1 00:00:00,040 --> 00:00:02,160 Jeremy, welcome to New York City podcast. 2 00:00:02,160 --> 00:00:05,320 And we've been we've been friends or friendly online, I 3 00:00:05,320 --> 00:00:08,760 think for a while. You are the the founder of fast 4 00:00:08,760 --> 00:00:11,480 dot AI, which is this incredible course that's online. 5 00:00:12,000 --> 00:00:13,600 We both taught large online courses. 6 00:00:13,600 --> 00:00:15,200 So we kind of have have talked about that. 7 00:00:15,400 --> 00:00:17,720 You're the founder of Answer dot AI before that, I think you were 8 00:00:17,720 --> 00:00:21,440 at Cagle, right? And you're Australian. 9 00:00:22,240 --> 00:00:24,240 You have an interest in biomedicine. 10 00:00:24,720 --> 00:00:27,880 And I think we're also into, I mean peace and trade broadly 11 00:00:27,880 --> 00:00:32,159 internationalism and and so on. Give me the spiel is that is 12 00:00:32,479 --> 00:00:33,960 Danielle. Everything is that or give me 13 00:00:33,960 --> 00:00:35,880 Jeremy on Jeremy. Yeah, no, pretty much. 14 00:00:35,880 --> 00:00:39,520 I mean, I say maybe fast AI, most people know us for the 15 00:00:39,520 --> 00:00:41,680 course because that's how most people interact with us, but 16 00:00:41,680 --> 00:00:46,480 that was only one quarter of it. So fast AI was all about trying 17 00:00:46,480 --> 00:00:50,320 to avoid a kind of massive centralization of power and 18 00:00:50,320 --> 00:00:54,120 inequality due to what my wife and I saw in 2012 is likely to 19 00:00:54,120 --> 00:00:57,280 be a rapid growth of AI. And so we want to say. 20 00:00:57,280 --> 00:01:00,600 Similar to open AI's mission. In theory, except we actually 21 00:01:00,600 --> 00:01:01,440 were open. Yeah. 22 00:01:01,440 --> 00:01:04,000 So we. Yeah. 23 00:01:04,000 --> 00:01:07,080 So we basically decided to get AI into the hands of as many 24 00:01:07,080 --> 00:01:11,240 people as possible, including people with few resources. 25 00:01:11,360 --> 00:01:15,880 And so we did a lot of research to figure out how to make AI 26 00:01:15,880 --> 00:01:19,080 more accessible because at that time, only 5 labs in the world. 27 00:01:19,120 --> 00:01:23,080 And yeah, the techniques to actually use AI in practice were 28 00:01:23,080 --> 00:01:25,880 not published. They were kind of like little 29 00:01:26,760 --> 00:01:30,280 recipes. Yeah, so my wife Rachel actually 30 00:01:30,280 --> 00:01:33,800 asked earlier when he was presenting in like 2012 or 31 00:01:33,800 --> 00:01:35,200 something about some of his work. 32 00:01:35,640 --> 00:01:38,760 And it's like, OK, so how did you actually do that bit? 33 00:01:38,760 --> 00:01:41,200 What weights did you use? How you know what fine shooting 34 00:01:41,200 --> 00:01:43,400 is to use? He's like, oh, we don't we don't 35 00:01:43,400 --> 00:01:45,960 publish any of that. That's our bag of tricks. 36 00:01:46,760 --> 00:01:48,880 So we were like, OK, this is not OK. 37 00:01:48,880 --> 00:01:53,000 Like this is this technology is going to change the world and 38 00:01:53,000 --> 00:01:55,520 it's requires a bag of tricks that you have to go to Stanford 39 00:01:55,520 --> 00:01:59,920 to learn, you know, So we figured out all the tricks and 40 00:01:59,920 --> 00:02:01,440 built a lot more tricks of our own. 41 00:02:02,120 --> 00:02:04,280 And then, you know, everybody then tried to make it all about 42 00:02:04,920 --> 00:02:06,440 money. So then Google eventually 43 00:02:06,440 --> 00:02:09,360 started creating TP us and stuff instead of saying like, oh, you 44 00:02:09,360 --> 00:02:11,280 can't. I remember Jeff Dean saying 45 00:02:11,960 --> 00:02:15,560 there's no point trying to do stuff with AI unless you're at 46 00:02:15,560 --> 00:02:18,080 Google because only we have this a compute. 47 00:02:18,080 --> 00:02:20,920 Yeah. And we beat them in a global 48 00:02:20,920 --> 00:02:22,520 competition to train Imagenet. You. 49 00:02:22,720 --> 00:02:25,720 Mean a taggle. No at that fast AI. 50 00:02:26,240 --> 00:02:27,160 Oh really? I didn't actually know. 51 00:02:27,200 --> 00:02:29,000 That yeah, yeah. There was a global competition 52 00:02:29,040 --> 00:02:32,120 called Donbench and we competed against Intel. 53 00:02:32,480 --> 00:02:37,880 They had like a cluster of 1000 Donbench DAWN Bench EE NCH. 54 00:02:37,920 --> 00:02:40,400 By the way, I love I, I, I'm friendly with Jeff Dean. 55 00:02:40,400 --> 00:02:41,400 I think he's amazing. And so. 56 00:02:41,520 --> 00:02:43,400 Yeah, yeah, yeah. So, so that, that's actually 57 00:02:43,400 --> 00:02:44,680 pretty. I mean, I'm sure he was 58 00:02:44,680 --> 00:02:46,000 impressed that you're able to do so much. 59 00:02:46,000 --> 00:02:47,680 Oh, yeah, no, he was, he was great about it. 60 00:02:47,680 --> 00:02:50,680 You know, they, they published a post, they published a paper and 61 00:02:50,680 --> 00:02:53,280 they credited us. And there's no hard feelings, 62 00:02:53,280 --> 00:02:56,280 you know, but we just want it. It's just we wanted to say like, 63 00:02:56,280 --> 00:03:00,280 no, you don't have to be a rich Google person to. 64 00:03:01,480 --> 00:03:02,480 You know how is that? Happening. 65 00:03:03,080 --> 00:03:06,520 Actually, maybe you can talk about that because like that's a 66 00:03:06,520 --> 00:03:10,200 little surprising to me because you know, obviously DeepSeek has 67 00:03:10,200 --> 00:03:14,520 brought costs down recently, but back then was it did you like 68 00:03:14,920 --> 00:03:18,680 obviously Google had massive amounts of clean data and huge 69 00:03:18,680 --> 00:03:23,280 compute resources and so on. Why could how could the student 70 00:03:23,280 --> 00:03:25,400 projects be competitive with Google during Dawn Bench? 71 00:03:25,960 --> 00:03:33,120 Because these big labs suffer from being over resourced. 72 00:03:33,520 --> 00:03:37,240 So in fact not as bad now, but particularly around that time 73 00:03:37,240 --> 00:03:39,800 and for the next few years at Google you're explicitly 74 00:03:39,800 --> 00:03:44,920 rewarded for using more compute. Where else we were like, hey, we 75 00:03:44,920 --> 00:03:48,040 don't have much money. Like we, we made no revenue, we 76 00:03:48,040 --> 00:03:50,160 had no grants. It was just my wife and I put 77 00:03:50,160 --> 00:03:52,960 our own money into fast AI can. You explain that to me. 78 00:03:52,960 --> 00:03:55,720 How are they rewarded for? Using so they were basically if 79 00:03:55,720 --> 00:03:59,960 you could use more TP us that's like a a a good tick on your 80 00:03:59,960 --> 00:04:01,200 performance. No, really. 81 00:04:01,200 --> 00:04:02,400 Yeah, wow. OK. 82 00:04:02,760 --> 00:04:07,040 So, you know, we came along and said, hey, like so for example. 83 00:04:08,440 --> 00:04:10,640 It's it's because they wanted people to use the TP us since 84 00:04:10,640 --> 00:04:11,720 they were. Yeah, and they wanted to like 85 00:04:11,720 --> 00:04:15,240 show off how big their, their rig was. 86 00:04:15,280 --> 00:04:17,839 And like you look at our big rig and these people using our big 87 00:04:17,839 --> 00:04:21,880 rig to do these big things. Well, for example, in in Dawn 88 00:04:21,880 --> 00:04:24,840 Bench, it was an image recognition competition, be as 89 00:04:24,840 --> 00:04:26,640 fast as you can to train a model. 90 00:04:27,920 --> 00:04:31,200 And the images were 224 by 224 pixels. 91 00:04:31,760 --> 00:04:35,800 And we thought like, OK, well, 90% of the time, the 1st 90% of 92 00:04:35,800 --> 00:04:40,240 training, we're going to train on 64 by 64 pixel downsized 93 00:04:40,240 --> 00:04:42,200 versions. Yeah, makes perfect sense. 94 00:04:42,200 --> 00:04:44,920 They look the same. You know the last 10% were used 95 00:04:44,920 --> 00:04:47,760 bigger ones. That 4X or 16X delta. 96 00:04:47,880 --> 00:04:50,320 Nobody else thought of that. You know, this is one of the 97 00:04:50,320 --> 00:04:54,160 many tricks we used and why would anybody like an open AI or 98 00:04:54,680 --> 00:04:58,160 Google try and do that? Because it's like, Oh, well, now 99 00:04:58,160 --> 00:05:00,280 we're not using our amazing DP us well. 100 00:05:00,720 --> 00:05:05,240 It's it's interesting because you know, that's actually I, 101 00:05:05,320 --> 00:05:08,200 I'm, I'm actually going to put out a little little comic on 102 00:05:08,200 --> 00:05:11,160 this actually on that, which is, you know, that meme about a 103 00:05:11,160 --> 00:05:16,160 secret third thing is, is people will say, Oh, you're not an X or 104 00:05:16,200 --> 00:05:18,560 AY, but a secret third thing. And they'll say it sarcastically 105 00:05:18,560 --> 00:05:20,120 like, Oh, you must be a Democrat or Republican. 106 00:05:20,120 --> 00:05:21,840 You're not a secret third thing, right. 107 00:05:22,240 --> 00:05:28,200 But actually, if you think about like, like a, like an image zero 108 00:05:28,200 --> 00:05:31,280 or one, one pixel is not enough to describe the complexity of an 109 00:05:31,280 --> 00:05:32,920 image. You need not just a secret third 110 00:05:32,920 --> 00:05:36,200 thing, but a secret 4th and 5th and thousandth and millionth and 111 00:05:36,200 --> 00:05:37,400 so on, right? Pixels. 112 00:05:37,720 --> 00:05:40,320 But you know, there is, there is a minimum necessary complexity, 113 00:05:40,560 --> 00:05:42,320 right? And it's interesting because 114 00:05:42,320 --> 00:05:45,240 obviously if you go all the way down to like a, you know, if you 115 00:05:45,240 --> 00:05:47,640 have the number of pixels all the way down to just one, you're 116 00:05:47,640 --> 00:05:50,080 not going to get enough, right? So it's an empirical question. 117 00:05:50,080 --> 00:05:53,560 Going from 2:56 to 64, it still works. 118 00:05:53,640 --> 00:05:55,280 I don't know, maybe going to 32 it still works. 119 00:05:55,280 --> 00:05:57,960 Maybe going to a fave icon, it even kind of still works. 120 00:05:57,960 --> 00:05:59,360 I don't know if you did that, if you. 121 00:05:59,360 --> 00:06:00,240 Absolutely. We did. 122 00:06:00,640 --> 00:06:05,440 And but I first just did it visually, you know, I just 123 00:06:05,440 --> 00:06:07,640 downscaled it and I looked and I was like, can I still see what 124 00:06:07,640 --> 00:06:08,320 that is? Right. 125 00:06:08,560 --> 00:06:10,960 And if I couldn't see it, then I thought computer probably won't 126 00:06:10,960 --> 00:06:12,240 be able to do as well. What was it? 127 00:06:12,240 --> 00:06:14,560 Was it like, was it 16? Was it 32? 128 00:06:14,560 --> 00:06:16,560 Was kind of. 6464. Yeah. 129 00:06:16,560 --> 00:06:21,120 OK, Yeah, at at 32 you. It's Squint. 130 00:06:21,320 --> 00:06:24,480 Yeah, it's it's, you can kind of see it's maybe a dog, but you 131 00:06:24,480 --> 00:06:25,760 can't see what kind of dog it is. 132 00:06:25,840 --> 00:06:27,120 I see. Interesting. 133 00:06:27,160 --> 00:06:29,240 Yeah. OK, so OK, I want to. 134 00:06:29,240 --> 00:06:31,200 Well, there's actually, I love, I love this. 135 00:06:31,360 --> 00:06:35,040 So first of all, I want to actually show you something. 136 00:06:35,200 --> 00:06:37,560 It'll jump around whatever. I want to show you something 137 00:06:37,560 --> 00:06:40,520 that we have done that I think is a compliment to Fast AI and 138 00:06:40,520 --> 00:06:43,640 this also. So I taught a MOOC in 2013 139 00:06:43,640 --> 00:06:45,800 called Startup Engineering. I'm a big fan of it, yeah. 140 00:06:45,800 --> 00:06:47,560 OK, great. So I did that with Vijay Pandey, 141 00:06:48,080 --> 00:06:50,240 my my colleague is. I'm a big fan of Vijay as well. 142 00:06:50,280 --> 00:06:51,760 Great. So he he's now at the bio fund. 143 00:06:51,760 --> 00:06:53,080 We've invested a lot of bio stuff together. 144 00:06:53,080 --> 00:06:54,280 So we have that overlap as well We're. 145 00:06:54,280 --> 00:06:57,680 Interested. So you and Steve Huffman created 146 00:06:57,680 --> 00:07:00,040 those two fantastic courses. I don't know if you ever looked 147 00:07:00,040 --> 00:07:00,800 I. Don't know Steve Huffman's 148 00:07:00,800 --> 00:07:01,160 course? What's? 149 00:07:01,160 --> 00:07:04,760 His yeah, so similar thing. They were both like kind of end 150 00:07:04,760 --> 00:07:08,240 to end, like how to make stuff. Oh, OK, got it. 151 00:07:08,240 --> 00:07:10,440 And. That's the Reddit founder, Yeah, 152 00:07:11,080 --> 00:07:12,520 Yeah, he's my friend also. I didn't actually. 153 00:07:12,520 --> 00:07:14,800 Know he did, of course. Yeah, so and neither of them are 154 00:07:14,800 --> 00:07:17,400 really available anymore and they're, you know, I free. 155 00:07:17,400 --> 00:07:18,880 Web Development course by Steve Huffman. 156 00:07:18,960 --> 00:07:20,680 Interesting. We need a, we need a, we need a 157 00:07:20,680 --> 00:07:23,560 modern one. Alright, OK, so how about this, 158 00:07:24,160 --> 00:07:26,840 maybe I'll do a refresher and we can we'll send it to the fast AI 159 00:07:26,840 --> 00:07:28,560 people put it online and something like that. 160 00:07:28,560 --> 00:07:31,000 I think that I do think a 2025 version. 