever ways to write a bit faster, maybe automate 00:00:10.199 --> 00:00:12.599 a single task. Right. A script here, a customer 00:00:12.599 --> 00:00:15.419 service reply there. Exactly. But there's this 00:00:15.419 --> 00:00:18.699 new class of builders, the sources are calling 00:00:18.699 --> 00:00:21.579 them AI CEOs, who are doing something completely 00:00:21.579 --> 00:00:23.980 different. They're building compounding systems. 00:00:24.239 --> 00:00:26.579 And that distinction, that's everything, isn't 00:00:26.579 --> 00:00:30.300 it? It is, because it unlocks this just... startling 00:00:30.300 --> 00:00:33.600 speed. We're seeing companies go from zero to 00:00:33.600 --> 00:00:38.179 $100 million in revenue faster than ever. I mean, 00:00:38.200 --> 00:00:40.000 ever. It's because they're building these self 00:00:40.000 --> 00:00:42.359 -improving loops instead of, you know, rigid 00:00:42.359 --> 00:00:44.979 business plans. That's the core of it. So in 00:00:44.979 --> 00:00:46.840 this deep dive, that's what we're unpacking. 00:00:46.840 --> 00:00:49.140 We're looking at the strategic thinking behind 00:00:49.140 --> 00:00:52.859 this kind of nonlinear success. We're going to 00:00:52.859 --> 00:00:54.899 get into the four specific feedback loops that 00:00:54.899 --> 00:00:57.960 create what the research calls real lasting wealth 00:00:57.960 --> 00:01:00.600 in the economy we're in right now. Welcome to 00:01:00.600 --> 00:01:03.259 the deep dive. Our mission today is pretty clear. 00:01:03.439 --> 00:01:06.540 We want to distill the core philosophy of what 00:01:06.540 --> 00:01:09.750 makes a successful AI CEO. based on the source 00:01:09.750 --> 00:01:11.930 material you sent over yeah so if you're looking 00:01:11.930 --> 00:01:14.290 to really understand what's driving this massive 00:01:14.290 --> 00:01:16.930 non -linear growth today you're in the right 00:01:16.930 --> 00:01:19.370 place and we kind of have to start by admitting 00:01:19.370 --> 00:01:22.390 something about our own brains the new economy 00:01:22.390 --> 00:01:26.269 was called the 2026 economy it's built on two 00:01:26.269 --> 00:01:29.450 things that we humans just aren't wired for and 00:01:29.450 --> 00:01:31.430 what are those things that grow incredibly fast 00:01:32.429 --> 00:01:34.909 and systems that improve themselves automatically. 00:01:35.269 --> 00:01:37.709 We think in straight lines. We do. We're wired 00:01:37.709 --> 00:01:40.129 for that gradual, predictable path. We plan for 00:01:40.129 --> 00:01:42.629 it. But that whole model is basically obsolete 00:01:42.629 --> 00:01:44.909 now. The wealth is in the loops, not the lines. 00:01:45.189 --> 00:01:47.069 That's it. And we're going to explore these four 00:01:47.069 --> 00:01:49.590 loops. There's the balance loop, the speed to 00:01:49.590 --> 00:01:51.769 revenue loop, the signal to innovation loop, 00:01:51.950 --> 00:01:54.670 and then maybe the most human one, the sweat 00:01:54.670 --> 00:01:56.599 equity loop. Understanding these, it's really 00:01:56.599 --> 00:01:58.659 a shift in thinking. You stop planning a product 00:01:58.659 --> 00:02:00.879 and you start building a kind of perpetual motion 00:02:00.879 --> 00:02:03.280 machine. So let's just start with the context, 00:02:03.319 --> 00:02:06.299 the sheer velocity we're seeing. Okay. The old 00:02:06.299 --> 00:02:08.439 growth models just don't fit anymore. I mean, 00:02:08.460 --> 00:02:11.979 look at the data. ChatGPT hit 100 million users 00:02:11.979 --> 00:02:16.139 in 60 days. 60 days. That number alone, it almost 00:02:16.139 --> 00:02:18.819 bends reality. That benchmark