161 00:07:31,000 --> 00:07:33,360 So actually, you know, let me tell you what I'm planning to do 162 00:07:33,360 --> 00:07:35,960 next on this. Well, so the reason I taught 163 00:07:35,960 --> 00:07:38,840 that course very similar I think in some ways to your, you know, 164 00:07:39,200 --> 00:07:44,880 kind of kind of thing is I know there's a lot of talent on the 165 00:07:44,880 --> 00:07:46,720 Internet, right? And actually really around the 166 00:07:46,720 --> 00:07:48,160 world. And you know how like the, you 167 00:07:48,480 --> 00:07:50,440 know, the kinds of the dark matter and like the Hubble 168 00:07:50,440 --> 00:07:53,960 telescope and you can find the dark matter around the globe or 169 00:07:54,080 --> 00:07:56,080 or or not the globe in the universe, right? 170 00:07:56,440 --> 00:07:58,200 So. Like gravitational lensing? 171 00:07:58,200 --> 00:07:59,360 Yeah, exactly. That's right. 172 00:07:59,400 --> 00:08:01,720 And so you need like a special telescope to see that, right. 173 00:08:01,960 --> 00:08:06,280 So by analogy, just a fun analogy, the mobile telescope, 174 00:08:06,280 --> 00:08:09,600 like the, the phones that billions of people now have, 175 00:08:10,000 --> 00:08:13,880 allow us to find if, if the, if the Hubble telescope allows us 176 00:08:13,880 --> 00:08:16,400 to find the dark matter, the mobile telescope, so to speak, 177 00:08:16,640 --> 00:08:19,520 allows us to find the dark talent around the world, right? 178 00:08:19,760 --> 00:08:22,320 Basically people who really have nothing other than their phone 179 00:08:22,720 --> 00:08:24,720 and their hunger to learn, right? 180 00:08:25,160 --> 00:08:28,200 And we can offer them a course, and that's like a Skyhook and a 181 00:08:28,200 --> 00:08:30,320 Bootstrap. That's what Fast AI was about as 182 00:08:30,320 --> 00:08:31,960 well. Like we really reached out to 183 00:08:32,679 --> 00:08:37,400 parts of India and Africa and stuff that had nothing. 184 00:08:37,400 --> 00:08:41,360 So we had like a guy from the Ivory Coast who was like asking 185 00:08:41,360 --> 00:08:45,120 like, is there some way to get this on CDs, 'cause we don't 186 00:08:45,120 --> 00:08:48,680 have Internet here. And yeah, turned out like one of 187 00:08:48,680 --> 00:08:50,360 our biggest markets was in Lagos. 188 00:08:50,960 --> 00:08:53,200 It's amazing. So actually I, I, I have a fair 189 00:08:53,200 --> 00:08:55,920 number of folks in, in Nigeria, basically anywhere there's 190 00:08:55,920 --> 00:08:59,200 Anglophones around the world, in India, Nigeria, in the 191 00:08:59,200 --> 00:09:01,600 Philippines, right? There's actually all these 192 00:09:01,600 --> 00:09:04,800 Anglophones meaning just, I, I do want to translate into other 193 00:09:04,800 --> 00:09:07,000 languages and so on. But I think that's like the V1. 194 00:09:07,040 --> 00:09:07,560 Right. Go ahead. 195 00:09:08,280 --> 00:09:12,080 Yeah, No, I mean it. And it, it, it was just like, 196 00:09:12,840 --> 00:09:16,000 it, it, there's all this talent around the world and it drives 197 00:09:16,000 --> 00:09:18,960 me crazy that it, it's not being used. 198 00:09:18,960 --> 00:09:21,320 You know, they're like picking coffee beans or whatever. 199 00:09:21,800 --> 00:09:25,680 And, and as you say, like they've got like so many of them 200 00:09:25,680 --> 00:09:28,800 were saying like how I'm training a particularly when 201 00:09:28,800 --> 00:09:30,560 collab, Google Paylab came along, they're like, I'm 202 00:09:30,560 --> 00:09:34,280 training a neural net on my phone, you know, through collab, 203 00:09:35,480 --> 00:09:37,200 you know, can you help me do this or that? 204 00:09:37,200 --> 00:09:39,480 And I'm just like, oh, this is great, you know? 205 00:09:39,480 --> 00:09:44,120 And so there was a, there was a young woman from Bangladesh, one 206 00:09:44,120 --> 00:09:46,400 of our first courses who contacted me. 207 00:09:47,480 --> 00:09:50,800 And she was like, Jeremy, you don't probably don't even know 208 00:09:50,800 --> 00:09:54,560 who I am, but I'm in Bangladesh and I'm a teenager. 209 00:09:54,560 --> 00:09:59,320 And she was like, I want to know if what I'm doing is OK because 210 00:09:59,320 --> 00:10:02,120 I feel shame. And she said, I don't know 211 00:10:02,120 --> 00:10:05,880 anybody else in my province that does anything with AII. 212 00:10:06,160 --> 00:10:08,440 Don't know any other girls that use computers. 213 00:10:09,200 --> 00:10:13,320 Everybody thinks I'm weird. I want you to know. 214 00:10:13,320 --> 00:10:16,880 I want to know if you think it's OK for me to do AI. 215 00:10:17,200 --> 00:10:19,440 Oh, she just needed the social encouragement. 216 00:10:19,440 --> 00:10:22,440 And I, and I, and I wrote back and I said, not only is it OK, 217 00:10:22,440 --> 00:10:26,680 but like, you know, you're going to put your province on the map, 218 00:10:26,840 --> 00:10:28,880 you know? And you know what, like a couple 219 00:10:28,880 --> 00:10:32,800 of years later, she wrote to me from Google in Silicon Valley. 220 00:10:32,800 --> 00:10:35,280 And she said, hey, thanks to you, I'm now a Google Scholar. 221 00:10:35,640 --> 00:10:37,600 They flew me over to San Francisco. 222 00:10:37,760 --> 00:10:40,360 What I like to do is I like to find these folks, mention them, 223 00:10:40,360 --> 00:10:43,080 train them, stand them up, and now they're leaders in their own 224 00:10:43,080 --> 00:10:44,640 communities. It's a, you know, quote, teach a 225 00:10:44,640 --> 00:10:47,000 man, teach a man to fish or teach a man to recognize an 226 00:10:47,000 --> 00:10:49,960 image of a fish, you know, right, so to speak, right. 227 00:10:50,640 --> 00:10:51,720 Actually, you know, you can use that. 228 00:10:51,720 --> 00:10:53,760 That's a good one liner. You know, 'cause you open with 229 00:10:53,800 --> 00:10:57,360 the you open with the bird thing from from XKCD. 230 00:10:57,520 --> 00:10:59,680 So teach a man to recognize an image of a fish or woman. 231 00:10:59,680 --> 00:11:01,680 You know, right? You know, the fish specifically 232 00:11:01,680 --> 00:11:04,640 you need to know is the tench, tench, Tench. 233 00:11:04,640 --> 00:11:07,880 Anybody who's understands computer vision knows about the 234 00:11:07,880 --> 00:11:10,120 tench. Yeah, because tench is the first 235 00:11:10,120 --> 00:11:13,040 Imagenet category. So anybody who's ever worked for 236 00:11:13,040 --> 00:11:13,840 the image net. Right. 237 00:11:14,200 --> 00:11:16,240 Yeah, yeah. So teach a man to recognize a 238 00:11:16,240 --> 00:11:17,840 tench, Yes, Yeah, that's. Good. 239 00:11:17,840 --> 00:11:19,640 That's right. Actually, that's like replaced 240 00:11:20,560 --> 00:11:21,840 Lena. Yes, exactly. 241 00:11:21,840 --> 00:11:25,160 Yes, that's right. OK, so let's see now why don't 242 00:11:25,160 --> 00:11:29,000 you give me the Jeremy life story So like before, so I know 243 00:11:29,240 --> 00:11:32,640 fast AI know Kaggle, I know answer AII know the COVID and 244 00:11:32,680 --> 00:11:35,840 and you know, masks what what's like What's the what's the so? 245 00:11:35,840 --> 00:11:40,720 Before Kaggle. So yeah, so Anthony and I kind 246 00:11:40,720 --> 00:11:45,200 of got Kaggle started in Melbourne in Australia, and then 247 00:11:45,200 --> 00:11:49,680 we flew out here. He had this crazy idea that 248 00:11:50,440 --> 00:11:54,240 venture capitalists in America would put money into our little 249 00:11:54,240 --> 00:11:55,840 start up. And I thought it was crazy. 250 00:11:55,840 --> 00:11:58,240 I thought there's no way. But he was right and I was 251 00:11:58,240 --> 00:12:00,280 wrong. It's like, OK, I'll, I'll come. 252 00:12:00,480 --> 00:12:02,880 I'll give it a go. But you know. 253 00:12:02,880 --> 00:12:04,600 Does Cagle have some? Is it? 254 00:12:04,600 --> 00:12:08,120 Is it an Australian is just sort of just like a funny word, 255 00:12:08,120 --> 00:12:10,000 made-up word? Just a made-up word. 256 00:12:10,080 --> 00:12:11,960 OK, Yeah. Like Google Cagle. 257 00:12:12,160 --> 00:12:15,440 OK, Yeah. And, and yeah, we, we, we spoke 258 00:12:15,440 --> 00:12:19,080 to some of your old colleagues. We spoke to Mark Marc Andreessen 259 00:12:20,840 --> 00:12:23,640 and it was interesting. At that time Andres and Horowitz 260 00:12:23,640 --> 00:12:26,480 hadn't done anything in machine learning and in the end they 261 00:12:26,480 --> 00:12:28,800 were very good about it. They passed on our round and 262 00:12:28,800 --> 00:12:31,560 they said, look, we don't know anything about machine learning. 263 00:12:31,640 --> 00:12:33,960 Maybe it's going to be a big deal, but we don't have anybody 264 00:12:33,960 --> 00:12:36,760 here that can judge that or not. But you know, so we ended up 265 00:12:37,160 --> 00:12:41,680 with like an old closler and other folks put the money in. 266 00:12:41,720 --> 00:12:45,880 But before that I had two startups in that I ran out of 267 00:12:45,880 --> 00:12:49,680 Australia. 1 was or Fastmail which became a very popular 268 00:12:49,680 --> 00:12:53,160 global e-mail company and then the other was called optimal 269 00:12:53,160 --> 00:12:56,320 decisions which if you're insurance you would definitely 270 00:12:56,320 --> 00:12:58,040 know and if you are not you definitely wouldn't. 271 00:12:58,080 --> 00:13:03,200 It basically trans changed how insurance companies price away 272 00:13:03,200 --> 00:13:06,240 from using just actuarial methods to using optimization 273 00:13:06,240 --> 00:13:08,160 based methods. Like convex optimization or 274 00:13:08,160 --> 00:13:09,640 something like that, no. Yeah, yeah. 275 00:13:09,640 --> 00:13:11,960 Just, you know, pretty classic optimization, But the key thing 276 00:13:11,960 --> 00:13:17,800 was to model elasticity and competitor price, not just risk. 277 00:13:18,120 --> 00:13:20,360 Because if all you do is model risk, all you can do is cost 278 00:13:20,360 --> 00:13:24,480 plus pricing, which as you know, is economically very suboptimal. 279 00:13:25,280 --> 00:13:29,200 So we make insurance companies a lot more profitable, which I 280 00:13:29,200 --> 00:13:31,040 have no pride over. In hindsight. 281 00:13:31,040 --> 00:13:34,360 I don't know why I spent years of my life working on that. 282 00:13:35,440 --> 00:13:41,280 But yeah, originally, I don't know, like coming out of school 283 00:13:42,040 --> 00:13:46,200 I was a bit lost to be honest, because like, I was interested 284 00:13:46,200 --> 00:13:48,120 in stuff that nobody else was interested in. 285 00:13:48,120 --> 00:13:51,680 So I was interested in like spreadsheets and databases and 286 00:13:51,680 --> 00:13:54,800 PCs. This is a bit over 30 years ago. 287 00:13:55,640 --> 00:13:58,680 I didn't know any other adults or kids that were interested in 288 00:13:58,680 --> 00:14:00,280 any of those things. You know, Australia. 289 00:14:00,400 --> 00:14:01,200 Yeah. OK. 290 00:14:01,560 --> 00:14:05,160 And there weren't any university courses you could go to that 291 00:14:05,160 --> 00:14:09,160 were about data. So I ended up doing philosophy, 292 00:14:09,960 --> 00:14:12,960 but I actually ended up not going to any classes because I 293 00:14:13,800 --> 00:14:18,080 happened to get a job at McKinsey and Company where they 294 00:14:18,480 --> 00:14:21,680 really appreciated this odd set of skills I had. 295 00:14:21,760 --> 00:14:27,120 So tell me about South McKenzie is actually interesting to me 296 00:14:27,120 --> 00:14:31,440 because there's the, let me give the negative and the positive 297 00:14:31,440 --> 00:14:35,360 view of McKenzie. So the negative view of McKenzie 298 00:14:35,360 --> 00:14:39,040 is, oh, you know you're hiring overpriced consultants to tell 299 00:14:39,040 --> 00:14:41,160 you to fire people and blah, blah, blah, blah, blah, right? 300 00:14:41,440 --> 00:14:44,840 And the positive view is it's something that takes young 301 00:14:44,840 --> 00:14:48,120 people and gives them lots of different kinds of business 302 00:14:48,120 --> 00:14:52,160 experience and, you know, lets them actually see the actual 303 00:14:52,160 --> 00:14:54,240 numbers of lots of businesses and actually trains people to 304 00:14:54,240 --> 00:14:56,560 make, of course, good slide decks and good presentations, 305 00:14:56,560 --> 00:14:58,920 but really to communicate well and understand the gears and 306 00:14:58,920 --> 00:15:02,360 nuts and bolts of businesses. And actually, when I've hired 307 00:15:02,360 --> 00:15:05,800 former McKinsey and Bain and so on people, they've actually done 308 00:15:05,800 --> 00:15:08,000 fairly well. They're they're very good 309 00:15:08,000 --> 00:15:11,040 non-technical athletes like power users or what have you, 310 00:15:11,040 --> 00:15:11,840 right? I don't know. 311 00:15:12,040 --> 00:15:14,520 Give me your thoughts on that. Oh, I mean, you know, I was in 312 00:15:14,520 --> 00:15:15,480 this unusual. Situation. 313 00:15:15,480 --> 00:15:16,920 Sorry to be negative, I didn't mean. 314 00:15:16,920 --> 00:15:19,400 It's just like the pro and oh, I love, I love like please, like 315 00:15:19,520 --> 00:15:20,760 challenge me. OK, go. 316 00:15:20,760 --> 00:15:22,880 If I say something worth challenging, challenge me 317 00:15:22,880 --> 00:15:25,520 because otherwise it's boring for everybody listening too and 318 00:15:25,520 --> 00:15:29,400 boring for me. Look, I started there when I was 319 00:15:29,400 --> 00:15:31,120 19, so. Oh. 320 00:15:31,160 --> 00:15:33,040 Really. Wow, That's, that's interesting. 321 00:15:33,040 --> 00:15:36,400 Yeah. So I was years younger than 322 00:15:36,400 --> 00:15:40,200 everybody else and for me it was eye opening and it was great 323 00:15:40,400 --> 00:15:43,640 because suddenly there were people who cared about what I 324 00:15:43,640 --> 00:15:46,280 did. And you're right, they're 325 00:15:46,280 --> 00:15:50,000 generally non-technical people. Just one of the reasons why as a 326 00:15:50,000 --> 00:15:53,680 19 year old I could be really successful there, you know. 327 00:15:55,720 --> 00:15:57,200 Did you feel you levelled up when you were there? 328 00:15:57,440 --> 00:15:59,760 Yes and no. It, it's funny you say it's this 329 00:15:59,760 --> 00:16:02,160 kind of polarizing thing. It was polarizing in my life 330 00:16:02,160 --> 00:16:03,960 too, right? Because at one level it's like I 331 00:16:04,160 --> 00:16:08,080 felt like, OK, I need to learn business because I didn't know 332 00:16:08,080 --> 00:16:10,680 any of that stuff. And I I wanted to create my own 333 00:16:10,680 --> 00:16:13,600 companies. Yeah, you're very commercial for 334 00:16:13,600 --> 00:16:15,720 a professor. Yeah, he's professor type. 335 00:16:15,880 --> 00:16:18,200 Yeah, yeah, yeah. Well, I mean, I never went into, 336 00:16:18,200 --> 00:16:20,760 I've never been a professional academic in my life. 337 00:16:20,800 --> 00:16:21,800 Right. But you've got, you've got the. 338 00:16:22,680 --> 00:16:24,160 I think we both have that disposition. 339 00:16:24,160 --> 00:16:25,680 Yeah, sure. No, absolutely. 340 00:16:26,120 --> 00:16:29,760 And so I was trying to learn business and by being at 341 00:16:29,760 --> 00:16:34,680 McKinsey, I, I did learn a lot about how business worked, but 342 00:16:34,680 --> 00:16:37,200 also in a lot of ways it's a very conservative organization 343 00:16:37,200 --> 00:16:39,320 because I, I was telling my colleagues at the time, Hey, 344 00:16:39,320 --> 00:16:43,960 this new Internet thing, I think it's going to be big, you know, 345 00:16:44,280 --> 00:16:46,760 and they'll be like, I don't know, Jeremy, this computer 346 00:16:46,760 --> 00:16:48,440 stuff, it's, this is pretty nerdy. 347 00:16:48,440 --> 00:16:52,200 It's like, what's it for? I'm like, I don't know exactly, 348 00:16:52,360 --> 00:16:56,960 but I feel like like very early 90s, I feel like a, it's going 349 00:16:57,000 --> 00:16:59,880 to impact business. And they're just like, no, look, 350 00:17:00,080 --> 00:17:01,920 let me explain how business works. 351 00:17:01,920 --> 00:17:05,119 You know, business is about relationships and strategy and 352 00:17:05,119 --> 00:17:09,800 capital and, you know, and in the end, like they were wrong, 353 00:17:10,480 --> 00:17:12,839 you know, and I didn't have the trust in myself. 354 00:17:13,280 --> 00:17:15,240 At the time, you didn't know whether you were wrong. 355 00:17:15,240 --> 00:17:18,880 Or I was sure I was wrong and I just kept trying to figure out 356 00:17:18,880 --> 00:17:21,880 why I'm so wrong. And I felt really upset with 357 00:17:21,880 --> 00:17:25,359 myself for being stupid that they everybody else can see it. 358 00:17:25,359 --> 00:17:28,960 It's so obvious that they're just like, look, Jeremy, let me 359 00:17:28,960 --> 00:17:30,760 try to explain it. I just couldn't get it. 360 00:17:32,080 --> 00:17:35,400 So I wish I had, you know, I stayed in consulting for 10 361 00:17:35,400 --> 00:17:36,600 years. Oh really? 362 00:17:36,600 --> 00:17:38,280 Wow. I should have done it just too, 363 00:17:38,520 --> 00:17:42,800 because that's enough. And like what I really learnt 364 00:17:42,800 --> 00:17:46,720 there was a sales. Like it's really grateful 365 00:17:46,720 --> 00:17:49,320 learning sales. What did you like? 366 00:17:50,200 --> 00:17:51,960 I don't know. What are the top 3 Five things 367 00:17:51,960 --> 00:17:53,520 you learn in McKinsey like sales? 368 00:17:53,640 --> 00:17:55,600 OK, Yeah. So I was and, and, and at IT 369 00:17:55,600 --> 00:17:57,440 county. So I went from from there to IT 370 00:17:57,440 --> 00:18:00,720 county. What what I learned was like, 371 00:18:00,760 --> 00:18:02,920 OK, it's all about change and influence, right? 372 00:18:03,040 --> 00:18:05,520 So it's not just sales, but it's a kind of sales. 373 00:18:05,520 --> 00:18:08,680 It's like you're trying to sell an idea or you're trying to sell 374 00:18:08,680 --> 00:18:14,800 a piece of work, whatever. So we were very careful about 375 00:18:14,800 --> 00:18:18,120 mapping out the organization, you know, so it's like, OK, we 376 00:18:18,120 --> 00:18:20,800 want to sell this piece of work next, or we want to help our 377 00:18:20,920 --> 00:18:24,960 client sell this idea. OK, who's everybody in the 378 00:18:24,960 --> 00:18:27,640 organization who's in any way a stakeholder who could have an 379 00:18:27,640 --> 00:18:31,080 opinion, who could cause this to succeed, who could cause this to 380 00:18:31,080 --> 00:18:32,920 fail? Like, OK, who do we know? 381 00:18:32,920 --> 00:18:39,320 Who knows that person and like extremely kind of careful and 382 00:18:39,320 --> 00:18:45,560 optimized process of creating change for human management, 383 00:18:45,560 --> 00:18:49,760 human connections. We brought professional actors 384 00:18:49,760 --> 00:18:53,600 in, like play the role of different types of clients and 385 00:18:53,600 --> 00:18:57,080 we would then interact with them and then, you know, then talk 386 00:18:57,080 --> 00:19:03,080 about what the results were. It was just way more intense 387 00:19:03,400 --> 00:19:06,480 human optimization than I'd ever conceived of. 388 00:19:06,480 --> 00:19:10,840 I'd always thought of that human side as being like, oh, some 389 00:19:10,840 --> 00:19:14,000 people are charismatic, you know, or, oh, some people are 390 00:19:15,040 --> 00:19:18,080 just good at convincing people. It's like, no, they're, they're 391 00:19:18,080 --> 00:19:20,400 skills. There's a science, there's a, 392 00:19:20,800 --> 00:19:23,280 there's a logic, There's a, there's like a different kind of 393 00:19:23,280 --> 00:19:26,920 logic to programming a computer. But if you want to get an 394 00:19:26,920 --> 00:19:31,400 organization to do a thing, you know, you have to know how to 395 00:19:31,400 --> 00:19:35,320 map it out and how to react. You know, in some ways it felt. 396 00:19:36,840 --> 00:19:39,760 Yeah. But in some ways it felt cold 397 00:19:39,760 --> 00:19:42,920 and kind of calculating and horrible to be like, oh, this 398 00:19:42,920 --> 00:19:44,880 human being. I don't seeing that as a human 399 00:19:44,880 --> 00:19:46,800 being. I'm seeing them as like this cog 400 00:19:46,800 --> 00:19:48,400 and this machine. And I'm going to use this 401 00:19:48,400 --> 00:19:52,560 process. But it totally worked, you know, 402 00:19:52,600 --> 00:19:56,360 And so it made me after a while, I, I changed my view of it. 403 00:19:56,360 --> 00:19:59,960 I was like, you know what? Like getting organizations to do 404 00:19:59,960 --> 00:20:02,680 things is important. It is also. 405 00:20:03,000 --> 00:20:07,320 Important and so if that involves treating people as 406 00:20:08,240 --> 00:20:14,480 machine parts sometimes because humans are very predictable, you 407 00:20:14,480 --> 00:20:18,880 know, and so if you learn how to manage different types of humans 408 00:20:18,880 --> 00:20:21,440 and different types of situations and like, you know, 409 00:20:21,520 --> 00:20:25,720 so like you get the one person, it'd be your kind of inside bowl 410 00:20:25,720 --> 00:20:29,680 who's like super who, who. And they've recognized that they 411 00:20:29,680 --> 00:20:31,880 can use you to advance their career. 412 00:20:31,880 --> 00:20:34,000 And then you talk to them specifically about how they can 413 00:20:34,000 --> 00:20:36,120 advance their career and then they tell you who's going to get 414 00:20:36,120 --> 00:20:38,880 in the way. And then you get 3 more people. 415 00:20:38,880 --> 00:20:42,440 And then you use that to put pressure on the 5th person who 416 00:20:42,440 --> 00:20:45,960 is well known to, you know, be somebody who likes following 417 00:20:45,960 --> 00:20:47,760 rather than leading. And you know, you structure it 418 00:20:47,760 --> 00:20:51,160 out, it'll play out. And at the end, it's like, OK, 419 00:20:51,760 --> 00:20:54,000 it happened, you know? It's funny, like the way you 420 00:20:54,000 --> 00:20:56,160 know, do you know Mark Craney at a 16 ZI don't know if you don't 421 00:20:56,160 --> 00:20:58,480 have him, he's a very different personality than you, but he 422 00:20:58,480 --> 00:21:04,360 also he's like a gruff Mormon a few words, but he's like a sales 423 00:21:04,360 --> 00:21:06,680 genius actually, right, and very similar. 424 00:21:06,680 --> 00:21:09,160 Like the way I think about it, that kind of reconciles all of 425 00:21:09,160 --> 00:21:12,720 it is it's a nested set of like win, win relationships all the 426 00:21:12,720 --> 00:21:14,360 way up to the organization level, right. 427 00:21:14,680 --> 00:21:18,160 Like the best kind of sales is when you are genuinely selling 428 00:21:18,160 --> 00:21:22,360 them something that will improve their business or their, their 429 00:21:22,360 --> 00:21:24,000 product or something in some way, right. 430 00:21:24,320 --> 00:21:26,880 And then it will also improve at a nested level of the career of 431 00:21:26,880 --> 00:21:28,880 this person who approves it. And so and so it's almost like a 432 00:21:29,200 --> 00:21:30,920 like a venture investment all the way through. 433 00:21:31,320 --> 00:21:33,920 And that is actually what I think is the reason that that 434 00:21:33,920 --> 00:21:36,120 will work is that's the most consistent kind of thing where 435 00:21:36,120 --> 00:21:39,040 even if you're flipping them to do it, they will like it in the 436 00:21:39,040 --> 00:21:40,280 medium to long run. Yeah. 437 00:21:40,280 --> 00:21:43,920 And if you're trying to have a dent on the world, you know, and 438 00:21:43,920 --> 00:21:47,920 you've got good ideas and develop good things, but you're 439 00:21:47,920 --> 00:21:52,880 unable to influence anybody to buy it or use it, then you're 440 00:21:52,880 --> 00:21:54,360 not going to have a dent on the world. 441 00:21:54,480 --> 00:21:56,200 Like that's actually, you know, it's funny, one of the I mean, 442 00:21:56,200 --> 00:21:57,960 there's a lot of great things about your course, but one of 443 00:21:57,960 --> 00:22:01,280 the best is the domain name fast at AI, right? 444 00:22:01,280 --> 00:22:03,400 Like I learn AI fast. Amazing. 445 00:22:03,400 --> 00:22:04,640 OK. That's what I want, right. 446 00:22:05,320 --> 00:22:07,840 So that's like an example of sort of an inbuilt marketing 447 00:22:07,840 --> 00:22:09,040 kind of thing, which is great, right. 448 00:22:09,200 --> 00:22:10,800 And I'm sure there was some thought into that, because lots 449 00:22:10,800 --> 00:22:11,560 of people could have. Named it. 450 00:22:11,560 --> 00:22:13,120 Oh, yeah. We did a lot of marketing stuff 451 00:22:13,120 --> 00:22:15,760 there. We we also, as far as I know, we 452 00:22:15,760 --> 00:22:18,800 were the first company in the world to do AB tests on our 453 00:22:18,800 --> 00:22:19,960 homepage. Oh, Zara. 454 00:22:19,960 --> 00:22:21,600 Interesting. I think we were also the first 455 00:22:21,600 --> 00:22:25,800 to have all the free e-mail accounts. 456 00:22:26,240 --> 00:22:30,200 A little photo would be added to every e-mail message, marketing 457 00:22:30,200 --> 00:22:32,160 the surface like we did a lot of little. 458 00:22:32,560 --> 00:22:34,600 Things like that. Things like that, little viral 459 00:22:34,600 --> 00:22:39,080 things that today everywhere. Yes, so, OK, great. 460 00:22:39,680 --> 00:22:42,760 Actually I want to show you something which is so we took. 461 00:22:42,920 --> 00:22:46,240 So let me describe problem and then solution and get your, your 462 00:22:46,240 --> 00:22:49,440 your thoughts right. So you and I have both taught 463 00:22:49,440 --> 00:22:52,800 large online courses, right? And the typical thing that 464 00:22:52,800 --> 00:22:56,200 happens with a large online course is people, it's a little 465 00:22:56,200 --> 00:23:00,000 bit like signing up for for like a workout, right? 466 00:23:00,240 --> 00:23:03,600 People aspirationally want to do it and then. 467 00:23:04,080 --> 00:23:07,200 They want to have done it. They want to have done it 468 00:23:07,200 --> 00:23:08,120 exactly. That's right and. 469 00:23:08,520 --> 00:23:11,400 Then they want to be the kind of person that would have done 470 00:23:11,400 --> 00:23:11,680 that. That's. 471 00:23:11,680 --> 00:23:12,600 Right. And there's something good out 472 00:23:12,600 --> 00:23:15,360 of that, right? But what happens is they sign up 473 00:23:15,360 --> 00:23:18,400 for and the problem is allocating the time or then if 474 00:23:18,400 --> 00:23:21,560 they have the time, the energy or the discouragement or what 475 00:23:21,560 --> 00:23:24,520 have you, there have been various mechanisms and so on to 476 00:23:24,520 --> 00:23:26,240 try to solve that, address that, right? 477 00:23:26,240 --> 00:23:28,560 There's like cohort based learning and you know, and so 478 00:23:28,560 --> 00:23:29,680 on. And those things work to an 479 00:23:29,680 --> 00:23:30,800 extent. What's so great? 