didn't even exist 00:02:18.819 --> 00:02:21.020 a few years ago. And the speed itself is getting 00:02:21.020 --> 00:02:24.340 faster. So Sora, for instance. got to a million 00:02:24.340 --> 00:02:26.800 downloads even quicker than ChatGPT did. Wow. 00:02:27.300 --> 00:02:30.020 And in the business world, this translates right 00:02:30.020 --> 00:02:33.800 to revenue. We're seeing startups like Cursor 00:02:33.800 --> 00:02:36.500 and Perplexity hitting $100 million in recurring 00:02:36.500 --> 00:02:39.599 revenue in, what, 12 to 18 months? Whoa. Just 00:02:39.599 --> 00:02:42.219 imagine trying to scale to a billion queries 00:02:42.219 --> 00:02:45.939 or $100 million in less than two years. It's 00:02:45.939 --> 00:02:47.680 genuinely hard to wrap your head around that 00:02:47.680 --> 00:02:50.000 kind of growth. It is. And it makes that old 00:02:50.000 --> 00:02:53.449 school five -year plan just... Completely obsolete. 00:02:53.490 --> 00:02:55.789 You move slow, you're just done. Absolutely. 00:02:55.969 --> 00:02:58.569 The sources are so clear on this. The speed isn't 00:02:58.569 --> 00:03:00.409 coming from some perfect launch plan. Because 00:03:00.409 --> 00:03:02.889 that plan is old by the time you launch it. Right. 00:03:03.009 --> 00:03:06.129 It comes from building these loops that learn 00:03:06.129 --> 00:03:08.610 and compound almost instantly. The loops are 00:03:08.610 --> 00:03:11.430 the plan. So that immediately brings up the question 00:03:11.430 --> 00:03:14.710 then, if the plan is obsolete, where on earth 00:03:14.710 --> 00:03:17.090 do you even start? You start with loop number 00:03:17.090 --> 00:03:20.830 one, the balance loop. And the idea here is simple. 00:03:21.120 --> 00:03:24.139 Stop chasing trends. Stop chasing the flavor 00:03:24.139 --> 00:03:27.180 of the month. Yeah. Instead, chase acute pain. 00:03:27.439 --> 00:03:30.159 So that's the balance. It's where your unique 00:03:30.159 --> 00:03:33.780 sort of unfair edge meets a real problem that 00:03:33.780 --> 00:03:35.780 customers are having every single week. Exactly. 00:03:35.900 --> 00:03:37.659 You focus only on your skills. You build something 00:03:37.659 --> 00:03:40.219 nobody buys. You focus only on the pain. You're 00:03:40.219 --> 00:03:42.539 just a commodity. Easy to copy. So let's define 00:03:42.539 --> 00:03:45.539 that edge. What is an asymmetric advantage? It's 00:03:45.539 --> 00:03:47.919 not just being talented, right? No. It's judgment 00:03:47.919 --> 00:03:50.439 that's been shaped by time. It's deep industry 00:03:50.439 --> 00:03:52.900 knowledge. Or maybe you mastered some really 00:03:52.900 --> 00:03:55.199 boring, messy process that no one else wanted 00:03:55.199 --> 00:03:58.020 to touch for years. That boring experience is 00:03:58.020 --> 00:04:00.219 actually the valuable part. It's the most valuable 00:04:00.219 --> 00:04:02.960 part because AI can now amplify it like never 00:04:02.960 --> 00:04:05.990 before. Right. And we see that in the GIP Investment 00:04:05.990 --> 00:04:10.009 Bank case study. They had 20 years of due diligence 00:04:10.009 --> 00:04:12.770 judgment. That was their edge. And when they 00:04:12.770 --> 00:04:15.629 used AI to automate, you know, 70 percent of 00:04:15.629 --> 00:04:18.009 their manual work, they did more than save time. 00:04:18.269 --> 00:04:20.529 They could suddenly do 40 percent more deals 00:04:20.529 --> 00:04:23.750 at a 60 percent lower cost. Their human learning 00:04:23.750 --> 00:04:26.209 became the AI's accelerator. OK, so that's the 00:04:26.209 --> 00:04:28.310 edge. What about the other side, the acute pain? 00:04:28.629 --> 00:04:31.069 Well, businesses don't pay for your vision. They 00:04:31.069 --> 00:04:33.379 pay to stop bleeding time and money. Like right 00:04:33.379 --> 00:04:35.199 now. So that has to be a weekly frustration. 