480 00:23:31,320 --> 00:23:34,000 Yes, so that that can work. But let me show you something 481 00:23:34,000 --> 00:23:37,520 that we did, which we call a learn a thon. 482 00:23:37,720 --> 00:23:39,560 When should you use a random forest? 483 00:23:40,440 --> 00:23:42,480 What is a confusion matrix? Don't know. 484 00:23:42,720 --> 00:23:44,640 What about collaborative filtering? 485 00:23:45,200 --> 00:24:05,560 Don't know. When should you use a random 486 00:24:05,560 --> 00:24:08,360 forest tabular data? And if you have a lot of like 487 00:24:08,360 --> 00:24:10,920 noisy features, what is the confusion matrix? 488 00:24:10,920 --> 00:24:14,400 It's like a table of actual answers against like the 489 00:24:14,400 --> 00:24:17,000 predicted answers and then comparing, you know like how 490 00:24:17,000 --> 00:24:19,440 often it gets it right and then when and how much it gets it 491 00:24:19,440 --> 00:24:21,320 wrong. What is collaborative filtering 492 00:24:21,400 --> 00:24:24,800 recommendation algorithm by clustering people or items or 493 00:24:24,800 --> 00:24:29,800 things by similarity. So basically we're going to do 494 00:24:29,800 --> 00:24:31,040 a, you know, updated version of that. 495 00:24:31,040 --> 00:24:33,960 But basically, so the fastest, so essentially, literally we 496 00:24:33,960 --> 00:24:36,400 took because what it's like about 10 hours, 11 hours of 497 00:24:36,400 --> 00:24:40,160 videos, right? So over 2 days, we said, OK, you 498 00:24:40,160 --> 00:24:43,520 really want to do fast AI. OK, sign up, come here. 9:00 AM 499 00:24:43,840 --> 00:24:48,400 on on Saturday morning and 9:00 to 9:00 Saturday, 9:00 to 9:00 500 00:24:48,400 --> 00:24:51,480 Sunday. They watch every single video 501 00:24:51,480 --> 00:24:53,680 start to finish. No phones, right? 502 00:24:54,000 --> 00:24:57,000 And then when it was time to go and type things in, you know, 503 00:24:57,000 --> 00:24:59,240 laptops out, do that. Absolutely. 504 00:24:59,240 --> 00:25:01,560 And it drives me crazy because so many people tell me like, oh, 505 00:25:01,560 --> 00:25:05,560 Jeremy, I I started your course. I meant to finish. 506 00:25:05,680 --> 00:25:07,240 You know, I've tried three times. 507 00:25:07,240 --> 00:25:09,320 I haven't managed to finish. I always think, like, look, 508 00:25:09,640 --> 00:25:13,160 yeah, you could just put aside one weekend and just binge it, 509 00:25:13,280 --> 00:25:16,600 you know, get it done. Yes, exactly. 510 00:25:16,680 --> 00:25:19,240 And I want to did I, did I show you the fellowship video? 511 00:25:20,040 --> 00:25:21,760 OK, hold on, take a look at this. 512 00:25:21,920 --> 00:25:25,280 OK, Global meritocracy is finally here because we're 513 00:25:25,280 --> 00:25:28,880 awarding $100,000 in funding for the new Network School 514 00:25:28,880 --> 00:25:32,480 Fellowship, and anyone from anywhere can apply, and they 515 00:25:32,480 --> 00:25:35,360 might well ask how. Well, you see, we've set up shop 516 00:25:35,400 --> 00:25:38,160 on an island right off the coast of Singapore in the new Special 517 00:25:38,160 --> 00:25:41,720 Economic Zone, and it has an enlightened immigration policy 518 00:25:41,960 --> 00:25:44,560 that means it's the perfect place to assemble a global 519 00:25:44,560 --> 00:25:47,200 community of tech founders and AI creators. 520 00:25:47,360 --> 00:25:50,620 And that's what we've done. We've set up housing, food, Co 521 00:25:50,620 --> 00:25:55,040 worker fitness classes, yoga, fast Wi-Fi, office pods, a 522 00:25:55,040 --> 00:25:58,640 state-of-the-art gym, healthy snacks, Starlink, A makerspace, 523 00:25:58,760 --> 00:26:01,360 a content studio, guest lectures from the most successful 524 00:26:01,360 --> 00:26:04,760 founders and investors in the world, nomad visas and help with 525 00:26:04,760 --> 00:26:07,160 everything else you might need. And we have funding too, if 526 00:26:07,160 --> 00:26:09,560 you're good. So go and apply for the Nervous 527 00:26:09,560 --> 00:26:13,480 Tool Fellowship now, atns.com. The only connection you need is 528 00:26:13,480 --> 00:26:17,360 an Internet connection. That's very inspiring. 529 00:26:17,440 --> 00:26:19,680 I want to come. Great, SO. 530 00:26:19,680 --> 00:26:22,680 Also, Malaysia's awesome, so go to Malaysia. 531 00:26:23,360 --> 00:26:25,280 That's right. So basically the combination of 532 00:26:25,280 --> 00:26:29,240 Singapore, Malaysia and the new Singapore Johor Special Economic 533 00:26:29,240 --> 00:26:31,160 Zone, You know it, it was one of the things where there was 534 00:26:31,160 --> 00:26:33,600 theory and then somebody had to put that into practice, right? 535 00:26:33,880 --> 00:26:37,200 So the theory is like Singapore has a lot of capital but doesn't 536 00:26:37,200 --> 00:26:39,440 have a lot of land. Malaysia's actually improving a 537 00:26:39,440 --> 00:26:42,880 lot, but but it is. I mean, Malaysia's got a good 538 00:26:42,880 --> 00:26:45,520 education system. It's a strong. 539 00:26:45,520 --> 00:26:49,240 Country very underrated and it's improving a lot and you can 540 00:26:49,240 --> 00:26:52,200 basically live a pretty good life there I I think and it's 541 00:26:52,200 --> 00:26:54,840 right next door, right? So Malaysia has land and has 542 00:26:54,840 --> 00:26:56,320 less? You literally drive there you. 543 00:26:56,360 --> 00:26:58,080 Literally drive there. I literally drive back and forth 544 00:26:58,080 --> 00:26:59,680 all the time, right? In fact, we're just like 30 545 00:26:59,680 --> 00:27:00,880 minutes from Singapore. Basically. 546 00:27:00,880 --> 00:27:02,800 You're just literally, you know, just go over the bridge pop. 547 00:27:02,800 --> 00:27:05,320 You can see, you can see Singapore directly from from it, 548 00:27:05,320 --> 00:27:07,440 right? So, and we'll have probably have 549 00:27:07,440 --> 00:27:09,640 a ferry or something back and forth that'll get down to like 550 00:27:09,640 --> 00:27:10,720 15 minutes or so. Yeah. 551 00:27:10,960 --> 00:27:14,160 So I want like these autonomous boat kind of things, right? 552 00:27:14,200 --> 00:27:15,200 So why not? Yeah. 553 00:27:15,320 --> 00:27:17,640 So those knock on road, let's get let's get that right. 554 00:27:17,840 --> 00:27:20,480 So this is something, what you're seeing in that video is 555 00:27:20,480 --> 00:27:22,560 something I've wanted to do for more than 10 years, right? 556 00:27:23,240 --> 00:27:26,320 And you just have to build all the overnight thing 10 years in 557 00:27:26,320 --> 00:27:29,880 the making. So certainly anybody who's like 558 00:27:29,880 --> 00:27:33,040 doing fast AI, who's taking the deep learning courses, we're 559 00:27:33,040 --> 00:27:35,520 looking for the kinds of people who completed your course and we 560 00:27:35,520 --> 00:27:37,160 can fund them and help them build things. 561 00:27:37,400 --> 00:27:40,280 And in particular, the thing about, so let me explain kind of 562 00:27:40,280 --> 00:27:42,520 the motivation behind what we're doing in network school, right? 563 00:27:42,840 --> 00:27:46,480 So a, it's very hard obviously now to get student visas, 564 00:27:46,480 --> 00:27:48,160 skilled worker visas into the US. 565 00:27:48,160 --> 00:27:50,480 It's, I mean, even like people who are tourist visas, like 566 00:27:50,480 --> 00:27:53,280 they're getting strip searched or crazy things happen. 567 00:27:53,360 --> 00:27:55,520 You saw, there's actually some Australian or what have you, 568 00:27:55,520 --> 00:27:57,560 like some terrible thing happened to them or that right? 569 00:27:58,040 --> 00:28:00,720 And I. Think every almost every country 570 00:28:00,720 --> 00:28:03,720 now has some stories, examples of people, citizens of their 571 00:28:03,720 --> 00:28:05,240 country that have been screwed around. 572 00:28:05,400 --> 00:28:07,920 Tourist visas. Student visas, Skilled worker 573 00:28:07,920 --> 00:28:11,760 visas like the and. In Southeast Asia, these 574 00:28:11,760 --> 00:28:15,200 countries are now competing for that talent with their digital 575 00:28:15,200 --> 00:28:18,280 visas, with their startup visas. Exactly. 576 00:28:18,280 --> 00:28:19,200 It's. So smart. 577 00:28:19,280 --> 00:28:21,360 This is exactly that's right. And this is the thing I was 578 00:28:21,400 --> 00:28:23,160 like. I want Australia to get on that 579 00:28:23,160 --> 00:28:24,840 boat too. You know, we've had this global 580 00:28:24,840 --> 00:28:26,880 talent visa in Australia, which is pretty good. 581 00:28:27,080 --> 00:28:29,880 It's so yeah, I have it. Everybody's needs to do this. 582 00:28:29,880 --> 00:28:33,040 You know the the country's offering digital nomad visas, 583 00:28:33,040 --> 00:28:33,520 right? Yeah. 584 00:28:33,520 --> 00:28:36,000 So there's this weird thing where the US is taking itself 585 00:28:36,120 --> 00:28:39,040 out of the global economy. Yeah, just as everybody else. 586 00:28:39,280 --> 00:28:41,280 Everybody else is diving in and. Exactly. 587 00:28:41,320 --> 00:28:46,200 That's all of America's big value creators are tech. 588 00:28:46,360 --> 00:28:48,120 That's right, exactly. And they're globally mobile 589 00:28:48,120 --> 00:28:50,120 because there's no silicon in Silicon Valley. 590 00:28:50,160 --> 00:28:53,840 No, we're not like mining. So our team answer AI is fully 591 00:28:53,840 --> 00:28:58,760 distributed. So we folks in in Turkey, Japan, 592 00:28:59,200 --> 00:29:02,520 earlier Ireland. If you ever want to look at 593 00:29:02,520 --> 00:29:05,960 them, we can host them in every school for a week or a month or 594 00:29:05,960 --> 00:29:07,320 something like this. And one of the things you want 595 00:29:07,320 --> 00:29:09,480 to do is like Co location for remote teams. 596 00:29:09,480 --> 00:29:12,160 That's a nice idea because like, we've got together for the first 597 00:29:12,160 --> 00:29:14,240 time ever in person here in Singapore. 598 00:29:14,240 --> 00:29:16,200 Oh, great. And, and, and we're all like, 599 00:29:16,280 --> 00:29:18,000 oh, it's so nice to spend a week together. 600 00:29:18,640 --> 00:29:23,080 Eric Reese and I at Answer AI, we did something a bit unusual. 601 00:29:23,240 --> 00:29:25,640 We decided to only have one policy. 602 00:29:26,720 --> 00:29:31,560 And our only policy at Answer AI is to only have one policy. 603 00:29:33,080 --> 00:29:37,040 OK, what is that policy? The policy is to only have one 604 00:29:37,040 --> 00:29:40,080 policy. Oh, it's very mad. 605 00:29:40,080 --> 00:29:41,720 Is this like one of those recursive kind of things? 606 00:29:41,920 --> 00:29:43,320 Go ahead. I'm done, Dave. 607 00:29:43,320 --> 00:29:46,680 We only have one policy and it's to only have one policy, so it 608 00:29:46,680 --> 00:29:49,640 can't have no policies because that's a policy. 609 00:29:50,960 --> 00:29:51,840 OK. OK. 610 00:29:52,200 --> 00:29:55,400 So we have no policies other than the policy that we're only 611 00:29:55,400 --> 00:29:57,160 going to have one policy. I see. 612 00:29:57,200 --> 00:29:57,920 OK. Got it. 613 00:29:58,320 --> 00:30:00,440 Why? Well, policies, they're like 614 00:30:00,440 --> 00:30:01,880 ideologies. They're like, they're these 615 00:30:01,880 --> 00:30:05,200 fixed things which say like, oh, you can turn your brain off now 616 00:30:05,880 --> 00:30:10,480 because, because we've decided X, you know, in this situation, 617 00:30:10,480 --> 00:30:15,840 this is how you meant to behave. Like, I am equally sceptical of 618 00:30:16,040 --> 00:30:19,480 ideologies and policies and all of these cognitive shortcuts 619 00:30:19,480 --> 00:30:24,400 that basically say like, oh, I believe in this thing because 620 00:30:24,400 --> 00:30:26,520 that's what my ideology says, you know, or. 621 00:30:26,520 --> 00:30:28,640 Yes. So let me give an analogy or a 622 00:30:28,800 --> 00:30:31,080 way of thinking about this that I have from the Network State 623 00:30:31,080 --> 00:30:35,040 book, which is, you know, like programming paradigms, you can 624 00:30:35,040 --> 00:30:38,040 have imperative programming, functional programming, 625 00:30:38,040 --> 00:30:40,280 declarative programming and so on and so forth, right? 626 00:30:40,600 --> 00:30:45,680 And for certain problem demands, you know, certain style, it just 627 00:30:45,720 --> 00:30:49,080 makes it very easy and concise to solve that problem domain, 628 00:30:49,120 --> 00:30:51,400 right? But then you also want like a 629 00:30:51,400 --> 00:30:54,120 multi paradigm language like like Python or something like 630 00:30:54,320 --> 00:30:56,960 with Haskell, you know, you can just do everything as F of G of 631 00:30:56,960 --> 00:30:58,920 H of X and you can actually get far with that. 632 00:30:59,200 --> 00:31:01,720 But it's sometimes nice to do things in an imperative style or 633 00:31:01,720 --> 00:31:06,080 what have you, right. And and so that's how I think 634 00:31:06,080 --> 00:31:07,720 about political paradigms, right? 635 00:31:07,720 --> 00:31:11,760 Like I never announced she is, you know, I'm not a big UFC guy, 636 00:31:11,960 --> 00:31:14,920 but like Ultimate Fighting Championship is some people are 637 00:31:14,920 --> 00:31:19,320 using grappling, some boxing, some Muay Thai and it's 638 00:31:19,320 --> 00:31:22,400 situational as to just solve this with a kick or a punch, 639 00:31:22,400 --> 00:31:23,960 right? Just solve this as functional or 640 00:31:23,960 --> 00:31:26,480 imperative. And I think like Lee Kuan Yew 641 00:31:26,480 --> 00:31:29,080 was someone who is like that, where he understood many 642 00:31:29,080 --> 00:31:32,240 different political schools of thought and then he just like 643 00:31:32,240 --> 00:31:35,760 applied the right technique that was sort of self consistent in 644 00:31:35,760 --> 00:31:38,560 that school of thought for that situation, right? 645 00:31:39,000 --> 00:31:43,280 And so that's like the beyond ideology thing, which is you're 646 00:31:43,280 --> 00:31:45,160 aware of a lot of these different things and you 647 00:31:45,240 --> 00:31:48,680 situationally figure out which one is appropriate and you use 648 00:31:48,680 --> 00:31:51,400 that because. Andrew, you know, you're 649 00:31:51,400 --> 00:31:58,440 constantly curious and interested and you know what you 650 00:31:58,440 --> 00:32:03,400 care about is doing a good job. You know, rather than being 651 00:32:03,400 --> 00:32:06,520 consistent with other members of your tribe, most humans are 652 00:32:06,520 --> 00:32:08,920 mainly interested in being consistent with other members of 653 00:32:08,920 --> 00:32:10,680 their tribe. That's right, the number one 654 00:32:10,680 --> 00:32:12,800 driving force. And, and the thing about that is 655 00:32:12,800 --> 00:32:14,680 there's a, there's a meta rationality to that. 656 00:32:14,680 --> 00:32:17,400 I think it's kind of like, do you know, like evolutionary game 657 00:32:17,400 --> 00:32:21,360 theory, right? So like you can imagine you have 658 00:32:21,360 --> 00:32:24,840 two populations of people who are conformists and dissidents, 659 00:32:24,840 --> 00:32:27,400 so to speak, right? And the distance are constantly 660 00:32:27,400 --> 00:32:29,880 exploring and they're taking high risk and sometimes they 661 00:32:29,880 --> 00:32:32,400 fall off a Cliff and sometimes they have reward and the tribe 662 00:32:32,400 --> 00:32:34,920 follows them, right? And the conformists are just, 663 00:32:35,400 --> 00:32:38,800 you know, they're like, this is this is risk capital. 664 00:32:38,920 --> 00:32:41,480 And this is just, you know, stay home money or what have you, so 665 00:32:41,480 --> 00:32:44,080 to speak, right? So you can make an argument for 666 00:32:44,080 --> 00:32:47,360 a portfolio strategy as to why you want a small number of 667 00:32:47,360 --> 00:32:50,280 dissidents who are sometimes wrong or they're wrong or 668 00:32:50,280 --> 00:32:52,840 contrarians or whatever you want to call entrepreneurs, right? 669 00:32:53,280 --> 00:32:56,320 And then most people should actually like go the tribe so 670 00:32:56,320 --> 00:32:59,080 they don't run off a Cliff, but they could actually find, you 671 00:32:59,080 --> 00:33:01,200 know, a better, better pasture or something over here. 672 00:33:01,200 --> 00:33:03,800 That's that's one way of thinking about the respective 673 00:33:03,800 --> 00:33:04,480 balance. Go ahead. 674 00:33:04,480 --> 00:33:05,520 Yeah. I mean, I'm kind of curious 675 00:33:05,520 --> 00:33:08,680 about this because like globally, somehow every 676 00:33:08,680 --> 00:33:13,320 jurisdiction has settled on the same education system and the 677 00:33:13,320 --> 00:33:17,560 education system teaches children to be conformist. 678 00:33:17,920 --> 00:33:21,600 Yes. If you, if you, you know the 679 00:33:21,600 --> 00:33:24,880 test tests whether you can feedback the things you are 680 00:33:24,880 --> 00:33:28,000 taught in the way that you are taught them, you will get 681 00:33:28,000 --> 00:33:30,280 rewarded if you do what you're told. 682 00:33:30,560 --> 00:33:34,640 And like, I'm kind of curious about how much of this thing we 683 00:33:34,640 --> 00:33:37,840 see in the world is because every single child basically in 684 00:33:37,840 --> 00:33:41,160 the Western world at least has learnt this same. 685 00:33:41,160 --> 00:33:43,680 Do you know their behaviors? Have you heard that it comes to 686 00:33:43,680 --> 00:33:45,560 the Prussian educational system? Yeah. 687 00:33:45,840 --> 00:33:47,520 OK Do you know what preceded that? 688 00:33:47,880 --> 00:33:49,800 No. OK, so there's this great book 689 00:33:50,440 --> 00:33:52,480 we can put it on the screen called called the Craft 690 00:33:52,480 --> 00:33:53,800 Apprentice. OK. 691 00:33:54,280 --> 00:33:58,080 And one of my macro kind of theories of the world is that 692 00:33:58,320 --> 00:34:00,280 history is running in reverse. And I can show you a bunch of 693 00:34:00,280 --> 00:34:02,840 graphs on that or what have you. But literally like AU curve 694 00:34:03,160 --> 00:34:06,200 where in many ways our future is more like our past, like more 695 00:34:06,200 --> 00:34:09,360 like, let's say the 1850s and then eventually the 1750s and 696 00:34:09,360 --> 00:34:11,320 the 1950s. Like there's a lot of U curves 697 00:34:11,320 --> 00:34:13,679 which have their minimum or maximum in 1950. 698 00:34:13,679 --> 00:34:15,320 And I can I can show you some graphs on that. 699 00:34:16,000 --> 00:34:19,760 And so one premise of that is like prior to the Prussian 700 00:34:19,760 --> 00:34:22,639 educational system, which was which is what we currently know 701 00:34:22,639 --> 00:34:25,120 K through 12 and so on, that was all set up. 702 00:34:25,320 --> 00:34:28,440 It was inspired by Bismarck after German unification to have 703 00:34:28,440 --> 00:34:31,920 all the children get basically the same software in their 704 00:34:31,920 --> 00:34:33,199 heads. It's like, you know how with 705 00:34:33,199 --> 00:34:35,719 Windows, you have like the default install that comes off 706 00:34:35,719 --> 00:34:38,639 the factory and then you have like, you know, Windows premium 707 00:34:38,639 --> 00:34:40,960 ultimate, maybe for college graduates, and then you have the 708 00:34:40,960 --> 00:34:43,040 service packs from, you know, mainstream media. 709 00:34:43,560 --> 00:34:44,960 That's how I kind of think about right, right. 710 00:34:45,440 --> 00:34:47,880 And, and there's a reason for that because then everybody kind 711 00:34:47,880 --> 00:34:51,760 of has the same references. They they salute the flag and 712 00:34:51,760 --> 00:34:53,560 you know, they've just got the same basic install and they can 713 00:34:53,560 --> 00:34:55,360 interoperate, right? There's there's a rationale for 714 00:34:55,360 --> 00:34:57,560 that. It's how you, it's a softer part 715 00:34:57,560 --> 00:35:00,120 of constructing a nation. In fact, arguably that's even as 716 00:35:00,120 --> 00:35:01,720 important as quote, the hardware part, right? 717 00:35:02,280 --> 00:35:04,520 Which is like the physical territory and, and the people 718 00:35:04,520 --> 00:35:06,200 and so on. But before that, there's a 719 00:35:06,200 --> 00:35:09,000 different system, which was all based on apprenticeship. 720 00:35:09,200 --> 00:35:12,400 And they would start working from an early age and they would 721 00:35:12,400 --> 00:35:14,840 just learn practical skills very, very early on. 722 00:35:15,360 --> 00:35:18,480 Or they'd be like Jebediah and Abigail would have 12 kids and 723 00:35:18,480 --> 00:35:21,560 they'd all be working on the farm and they'd be like mini 724 00:35:21,560 --> 00:35:24,600 industrial robots, so to speak, picking fruit or something like 725 00:35:24,600 --> 00:35:26,960 that, you know, mending fences very, very early, early on. 726 00:35:27,200 --> 00:35:30,360 So the entire concept of extended adolescence wasn't 727 00:35:30,360 --> 00:35:31,600 there. The concept of being on your 728 00:35:31,600 --> 00:35:34,360 parents health insurance till 26 or whatever, it wasn't there. 729 00:35:34,800 --> 00:35:38,920 And now the reason that that stuff got introduced in part is 730 00:35:38,920 --> 00:35:43,120 because I think in the, in the late 1800s, with the advent of 731 00:35:43,120 --> 00:35:46,840 like industrialization in factories, these kids were no 732 00:35:46,840 --> 00:35:49,400 longer under the supervision of their parents or people the 733 00:35:49,400 --> 00:35:51,800 parents knew. They were under the supervision 734 00:35:51,800 --> 00:35:55,480 of factory owners who would push them too hard, right? 735 00:35:55,680 --> 00:35:58,720 Like these were like the child labour factories, you know, and 736 00:35:58,720 --> 00:36:01,320 so and so forth. And that was a dis alignment 737 00:36:01,560 --> 00:36:03,920 between like the interest of the factory owner and interest of 738 00:36:03,920 --> 00:36:05,000 the kids. That's when the child labour 739 00:36:05,000 --> 00:36:07,640 laws were passed and so on. I mean, that took a long time. 740 00:36:07,720 --> 00:36:09,080 It took a long time. It's like, what was it like 741 00:36:09,120 --> 00:36:11,960 60-70 years? Britain was the first in the 742 00:36:11,960 --> 00:36:13,960 world to introduce child labour laws but. 743 00:36:14,440 --> 00:36:16,680 Yes. Still took much longer than it 744 00:36:16,680 --> 00:36:17,880 should have. That's right, this little 745 00:36:17,880 --> 00:36:20,080 Dickensian kind of era or what have you, right. 746 00:36:20,360 --> 00:36:23,520 So then, so now there's a good to that at first, but then 747 00:36:23,760 --> 00:36:26,920 that's what actually led to the modern era of adolescence. 748 00:36:26,920 --> 00:36:30,720 And you know, I'm having fun as a kid for a long period of time. 749 00:36:30,760 --> 00:36:34,000 And now we have this extremely extended adolescence and 750 00:36:34,000 --> 00:36:36,840 training period where some people are like students as 751 00:36:36,840 --> 00:36:39,560 doctors all the way up into their 30s before they start 752 00:36:39,560 --> 00:36:41,280 their career. And they're almost middle age 753 00:36:41,280 --> 00:36:44,120 before they, you know, and I think that the, the, the 754 00:36:44,120 --> 00:36:47,800 corrective to that is 'cause everything good, you can always 755 00:36:47,800 --> 00:36:51,640 overdo it, right? And so you can go from quote, 756 00:36:51,720 --> 00:36:54,720 you know, like being opposing to child labor to not allowing 757 00:36:54,720 --> 00:36:57,960 people to even work until their, their 30s as a doctor, for 758 00:36:57,960 --> 00:37:01,200 example, right. So I think the opposite of that, 759 00:37:01,200 --> 00:37:05,880 the thesis antithesis synthesis is when the kid is at home and 760 00:37:05,880 --> 00:37:07,680 they're under the supervision of their parent, but they're able 761 00:37:07,680 --> 00:37:10,160 to start earning online by doing development, software 762 00:37:10,160 --> 00:37:13,480 development and so on. Even 1012 years ago, I had a 763 00:37:13,480 --> 00:37:15,280 bunch of kids. Some of my best students at at 764 00:37:15,280 --> 00:37:20,240 Stanford 1012 years ago were, were kids who had actually 765 00:37:20,240 --> 00:37:23,400 earned their first dollar doing online programming in their 766 00:37:23,400 --> 00:37:24,480 teens. Right. 767 00:37:24,920 --> 00:37:27,640 And it's not Even so much about the amount of money it is that 768 00:37:28,440 --> 00:37:29,720 it's that the market is a greater. 769 00:37:29,720 --> 00:37:31,480 This is how I think with with have you seen the grade 770 00:37:31,480 --> 00:37:33,320 inflation graphs? Yeah. 771 00:37:33,320 --> 00:37:36,120 So like, you know, you put that on screen, but basically kind of 772 00:37:36,120 --> 00:37:39,280 crazy. Everybody gets a 4.0. 773 00:37:39,280 --> 00:37:40,840 Basically students are the customers. 774 00:37:40,840 --> 00:37:42,320 So they're basically buying a job. 775 00:37:42,480 --> 00:37:45,120 And so how do you, how do you deal with that? 776 00:37:45,120 --> 00:37:47,200 And my answer is the market is a grader, right? 777 00:37:47,360 --> 00:37:49,520 So now you have kids that are doing software, they can't hurt 778 00:37:49,520 --> 00:37:51,920 themselves like in a factory, they're under supervision 779 00:37:51,920 --> 00:37:54,720 because they're working remote at home, but they're also like 780 00:37:54,720 --> 00:37:56,760 apprenticing, right? I think we network school. 781 00:37:56,760 --> 00:37:59,040 We also want to make that happen where now they're in a friendly 782 00:37:59,040 --> 00:38:01,120 environment along a bunch of other adults. 783 00:38:01,120 --> 00:38:04,000 So they're they can run around and and roam and so on. 784 00:38:04,520 --> 00:38:06,800 And then they can level up. They can be next to an 785 00:38:06,800 --> 00:38:08,880 electrical engineer, next to a mechanical engineer as they're 786 00:38:08,880 --> 00:38:11,480 building robots and stuff like that and just help them with 787 00:38:11,480 --> 00:38:13,880 small things, right? And they start to see what the 788 00:38:14,360 --> 00:38:17,560 like what adults are doing. And it's not just being, you 789 00:38:17,560 --> 00:38:19,320 know, sitting at a desk the whole day, right? 790 00:38:19,320 --> 00:38:21,120 So let me pause there. That's kind of how I'm thinking 791 00:38:21,120 --> 00:38:22,400 about part of the future education. 792 00:38:22,400 --> 00:38:24,200 Maybe you have some thoughts? I have a lot of thoughts. 