00:04:35.259 --> 00:04:38.680 A weekly or even daily frustration. An accountant 00:04:38.680 --> 00:04:41.040 wasting five hours every week on reconciliation. 00:04:41.600 --> 00:04:45.019 A realtor losing deals, chasing down paper signatures. 00:04:45.160 --> 00:04:48.199 Or new problems. Not at all. But they're consistently 00:04:48.199 --> 00:04:51.339 painful. And that pain creates urgency, which 00:04:51.339 --> 00:04:53.899 makes people willing to pay. So the formula is 00:04:53.899 --> 00:04:57.699 simple. Your edge plus real pain plus fast testing. 00:04:57.759 --> 00:05:00.160 That's a business worth building. So if that 00:05:00.160 --> 00:05:03.060 unique expertise is key, what's the real value 00:05:03.060 --> 00:05:06.600 of focusing on that messy, kind of unexciting 00:05:06.600 --> 00:05:08.939 pain? It just guarantees you're building something 00:05:08.939 --> 00:05:11.300 people are actually desperate to solve. Okay, 00:05:11.339 --> 00:05:13.480 that brings us to the second loop. So if loop 00:05:13.480 --> 00:05:16.819 one is about what to solve, loop two, the speed 00:05:16.819 --> 00:05:19.800 to revenue loop, is all about the how fast. And 00:05:19.800 --> 00:05:23.720 here, speed beats perfection. Every time. And 00:05:23.720 --> 00:05:26.699 the case study for this is Cursor, right? Zero 00:05:26.699 --> 00:05:29.519 to 100 million in 18 months. By shipping new 00:05:29.519 --> 00:05:32.339 features every single day. Every day. I mean, 00:05:32.339 --> 00:05:34.199 how is that even possible? What's the engine 00:05:34.199 --> 00:05:36.560 behind that speed? They call it the hidden engine. 00:05:36.699 --> 00:05:40.139 It's a meta loop. The team at Cursor uses Cursor 00:05:40.139 --> 00:05:43.199 to build Cursor. Ah, so they're their own first 00:05:43.199 --> 00:05:45.699 customer. Instantly. They ship a feature, use 00:05:45.699 --> 00:05:48.019 it themselves, find a bug and fix it, often on 00:05:48.019 --> 00:05:50.620 the same day. So they completely cut out those 00:05:50.620 --> 00:05:53.899 long feedback cycles. No endless meetings, no 00:05:53.899 --> 00:05:56.240 waiting weeks for user feedback. The pain is 00:05:56.240 --> 00:05:58.500 immediate because they live inside the product. 00:05:58.699 --> 00:06:00.860 That's their competitive mode. And that's so 00:06:00.860 --> 00:06:03.199 important for anyone listening because long -term 00:06:03.199 --> 00:06:06.220 planning is just so fragile now. It is. A new 00:06:06.220 --> 00:06:09.360 model release, like a hypothetical ChatGPT -6 00:06:09.360 --> 00:06:12.000 or Gemini 4, could just wipe out your advantage 00:06:12.000 --> 00:06:13.720 in a few weeks. You can't just disappear for 00:06:13.720 --> 00:06:15.980 six months to build in secret anymore. But I 00:06:15.980 --> 00:06:17.860 mean, doesn't that just lead to burnout? How 00:06:17.860 --> 00:06:20.699 do you maintain an edge when the ground is constantly 00:06:20.699 --> 00:06:23.100 shifting underneath you? Your advantage is the 00:06:23.100 --> 00:06:26.100 speed of your learning. By iterating daily, you 00:06:26.100 --> 00:06:28.819 prioritize learning and revenue. And that becomes 00:06:28.819 --> 00:06:32.060 your moat. not the code itself welcome back to 00:06:32.060 --> 00:06:34.060 the deep dive we just covered the need to move 00:06:34.060 --> 00:06:36.279 incredibly fast with the speed to revenue loop 00:06:36.279 --> 00:06:39.240 