793 00:38:25,280 --> 00:38:28,040 Yeah. So I mean, I, I know a lot of 794 00:38:28,320 --> 00:38:32,240 kids who are in that kind of interesting group who are 795 00:38:32,240 --> 00:38:34,920 basically ready to go to university when they're like 11 796 00:38:34,920 --> 00:38:38,400 or 12 and adults all try to stop them. 797 00:38:39,120 --> 00:38:41,200 Oh, interesting. It's like we don't for some 798 00:38:41,200 --> 00:38:43,040 reason. Well, the vast majority of 799 00:38:43,040 --> 00:38:46,600 adults I deal with don't want children to learn. 800 00:38:46,800 --> 00:38:49,520 When they're ready to learn, they have to learn at the speed 801 00:38:49,520 --> 00:38:51,640 which they're expected. To learn they want a speed 802 00:38:51,640 --> 00:38:54,240 limit. Yeah, and they assume any kid 803 00:38:54,240 --> 00:38:57,720 that's keen to learn more, it must be the parents fault that 804 00:38:58,320 --> 00:39:00,920 they're pushing them. Kids, kids are not allowed to 805 00:39:00,920 --> 00:39:04,800 have curiosity and drive and passion. 806 00:39:06,160 --> 00:39:10,040 But actually not every kid learns everything at the same 807 00:39:10,200 --> 00:39:11,280 speed. Yeah. 808 00:39:11,280 --> 00:39:15,160 So I'm very interested in like, how do we help this, that talent 809 00:39:15,680 --> 00:39:20,320 at the much younger age, not because I want to like make them 810 00:39:20,320 --> 00:39:23,600 more productive or whatever, but it's because I know so many of 811 00:39:23,600 --> 00:39:27,560 these kids are equally unhappy when they're artificially held 812 00:39:27,560 --> 00:39:29,240 back. And I want to all, you know, 813 00:39:29,240 --> 00:39:33,280 help them all have the opportunity to, to have that 814 00:39:33,280 --> 00:39:36,080 excitement of feeling like they're achieving their 815 00:39:36,080 --> 00:39:38,360 potential, that they're, that they're just really happy with 816 00:39:38,360 --> 00:39:41,120 the things they're building. So I've got a kid, you know, 817 00:39:41,120 --> 00:39:47,080 she's 9. And she's, we, we let her 818 00:39:47,200 --> 00:39:50,600 basically have whatever opportunities she wants, you 819 00:39:50,600 --> 00:39:54,200 know, and she chooses her curriculum and she chooses what 820 00:39:54,200 --> 00:39:56,600 she does. And she's happy for us to 821 00:39:56,800 --> 00:39:59,560 provide her some guidance as well, you know? 822 00:39:59,560 --> 00:40:03,520 But we don't force her to do anything. 823 00:40:03,600 --> 00:40:09,000 And yeah, she's got this great cohort of friends all around the 824 00:40:09,000 --> 00:40:13,800 world now who learn in this way and are all doing it at their 825 00:40:13,800 --> 00:40:17,000 own speed. Obviously with AI, there's a lot 826 00:40:17,000 --> 00:40:21,160 of opportunities to help more and more of these kinds of kids 827 00:40:21,680 --> 00:40:26,280 develop as they're ready, you know, and, and get a much more 828 00:40:26,840 --> 00:40:32,600 customized, personalized, dynamic education experience, 829 00:40:33,400 --> 00:40:37,280 one that's not focused on conformity or authority. 830 00:40:37,600 --> 00:40:39,200 You know, sometimes my daughter comes back. 831 00:40:39,200 --> 00:40:43,280 She's like, she does lots and lots of extracurricular things. 832 00:40:43,280 --> 00:40:45,320 You know, one of them is trampolining, and she comes back 833 00:40:45,320 --> 00:40:46,960 from trampolining. Sometimes she'll be like, oh, I 834 00:40:46,960 --> 00:40:48,960 got a Gold Star for good behavior. 835 00:40:49,280 --> 00:40:51,400 Isn't that great? And I would say, like, I don't 836 00:40:51,400 --> 00:40:53,840 know, I'm not sure I want you to have great behavior. 837 00:40:54,240 --> 00:40:59,120 You know, why do you think that's so important to have 838 00:40:59,120 --> 00:41:04,520 great behavior? Well, well, of course it depends 839 00:41:04,560 --> 00:41:08,240 obviously like a layer of dissidence and so on on top of a 840 00:41:08,240 --> 00:41:10,400 fundamentally pro social attitude is good. 841 00:41:10,600 --> 00:41:13,040 But if if people are like anti social and they're littering or 842 00:41:13,040 --> 00:41:14,840 they're, you know, yelling in the street, that's. 843 00:41:14,840 --> 00:41:18,400 No, exactly. It's, it's it's not necessarily, 844 00:41:18,440 --> 00:41:21,160 you know, being the best behaved kid in the class and getting the 845 00:41:21,160 --> 00:41:23,320 Gold Star that week is not necessarily the great thing. 846 00:41:23,320 --> 00:41:25,600 It's not, it's something not something I want her to be proud 847 00:41:25,600 --> 00:41:26,240 of. Right, Right. 848 00:41:26,800 --> 00:41:31,920 You know, yeah, she's incredibly pro social, she's incredibly 849 00:41:31,920 --> 00:41:34,600 kind, she's incredibly generous. But that doesn't mean she has to 850 00:41:34,600 --> 00:41:36,680 do everything she's told as soon as she's told to do it. 851 00:41:36,680 --> 00:41:37,240 That's. Right. 852 00:41:37,320 --> 00:41:38,720 And, and this is, it's funny you see us because. 853 00:41:39,040 --> 00:41:41,800 Basically, particularly for a girl, like, like, like, like 854 00:41:41,840 --> 00:41:46,600 girls are particularly taught to, to like fit in and do what 855 00:41:46,600 --> 00:41:49,520 they're told. And I don't want her to be 856 00:41:49,520 --> 00:41:51,840 somebody in society who just fits in and does what she's 857 00:41:51,840 --> 00:41:54,200 told. I think, I think this concept of 858 00:41:54,200 --> 00:41:59,360 like the balance and so on where it's like, you know, as you 859 00:41:59,360 --> 00:42:01,720 said, they're pro social and they're kind, but they also 860 00:42:01,720 --> 00:42:04,080 don't obey every single command and so, so. 861 00:42:04,240 --> 00:42:07,320 Yeah, I, I tend to focus on empathy with my daughter, which 862 00:42:07,320 --> 00:42:08,840 maybe ends up in a similar place. 863 00:42:09,120 --> 00:42:11,520 You know, just like, particularly for younger kids, 864 00:42:11,520 --> 00:42:13,920 empathy doesn't necessarily come as easily. 865 00:42:13,920 --> 00:42:17,560 So I have to kind of say like, OK, you thought that was funny. 866 00:42:18,480 --> 00:42:22,320 Now can you try to imagine what that person's situation was? 867 00:42:22,400 --> 00:42:25,440 Do you think they would have found it funny if you were them 868 00:42:25,440 --> 00:42:28,120 in that situation? Has anything similar happened to 869 00:42:28,120 --> 00:42:30,440 you before? And eventually someone just 870 00:42:30,880 --> 00:42:34,080 said, wow, did I just do that thing to them that that other 871 00:42:34,080 --> 00:42:35,640 person did to me? That made me sad. 872 00:42:35,640 --> 00:42:39,240 Like, oh wow, I feel so sad. I didn't want to make upset that 873 00:42:39,240 --> 00:42:41,880 person. It, it's funny because you know, 874 00:42:41,880 --> 00:42:44,520 sometimes you can get to like, just like with religions, you 875 00:42:44,520 --> 00:42:47,000 can often get to a similar behavior pattern by different 876 00:42:47,000 --> 00:42:50,160 kinds of religion. So I had a, a recent tweet a 877 00:42:50,160 --> 00:42:52,720 little bit viral on, on, on actually that exact topic of 878 00:42:52,720 --> 00:42:56,400 empathy. And essentially what I said is 879 00:42:56,600 --> 00:43:00,720 because I was, I was talking to conservatives and I was saying, 880 00:43:00,720 --> 00:43:03,680 look, empathy is actually a useful concept even for a 881 00:43:03,680 --> 00:43:05,640 completely cold blooded capitalist, right? 882 00:43:06,080 --> 00:43:07,520 Why? Because you have to understand 883 00:43:07,520 --> 00:43:09,960 other guy's point of view and their win win, right? 884 00:43:10,400 --> 00:43:15,000 And a lot of the like, especially in today's America, 885 00:43:15,000 --> 00:43:18,400 they've gotten themselves in the in the mental state. 886 00:43:18,640 --> 00:43:20,520 They think everybody's exploiting them, everybody's 887 00:43:20,520 --> 00:43:23,760 ripping them off, right? And that like Australia's an 888 00:43:23,760 --> 00:43:27,480 enemy and Canada's an enemy and Vietnam is an enemy and 889 00:43:27,480 --> 00:43:31,600 whatever, right? And it's like, you know, lots of 890 00:43:31,600 --> 00:43:34,640 people are just neutral, right? They're just business partners 891 00:43:34,640 --> 00:43:36,160 or they're just like living their lives. 892 00:43:36,160 --> 00:43:37,760 And you don't have to like fight. 893 00:43:37,760 --> 00:43:39,400 And you can't fight the entire world. 894 00:43:39,400 --> 00:43:42,120 And you also have to have some understanding of, OK, what's 895 00:43:42,120 --> 00:43:43,680 their win? And how can we get to a win win? 896 00:43:43,680 --> 00:43:45,800 And often a win win is more profitable for both parties 897 00:43:45,800 --> 00:43:47,600 involved and and so on and so forth, right? 898 00:43:48,080 --> 00:43:49,360 So. You can. 899 00:43:49,400 --> 00:43:54,120 And actually, altruism is programmed into us like this is 900 00:43:54,120 --> 00:43:57,360 something we've discovered, like evolutionarily being programmed 901 00:43:57,360 --> 00:44:00,440 into all of us. To not be altruistic is to fight 902 00:44:00,760 --> 00:44:05,440 against your basic instincts. And that's really dangerous 903 00:44:05,440 --> 00:44:09,480 because when you fight against things that evolution has 904 00:44:09,480 --> 00:44:13,160 programmed you to do, you're creating a new unstable 905 00:44:13,160 --> 00:44:16,840 equilibrium, you know? So why has that happened? 906 00:44:16,840 --> 00:44:21,360 Well, presumably there were plenty of groups that had no 907 00:44:21,360 --> 00:44:24,560 altruism in their villages. You know, just genetically they 908 00:44:24,800 --> 00:44:26,240 didn't have that as part of their DNA. 909 00:44:26,240 --> 00:44:27,480 They can. Cooperate and they died out. 910 00:44:27,600 --> 00:44:32,400 They died out, you know. And so we, we as a species, you 911 00:44:32,400 --> 00:44:36,000 know, we're not perfect, right? But you don't want to 912 00:44:36,000 --> 00:44:39,360 underestimate the power of what we're born with, you know, when 913 00:44:39,360 --> 00:44:41,880 we're born, you know, altruism is not weakness. 914 00:44:42,200 --> 00:44:46,040 Altruism is, is strength. These are the people that 915 00:44:46,040 --> 00:44:48,240 survived. And if you want to fight against 916 00:44:48,240 --> 00:44:50,680 that, then you're fighting against a basic survival 917 00:44:50,680 --> 00:44:52,800 instinct. Also, it's, it's nigh on 918 00:44:52,800 --> 00:44:57,800 impossible to design and organize such a complex system. 919 00:44:57,880 --> 00:45:02,640 They, they arise over a very long period of time to create 920 00:45:02,640 --> 00:45:05,800 these marvelously stable equilibria. 921 00:45:05,960 --> 00:45:10,040 You know, and this is what kind of terrifies me at the moment is 922 00:45:10,040 --> 00:45:14,120 there are so many opportunities to destabilize the equilibrium 923 00:45:14,160 --> 00:45:17,360 right now, you know, with with technology and the connectivity 924 00:45:17,360 --> 00:45:20,360 we have. And historically, each time you 925 00:45:20,360 --> 00:45:24,920 get a previously stable equilibrium is damaged, 926 00:45:25,120 --> 00:45:29,800 sometimes ending up with, you know, hundreds of years of 927 00:45:29,840 --> 00:45:33,160 societal misery. And so I always just like, I'm 928 00:45:33,160 --> 00:45:37,680 definitely very keen to see change and growth, but I want 929 00:45:37,680 --> 00:45:44,040 people to understand the power of where we're at and know how 930 00:45:44,160 --> 00:45:47,400 hard it was to get there. And, and to know enough history 931 00:45:47,560 --> 00:45:50,720 to know that, you know, creating, you know, 932 00:45:50,720 --> 00:45:54,120 destabilizing an equilibrium creates a power vacuum. 933 00:45:54,680 --> 00:46:00,120 And there are certain people who are extremely motivated and good 934 00:46:00,120 --> 00:46:02,200 at taking advantage of power vacuums. 935 00:46:02,600 --> 00:46:04,440 And the pair, the people you definitely don't want to 936 00:46:04,440 --> 00:46:08,760 empower, you know, I don't know, like somehow Singapore did an 937 00:46:08,760 --> 00:46:10,680 amazing job. Like the one country in the 938 00:46:10,680 --> 00:46:13,640 world that like, I think they just got lucky with Lee Kuan 939 00:46:13,640 --> 00:46:15,480 Yew, you know what I mean? They ended up with a guy who's 940 00:46:16,000 --> 00:46:18,640 kind of incorruptible. He doesn't have a huge chip on 941 00:46:18,640 --> 00:46:21,880 his shoulder. He just cares about outcomes. 942 00:46:22,200 --> 00:46:25,680 Most places around the world in that situation end up with, you 943 00:46:25,680 --> 00:46:30,760 know, basically a, you know, deeply insecure, chip on their 944 00:46:30,760 --> 00:46:35,120 shoulder power hungry person. You know, it's funny about Lee 945 00:46:35,120 --> 00:46:37,280 Kuan Yew, which I think is very underappreciated. 946 00:46:37,840 --> 00:46:43,120 Is he like he could argue his case in English? 947 00:46:43,240 --> 00:46:45,680 I think this is the most underappreciated aspect of Lee 948 00:46:45,680 --> 00:46:49,400 Kuan Yew because he would argue his case in English that he 949 00:46:49,400 --> 00:46:51,680 could argue on the global stage, right? 950 00:46:51,920 --> 00:46:53,800 Other people understood at least his point of view. 951 00:46:53,800 --> 00:46:56,520 He can make it cogently, he could do it in short form, he 952 00:46:56,520 --> 00:46:59,760 could do it in long form, sound bites and then, you know, long 953 00:46:59,880 --> 00:47:02,160 speeches extemporaneously or in policy papers. 954 00:47:02,800 --> 00:47:05,520 And he made sure that Singapore won the argument. 955 00:47:06,080 --> 00:47:09,800 If you win the argument, then you often don't have to fight, 956 00:47:10,600 --> 00:47:12,600 right? Because there's like that swing 957 00:47:12,600 --> 00:47:14,200 vote in the middle. He's like, you know what, he has 958 00:47:14,200 --> 00:47:16,160 a point here. We should do it his way and so 959 00:47:16,160 --> 00:47:18,840 on and so forth, right? And I feel that, for example, 960 00:47:18,840 --> 00:47:22,240 there's other other folks in East Asia who delivered 961 00:47:22,240 --> 00:47:25,040 comparable economic results to LKY, right? 962 00:47:25,840 --> 00:47:29,040 For example, in, in South Korea or in Taiwan or what have you. 963 00:47:29,040 --> 00:47:32,160 But they couldn't make their argument in English, right? 964 00:47:32,160 --> 00:47:35,280 That's a really exceptional aspect of they could speak in 965 00:47:35,280 --> 00:47:38,480 Korean, they could speak in Chinese, but like they couldn't, 966 00:47:38,680 --> 00:47:40,880 they couldn't make their case on a global stage, right? 967 00:47:41,400 --> 00:47:45,600 And and I think that's very underrated and it's something I 968 00:47:45,600 --> 00:47:49,480 think about a lot because so let me let me actually slightly 969 00:47:49,480 --> 00:47:51,840 counter argue with you on the power vacuum thing, which is 970 00:47:51,840 --> 00:47:54,480 there, right. I think that we are about to 971 00:47:54,480 --> 00:47:59,560 enter a period where the the future is China versus the 972 00:47:59,560 --> 00:48:00,800 Internet. Should I elaborate on what I 973 00:48:00,800 --> 00:48:02,760 mean by that? I know versus the Internet. 974 00:48:02,760 --> 00:48:07,080 China versus the Internet. So the 20th century was sort of 975 00:48:07,080 --> 00:48:10,160 a symmetric thing, you know, almost like basketball at the 976 00:48:10,160 --> 00:48:12,840 Final four plays. And it then ends up as US versus 977 00:48:12,840 --> 00:48:15,680 USSR. Everybody slugs it out, right? 978 00:48:15,920 --> 00:48:18,480 Sean Mcmeekin has this book called Stalin's War where he 979 00:48:18,480 --> 00:48:21,360 kind of makes the point that World War One, World War 2 can 980 00:48:21,360 --> 00:48:24,040 be seen almost as like a 30 years war, like an extended bar 981 00:48:24,040 --> 00:48:27,440 brawl with people like smashing chairs over each other's heads 982 00:48:27,440 --> 00:48:30,360 all around the world, right? And then it kind of lands up as 983 00:48:30,360 --> 00:48:33,520 the US versus USSR, right, with Japan and Germany eliminated 984 00:48:33,520 --> 00:48:38,000 and, and, and, and other powers too, UUSUK, France, blah, blah, 985 00:48:38,000 --> 00:48:39,320 right? I think this century is going to 986 00:48:39,320 --> 00:48:43,000 be different where it's a not a symmetric thing, but asymmetric 987 00:48:43,320 --> 00:48:46,600 like China and the Internet are. I think the balancing things and 988 00:48:46,600 --> 00:48:49,280 China is obvious and I think the Internet is not obvious. 989 00:48:49,280 --> 00:48:50,560 What do you mean with China is obvious? 990 00:48:50,560 --> 00:48:55,200 China, if you take the quote American empire, I think China 991 00:48:55,200 --> 00:49:00,400 inherits the manufacturing and the money and the military, not 992 00:49:00,400 --> 00:49:03,000 all the money, but the manufacturing, the military and 993 00:49:03,280 --> 00:49:06,720 really the might of it globally like what the alliances and so 994 00:49:06,720 --> 00:49:08,720 on. The world is after this tariff 995 00:49:08,720 --> 00:49:10,960 thing recentralizing around China, right? 996 00:49:10,960 --> 00:49:12,520 Quickly it'll. Be interesting to see how 997 00:49:13,320 --> 00:49:16,160 eminent that is, but it's it's something very deep happening 998 00:49:16,160 --> 00:49:16,760 there. Yeah. 999 00:49:16,760 --> 00:49:18,440 So, so I think what's going? To happen. 1000 00:49:18,480 --> 00:49:21,040 And it's not just economically, also culturally. 1001 00:49:21,040 --> 00:49:24,000 You know, America's cultural power has been enormous. 1002 00:49:24,320 --> 00:49:25,600 It, it has been. That's right. 1003 00:49:25,720 --> 00:49:29,320 And now in Australia, I'm seeing people being like, oh, America's 1004 00:49:29,800 --> 00:49:32,240 kind of cringe now. It's cringe now, that's right. 1005 00:49:32,240 --> 00:49:37,720 But I think that the other air that the less visible, but as 1006 00:49:37,720 --> 00:49:40,920 important air is the Internet, which it has the people, the 1007 00:49:40,920 --> 00:49:43,200 values and the language. OK. 1008 00:49:43,200 --> 00:49:47,080 And the reason I say that is the only thing that has economic 1009 00:49:47,080 --> 00:49:50,160 scale comparable to China is actually the Internet like so 1010 00:49:50,160 --> 00:49:52,400 that's that why why am I into crypto? 1011 00:49:52,400 --> 00:49:55,240 I'm into crypto because everybody in the Internet is 1012 00:49:55,240 --> 00:49:57,880 equal, meaning you're peer-to-peer. 1013 00:49:57,880 --> 00:49:59,320 You can send packets back and forth. 1014 00:49:59,320 --> 00:50:01,280 You have the same property rights, you have the same 1015 00:50:01,280 --> 00:50:03,600 contract law, right? You have the same monetary 1016 00:50:03,600 --> 00:50:06,000 policy. And so whatever you were born 1017 00:50:06,000 --> 00:50:09,920 into, you can opt in to a system of law that is superior to the 1018 00:50:09,920 --> 00:50:13,760 one that you were born into. And it's like emigrating to at 1019 00:50:13,760 --> 00:50:15,760 least half of what a government is, right? 1020 00:50:15,760 --> 00:50:17,680 It's not the land, it's not the physical territory yet. 1021 00:50:17,680 --> 00:50:20,320 I'll come to that. But it's at least the property 1022 00:50:20,320 --> 00:50:22,760 rights. And you have to have some 1023 00:50:22,760 --> 00:50:24,640 sacrifice. You have to buy some of the coin 1024 00:50:24,640 --> 00:50:26,520 or whatever. You start interacting with this, 1025 00:50:26,880 --> 00:50:30,600 now you have like a system of law that's often superior to the 1026 00:50:30,600 --> 00:50:34,560 one that you inherited, whether is in Nigeria or is in, you 1027 00:50:34,560 --> 00:50:35,960 know, Lebanon OR something like that. 1028 00:50:35,960 --> 00:50:37,360 These places have destroyed currencies. 1029 00:50:37,360 --> 00:50:38,520 They don't guard property rights. 1030 00:50:38,760 --> 00:50:42,440 Now you can finally save because you know, the, the, the, the 1031 00:50:42,440 --> 00:50:44,040 blockchain protects your savings, right? 1032 00:50:44,120 --> 00:50:47,680 So I think that the Internet has half of what we want, which is 1033 00:50:47,680 --> 00:50:50,320 it has a system of government and with all these blockchains, 1034 00:50:50,320 --> 00:50:53,720 multiple systems of government. And it actually compare it, one 1035 00:50:53,720 --> 00:50:55,320 of the ways to think about it is, you know, with early 1036 00:50:55,320 --> 00:50:57,520 America, it didn't actually think of itself as America at 1037 00:50:57,520 --> 00:50:59,120 first. They were British colonists, 1038 00:50:59,200 --> 00:51:00,400 right? They're they're, you know, like 1039 00:51:00,680 --> 00:51:03,440 the the, you know, Virginia colony, Massachusetts colony, 1040 00:51:03,520 --> 00:51:05,600 and they had a land of native people, but they didn't have a 1041 00:51:05,600 --> 00:51:07,960 government, right? Because the government was in 1042 00:51:07,960 --> 00:51:10,640 London and took a while for them to develop a sense of national 1043 00:51:10,640 --> 00:51:12,880 consciousness and realize, oh, that's actually not our 1044 00:51:12,880 --> 00:51:14,760 government. Our government is here, right? 1045 00:51:15,040 --> 00:51:16,920 So they had land people in government, they became America, 1046 00:51:17,000 --> 00:51:18,480 right? And I think the Internet is 1047 00:51:18,480 --> 00:51:21,400 evolving in the opposite way. It has a people and actually as 1048 00:51:21,400 --> 00:51:23,560 a government in the form of the blockchain, but doesn't it 1049 00:51:23,560 --> 00:51:26,520 doesn't need to have land. I think that's the next step. 1050 00:51:26,600 --> 00:51:30,800 And hopefully it won't be verses unfortunately, Jinping has moved 1051 00:51:30,800 --> 00:51:36,280 into a power vacuum in China. Prior to that, actually, China 1052 00:51:36,840 --> 00:51:39,880 was much more of a democracy than people realized. 1053 00:51:41,200 --> 00:51:44,160 Talk about this. Well, I think a lot of people 1054 00:51:44,160 --> 00:51:47,320 don't understand how the political situation in China 1055 00:51:47,320 --> 00:51:50,080 worked. So there was a lot of voting, 1056 00:51:50,520 --> 00:51:53,920 but unlike most Western democracies, the voting was 1057 00:51:53,920 --> 00:51:56,680 entirely within the the party party. 1058 00:51:57,480 --> 00:51:58,960 And people might think, oh, that's not very big. 1059 00:51:58,960 --> 00:52:00,400 It's. Actually, 100 million people. 1060 00:52:00,800 --> 00:52:01,640 Chase Guy was very big. Yeah. 1061 00:52:01,640 --> 00:52:03,720 And then you go. To the and and it's not and like 1062 00:52:04,160 --> 00:52:07,840 of my so I spent a lot of time in China and with a lot of 1063 00:52:08,240 --> 00:52:11,640 really great people in China, young people and the vast 1064 00:52:11,640 --> 00:52:16,160 majority of the best of the people, most what they wanted to 1065 00:52:16,160 --> 00:52:21,160 do was to get into the party. Well, the the not commenting on 1066 00:52:21,160 --> 00:52:24,120 whether this is good or bad, but it ends up with a kind of a 1067 00:52:24,120 --> 00:52:28,360 democracy of, you know, the the hardest working, most 1068 00:52:28,360 --> 00:52:30,680 intellectually capable people. Can I make a provocative 1069 00:52:30,760 --> 00:52:33,000 comment? So there's a book called The 1070 00:52:33,000 --> 00:52:36,240 Party Decides. The point of that book was the 1071 00:52:36,240 --> 00:52:39,960 American Union Party decides who's actually running on the 1072 00:52:39,960 --> 00:52:42,040 Democrat and Republican side. For many years there people have 1073 00:52:42,040 --> 00:52:44,120 said a choice non echo or whatever, right? 1074 00:52:44,640 --> 00:52:47,440 And so there's a similarity to that where there were quote, 1075 00:52:47,440 --> 00:52:49,320 smoke filled rooms where the candidate was determined. 1076 00:52:49,320 --> 00:52:51,920 And certainly with the recent Democrat primary, it was 1077 00:52:51,920 --> 00:52:54,440 something where basically the party determined who was 1078 00:52:54,440 --> 00:52:55,560 running. And and so and so and then 1079 00:52:55,560 --> 00:52:57,640 there's a whole disaster, the whole Biden comma thing. 1080 00:52:58,040 --> 00:53:01,280 So there's more similarity to the American system for many 1081 00:53:01,280 --> 00:53:04,240 years where there was essentially A uniparty that 1082 00:53:04,240 --> 00:53:06,680 decided like who who the candidates were. 1083 00:53:07,440 --> 00:53:10,920 Then then some would argue and now I'd say in a sense we've had 1084 00:53:10,920 --> 00:53:13,760 true democracy burst forth, but that's some people conceptualize 1085 00:53:13,760 --> 00:53:14,960 this democracy. Let me pause there. 1086 00:53:15,080 --> 00:53:18,040 Yeah. So yeah, so that's another whole 1087 00:53:18,040 --> 00:53:20,200 kind of worms I'll leave aside for a moment, which is that 1088 00:53:20,200 --> 00:53:22,600 actually, yeah, there's there's actually a lot more conspiracies 1089 00:53:22,600 --> 00:53:24,520 in the world than people realize. 1090 00:53:24,520 --> 00:53:25,920 Hey, there's a lot of smoke filled rooms. 1091 00:53:25,920 --> 00:53:27,240 I've been in plenty of them. Yeah. 1092 00:53:27,760 --> 00:53:30,600 But I think I just wanted to mention is the the the the thing 1093 00:53:30,600 --> 00:53:34,640 that was missing in what you said is the is the key power for 1094 00:53:34,640 --> 00:53:36,920 me. The key issue for me which is 1095 00:53:36,920 --> 00:53:38,800 the presence of positive feedback loops. 1096 00:53:38,800 --> 00:53:40,480 And I want to say positive feedback loop. 1097 00:53:40,480 --> 00:53:43,280 I don't mean good feedback loop, I mean a feedback loop which 1098 00:53:43,280 --> 00:53:44,960 goes back and causes more of itself. 1099 00:53:45,040 --> 00:53:49,320 So power and wealth. Like like viral reproduction? 1100 00:53:49,760 --> 00:53:52,320 Something like that. But like power and wealth are 1101 00:53:52,320 --> 00:53:54,480 naturally positive feedback loops. 1102 00:53:55,280 --> 00:53:58,160 Getting more power puts you in a position to be able to get more 1103 00:53:58,160 --> 00:54:00,040 power. Getting more wealth puts you in 1104 00:54:00,040 --> 00:54:02,200 a position to get more wealth. And then you've got the cross 1105 00:54:02,200 --> 00:54:03,960 correlation. Getting more power helps you get 1106 00:54:03,960 --> 00:54:05,520 more wealth. Getting more wealth helps you 1107 00:54:05,520 --> 00:54:07,960 get more power. I talked to him earlier about 1108 00:54:07,960 --> 00:54:09,680 the importance of a stable equilibrium. 