now let's talk about loop number three the signal 00:06:39.240 --> 00:06:41.620 to innovation loop this one is about survival 00:06:41.620 --> 00:06:44.240 really when any feature can be copied in a matter 00:06:44.240 --> 00:06:47.120 of weeks your only protection is listening to 00:06:47.120 --> 00:06:49.379 reality listening to what your users actually 00:06:49.379 --> 00:06:52.540 do exactly and the best contrast here is youtube 00:06:52.540 --> 00:06:56.199 versus uh the startup quibi youtube is basically 00:06:56.199 --> 00:06:58.920 a giant learning machine Every click, every pause, 00:06:59.040 --> 00:07:01.720 every scroll is a signal it uses to adjust. It's 00:07:01.720 --> 00:07:04.160 a real -time system. It's not using focus groups 00:07:04.160 --> 00:07:06.300 to guess what people want. It's just watching 00:07:06.300 --> 00:07:08.959 behavior and constantly tweaking itself. That 00:07:08.959 --> 00:07:12.220 loop never stops. Never. Now, compare that to 00:07:12.220 --> 00:07:15.459 Quibi. They launched with almost $2 billion. 00:07:16.480 --> 00:07:20.019 Huge celebrities, a very confident vision. Which 00:07:20.019 --> 00:07:22.500 was short, premium shows for your phone. Simple 00:07:22.500 --> 00:07:25.519 idea. But the engagement data, right from the 00:07:25.519 --> 00:07:27.959 start, was weak. And they trusted their vision 00:07:27.959 --> 00:07:30.699 more than the signals? They did. Instead of adjusting 00:07:30.699 --> 00:07:33.379 to the reality in front of them, they just defended 00:07:33.379 --> 00:07:35.920 the original idea. And the company was gone in 00:07:35.920 --> 00:07:38.439 six months. So the money didn't save them. The 00:07:38.439 --> 00:07:41.240 big names didn't save them. No. The insight is 00:07:41.240 --> 00:07:44.379 that features and price won't save you. The only 00:07:44.379 --> 00:07:46.779 real edge is how fast you notice and respond 00:07:46.779 --> 00:07:49.699 to those signals. So why is responding to user 00:07:49.699 --> 00:07:52.579 signals more important now than it was, say, 00:07:53.149 --> 00:07:55.790 five years ago because features get copied so 00:07:55.790 --> 00:07:58.569 fast that tight signal loop is the only thing 00:07:58.569 --> 00:08:00.269 that lets you pivot and protect the business 00:08:00.269 --> 00:08:03.209 from decay that brings us to the final loop which 00:08:03.209 --> 00:08:05.389 might be the most important one the sweat equity 00:08:05.389 --> 00:08:07.569 loop yeah this one is very human the old advice 00:08:07.569 --> 00:08:09.810 was always hire smart people and get out of the 00:08:09.810 --> 00:08:12.689 way but in an ai first world that can actually 00:08:12.689 --> 00:08:16.610 destroy your business it can because ai products 00:08:16.610 --> 00:08:20.430 aren't set it and forget it they drift they slowly 00:08:20.430 --> 00:08:22.980 get worse if a human isn't paying close attention. 00:08:23.240 --> 00:08:25.680 You see things like prompt degradation, messy 00:08:25.680 --> 00:08:29.000 data, edge cases that just explode. And for anyone 00:08:29.000 --> 00:08:31.199 who hasn't seen prompt degradation, it's basically 00:08:31.199 --> 00:08:34.799 when the system starts to forget its instructions. 00:08:34.919 --> 00:08:38.500 The output gets more generic, less useful. By 00:08:38.500 --> 00:08:40.360 the time it's obvious, your users are probably 00:08:40.360 --> 00:08:43.759 already gone. You've lost them. You know, I still 00:08:43.759 --> 00:08:46.120 wrestle with pump drift myself, even on small 00:08:46.120 --> 00:08:49.470 personal projects. It takes that... that invisible 00:08:49.470 --> 00:08:52.509 manual effort, that sweat equity, to stay close 00:08:52.509 --> 00:08:54.629 and test the outputs and just notice when something 00:08:54.629 --> 00:08:56.850 feels a little off. And that invisible effort, 00:08:56.990 --> 00:08:58.710 those late nights running tests, that turns into 00:08:58.710 --> 00:09:01.350 insight. It sharpens your judgment. And that 00:09:01.350 --> 00:09:03.529 judgment is the one thing AI cannot take from 00:09:03.529 --> 00:09:05.850 you. So failure is just part of the loop. It 00:09:05.850 --> 00:09:08.389 turns effort into experience. Exactly. Money 00:09:08.389 --> 00:09:11.090 without judgment has gone fast. But judgment 00:09:11.090 --> 00:09:14.049 earned through that effort, that sticks. So how 00:09:14.049 --> 00:09:16.009 do founders balance staying that close to the 00:09:16.009 --> 00:09:18.879 product with the need to... actually scale the 00:09:18.879 --> 00:09:21.740 business. Well, the visible scaling, it actually 00:09:21.740 --> 00:09:24.919 relies on that invisible sweat equity to keep 00:09:24.919 --> 00:09:27.440 all the automated systems sharp and alive. We've 00:09:27.440 --> 00:09:29.860 covered a lot here, the four loops, balance, 00:09:30.080 --> 00:09:33.480 speed, signal, and sweat equity. But I think 00:09:33.480 --> 00:09:35.799 the biggest payoff that's often missed isn't 00:09:35.799 --> 00:09:38.740 just the revenue. It's the understanding you 00:09:38.740 --> 00:09:40.379 build along the way. That's the part that no 00:09:40.379 --> 00:09:42.440 one else can copy. Right. And to make this really 00:09:42.440 --> 00:09:45.039 actionable for you, the listener, we pulled four 00:09:45.039 --> 00:09:47.809 steps directly from the material. Okay. First, 00:09:47.870 --> 00:09:50.669 for balance. Just write down what you are uniquely 00:09:50.669 --> 00:09:55.590 good at and then list three painful weekly problems 00:09:55.590 --> 00:09:57.549 that you hear people complain about. Simple enough. 00:09:57.730 --> 00:10:01.190 Second, for speed. Just ship something small 00:10:01.190 --> 00:10:02.990 every single week for the next month. It doesn't 00:10:02.990 --> 00:10:05.509 have to be perfect. Just keep moving. Third, 00:10:05.610 --> 00:10:09.169 signal. Track one real action user's take in 00:10:09.169 --> 00:10:11.590 your system. Not a vanity metric like likes, 00:10:11.629 --> 00:10:14.769 but actual behavior. Yeah. And finally, sweat. 00:10:15.360 --> 00:10:17.659 Find that one messy, degrading problem you've 00:10:17.659 --> 00:10:19.940 been avoiding and just spend some real time fixing 00:10:19.940 --> 00:10:22.250 it this week. The goal is to build systems that 00:10:22.250 --> 00:10:24.470 learn faster than the world is changing. That's 00:10:24.470 --> 00:10:26.730 the whole game. So here's the final thought for 00:10:26.730 --> 00:10:29.169 you to think about as we wrap up. If that's the 00:10:29.169 --> 00:10:32.350 ultimate goal, what's the one crucial human skill 00:10:32.350 --> 00:10:35.809 beyond just technical ability that you personally 00:10:35.809 --> 00:10:39.009 might be neglecting? The one that, if you sharpened 00:10:39.009 --> 00:10:41.649 it with real effort, could become the most valuable, 00:10:41.830 --> 00:10:44.309 non -automatable part of your own personal system. 00:10:44.549 --> 00:10:46.629 That wisdom that you build through effort, that's 00:10:46.629 --> 00:10:48.980 the real prize in this economy. Keep building, 00:10:49.139 --> 00:10:51.440 keep learning, and you'll stay ahead. Thank you 00:10:51.440 --> 00:10:53.240 for joining us for this deep dive. We'll talk 00:10:53.240 --> 00:10:53.779 to you next time.