1109 00:54:10,320 --> 00:54:14,680 How can you get a stable equilibrium in a situation where 1110 00:54:14,680 --> 00:54:17,360 somebody getting ahead can let them get more ahead? 1111 00:54:17,440 --> 00:54:19,960 Right, There's a the. Compounding interest. 1112 00:54:19,960 --> 00:54:21,840 There's a huge tension here, right? 1113 00:54:22,360 --> 00:54:26,880 And this is where democracy and capitalism and the market 1114 00:54:27,120 --> 00:54:31,280 economy come into a huge, a huge problem, right? 1115 00:54:31,280 --> 00:54:36,600 Which is if if you allow those positive feedback loops that 1116 00:54:36,600 --> 00:54:42,960 happen, then you end up with people who have incredible 1117 00:54:42,960 --> 00:54:45,800 riches and incredible power because they're on the right 1118 00:54:45,800 --> 00:54:47,520 side of that feedback loop you. Know. 1119 00:54:47,840 --> 00:54:51,960 Yes, OK. And so it's a natural is that 1120 00:54:51,960 --> 00:54:57,440 equilibrium and it's not compatible with actual market 1121 00:54:57,440 --> 00:55:01,800 forces or with democracy because you're now in a situation where 1122 00:55:01,800 --> 00:55:05,520 you can like you can buy the media, you know, you can, or 1123 00:55:05,520 --> 00:55:08,840 nowadays like the social networks or whatever you can, 1124 00:55:09,160 --> 00:55:11,800 you know, all the odds in your favor. 1125 00:55:12,400 --> 00:55:17,440 And that is not, again, that's not a resilient state to be in. 1126 00:55:17,440 --> 00:55:22,240 So somehow many societies in the world have managed to create 1127 00:55:22,280 --> 00:55:27,320 sophisticated complex equilibria that have avoided this for 1128 00:55:27,320 --> 00:55:31,200 decades, you know, but it's not the natural state of things. 1129 00:55:31,320 --> 00:55:34,960 The natural state of things is for there to be, you know, 1 1130 00:55:34,960 --> 00:55:39,280 incredibly wealthy and powerful person that you know is there 1131 00:55:39,280 --> 00:55:41,440 because of the pair of positive feedback. 1132 00:55:41,560 --> 00:55:45,360 OK, so let me let me disagree with that in two ways and then 1133 00:55:45,640 --> 00:55:47,160 maybe a counter argument. Counter argument. 1134 00:55:47,480 --> 00:55:50,400 The 1st is there's this saying like shirt sleeves to shirt 1135 00:55:50,400 --> 00:55:52,200 sleeves in three generations, right? 1136 00:55:52,520 --> 00:55:56,360 Which is to say that like this guy, he starts a factory, his 1137 00:55:56,360 --> 00:56:00,200 son inherits it, and his discipit grandson puts a fortune 1138 00:56:00,200 --> 00:56:03,080 up his nose and, you know, does drugs and, you know, basically 1139 00:56:03,080 --> 00:56:04,360 spends down the whole thing, right? 1140 00:56:04,880 --> 00:56:07,240 And this is like the resource curse concept where when people 1141 00:56:07,240 --> 00:56:11,360 get too wealthy or too powerful, these get extremely lazy. 1142 00:56:11,680 --> 00:56:14,240 They forget cause and effect, especially if they're two or 1143 00:56:14,240 --> 00:56:17,360 three generations out and they don't even know what hard work 1144 00:56:17,560 --> 00:56:19,760 resulted in that fortune in the 1st place. 1145 00:56:20,000 --> 00:56:21,040 And these blow the whole thing up. 1146 00:56:21,520 --> 00:56:23,480 And that's actually what's happening with the US right now. 1147 00:56:23,720 --> 00:56:26,360 Like in in many ways, I think the people who are currently 1148 00:56:26,360 --> 00:56:28,840 running the US government are not founders. 1149 00:56:28,840 --> 00:56:31,360 They're heirs. They've inherited the system 1150 00:56:31,360 --> 00:56:34,240 that, like better people set up decades and decades ago. 1151 00:56:34,240 --> 00:56:35,800 They don't even understand how it works. 1152 00:56:35,800 --> 00:56:37,160 It's like a factory they've inherited. 1153 00:56:37,160 --> 00:56:40,120 And they don't understand how it produces widgets or how it 1154 00:56:40,120 --> 00:56:41,880 maintains global order, global peace. 1155 00:56:42,240 --> 00:56:44,840 And they just think I'm big and powerful and they don't 1156 00:56:44,840 --> 00:56:48,120 understand why it exists. I think that's true, but it 1157 00:56:48,120 --> 00:56:52,400 doesn't matter because the, what the data shows is that over 1158 00:56:52,600 --> 00:56:55,960 multiple hundreds of years periods, the wealthy families 1159 00:56:55,960 --> 00:57:00,600 say the wealthy families and, and at like highest levels of 1160 00:57:00,600 --> 00:57:04,520 power, you know, you like, if you look at the history of the, 1161 00:57:04,880 --> 00:57:09,040 you know, English royal family, whatever, or Chinese emperors, 1162 00:57:09,040 --> 00:57:12,600 like they stay there for hundreds of years, you know, and 1163 00:57:12,600 --> 00:57:16,720 they create, they create feudal systems underneath themselves, 1164 00:57:16,720 --> 00:57:19,920 which are critical for establishing loyalty and all 1165 00:57:19,920 --> 00:57:23,760 that. That's the more natural state of 1166 00:57:23,760 --> 00:57:26,920 things that things fall into. Unless you can. 1167 00:57:29,680 --> 00:57:32,280 OK, so I'm on a counter argument set from an argument that I 1168 00:57:32,280 --> 00:57:34,920 think is interesting at least maybe to to, you know, maybe 1169 00:57:34,920 --> 00:57:37,440 you'll disagree. So if you have an heir or you, 1170 00:57:37,480 --> 00:57:40,640 you let's say you have a like a Genghis Khan, right? 1171 00:57:40,880 --> 00:57:42,720 They have two. Like they have a child, they've 1172 00:57:42,720 --> 00:57:45,200 got half their DNA, then another child, they've got a fourth, 1173 00:57:45,200 --> 00:57:46,800 then their child, they've got an eighth, right. 1174 00:57:47,360 --> 00:57:49,440 And most of the time people don't have an exponentially 1175 00:57:49,440 --> 00:57:53,280 increasing number of children. So that means that that fortune, 1176 00:57:53,280 --> 00:57:57,360 for example, would or, or whatever it is, it's very hard 1177 00:57:57,360 --> 00:58:01,600 to pass a fortune down many generations #1 and #2 is that 1178 00:58:01,600 --> 00:58:04,000 person almost doesn't even exist anymore because their genes are 1179 00:58:04,000 --> 00:58:06,520 being split up, diluted. Like, does the person even? 1180 00:58:06,760 --> 00:58:10,680 In what sense is somebody who's only 116th part of the same 1181 00:58:10,680 --> 00:58:11,920 family? Right. 1182 00:58:11,920 --> 00:58:15,240 I I think you're dramatically though over emphasizing the 1183 00:58:15,240 --> 00:58:20,600 importance of beings over context. 1184 00:58:21,280 --> 00:58:22,960 So like if. They're 4 generations down. 1185 00:58:22,960 --> 00:58:25,480 How is they've got a bunch of descendants, right? 1186 00:58:25,920 --> 00:58:29,000 The vast major of their descendants must like. 1187 00:58:29,000 --> 00:58:32,080 What does it even mean to say, a family across four or five 1188 00:58:32,080 --> 00:58:34,520 generations that family doesn't like? 1189 00:58:35,000 --> 00:58:37,120 Argues not how power is transferred, right? 1190 00:58:37,120 --> 00:58:41,120 So power is transferred by picking an air and then they 1191 00:58:41,120 --> 00:58:42,920 have an air and they have an air. 1192 00:58:43,200 --> 00:58:47,640 And then as soon as there's like a lack of a clear air, then you 1193 00:58:47,640 --> 00:58:50,560 get 100 years of war and then somebody wins. 1194 00:58:50,600 --> 00:58:54,240 And now they have another, you know, err, err, err, like they, 1195 00:58:54,240 --> 00:58:57,680 that's the thing. They, they, they generate the 1196 00:58:57,680 --> 00:59:04,160 system of hierarchical loyalty and, and, and they do, like you 1197 00:59:04,160 --> 00:59:06,760 can see, historically, the people do maintain it. 1198 00:59:06,920 --> 00:59:10,320 But but I made two points. First is most of their errors 1199 00:59:10,600 --> 00:59:13,600 are not inheriting that fortune, so the the majority of the 1200 00:59:13,600 --> 00:59:16,360 family or the descendants or whatever are not right because 1201 00:59:16,400 --> 00:59:20,080 it would be divided. And the second is even this 4th 1202 00:59:20,080 --> 00:59:23,880 or 5th generation guy is now like 132nd Genghis Khan or, or 1203 00:59:23,880 --> 00:59:26,240 what have you. And so they may just not have 1204 00:59:26,240 --> 00:59:29,680 the, the zeal or the energy of the original Genghis, right? 1205 00:59:29,880 --> 00:59:31,880 Say lose. And then there's a new guy who, 1206 00:59:31,880 --> 00:59:34,680 who takes over, right? So basically what I'm saying is 1207 00:59:35,000 --> 00:59:39,640 it's almost like there's there's a huge tax like a 50% tax every 1208 00:59:39,640 --> 00:59:42,920 generation that makes it very hard to keep concentrating the 1209 00:59:42,920 --> 00:59:46,360 same stuff in the same because the same people don't even exist 1210 00:59:46,920 --> 00:59:49,000 3 or 4 even with there's some inbred but you think. 1211 00:59:49,440 --> 00:59:52,440 You've got the premise wrong. The premise is that what better 1212 00:59:52,440 --> 00:59:54,600 there is the genes. And what I'm saying is no 1213 00:59:54,600 --> 00:59:56,760 biology. What matters is the power of the 1214 00:59:56,760 --> 01:00:00,080 positive feedback loop. Power gets begets power. 1215 01:00:00,320 --> 01:00:03,560 It doesn't matter if my I'm 5 generations away from Genghis 1216 01:00:03,560 --> 01:00:06,640 Khan. What matters is I'm the king of 1217 01:00:06,640 --> 01:00:08,960 England or I am the king of France, right? 1218 01:00:09,600 --> 01:00:12,480 But, you know, like, you've, like, you saw what happened in 1219 01:00:12,480 --> 01:00:15,200 China. Hundreds of years of terrible 1220 01:00:15,200 --> 01:00:19,000 emperors, opium addicts destroying the country, they 1221 01:00:19,000 --> 01:00:22,920 still maintain the power, right? And the country went from like 1222 01:00:23,280 --> 01:00:27,440 during the Tang dynasty, the, you know, the vast majority of 1223 01:00:27,440 --> 01:00:31,880 GDP in the world was in China, Cultural Center was was in 1224 01:00:31,880 --> 01:00:34,280 China, scientific center was in China. 1225 01:00:35,400 --> 01:00:41,880 And then through power concentration, the civilization. 1226 01:00:43,720 --> 01:00:44,280 Sure. So. 1227 01:00:44,320 --> 01:00:46,920 So we don't want that to happen I guess. 1228 01:00:47,200 --> 01:00:49,000 So, so let me agree with you on that. 1229 01:00:49,000 --> 01:00:51,080 And I do think that there needs to be alternatives and so on and 1230 01:00:51,080 --> 01:00:52,760 so forth. I'll just make one other point, 1231 01:00:52,760 --> 01:00:57,160 which is if that person is only 132nd or 164th Genghis Khan, 1232 01:00:57,440 --> 01:01:03,280 then there were 31 or 63 other people or families that rose. 1233 01:01:03,720 --> 01:01:06,600 So, so like the mobility is actually there. 1234 01:01:06,880 --> 01:01:10,080 If there's, it's if it's a sufficiently exogamous society, 1235 01:01:10,640 --> 01:01:13,840 then all these folks did rise to become rulers because their 1236 01:01:13,840 --> 01:01:15,240 bloodlines actually did get up there. 1237 01:01:15,240 --> 01:01:17,360 So basically, what I'm essentially what I'm not what 1238 01:01:17,360 --> 01:01:21,840 I'm agreeing with you is the title got passed down, but the 1239 01:01:21,840 --> 01:01:27,320 family doesn't even exist beyond 564, whatever number of 1240 01:01:27,320 --> 01:01:29,640 generations, right? The family just gets diluted 1241 01:01:29,640 --> 01:01:31,360 out. Does that make any sense? 1242 01:01:31,400 --> 01:01:33,120 Yeah, but but that's what I'm saying. 1243 01:01:33,120 --> 01:01:36,640 It doesn't matter, right? What matters is that you that 1244 01:01:36,640 --> 01:01:40,600 the, the positive feedback loop created a power and wealth 1245 01:01:40,600 --> 01:01:44,080 concentration that was maintained for hundreds of years 1246 01:01:45,000 --> 01:01:48,160 and most people in the country suffered, right? 1247 01:01:48,440 --> 01:01:51,160 And that's a thing that we want to avoid. 1248 01:01:51,160 --> 01:01:54,040 And it's incredibly difficult to avoid because that's the natural 1249 01:01:54,040 --> 01:01:55,880 state of things. It's positive. 1250 01:01:55,880 --> 01:01:59,920 Positive feedback loops. Maybe this is an empirical 1251 01:01:59,920 --> 01:02:01,200 question. We can look at different 1252 01:02:01,200 --> 01:02:05,520 trajectories, but I think it is difficult to maintain that power 1253 01:02:05,520 --> 01:02:07,880 and wealth concentration without zeal. 1254 01:02:08,160 --> 01:02:11,560 And that zeal, if it's not there, what, people get fat and 1255 01:02:11,560 --> 01:02:13,720 happy a few generations out? Like we've seen that. 1256 01:02:13,720 --> 01:02:16,480 I mean, maybe, maybe we're just think of different kinds of 1257 01:02:16,480 --> 01:02:19,880 examples, right? And for example, in tech, it's 1258 01:02:19,880 --> 01:02:22,200 almost entirely quote, new money, right? 1259 01:02:22,600 --> 01:02:26,960 And what I find is that people who've inherited fortunes are 1260 01:02:26,960 --> 01:02:29,840 just lethargic, right? They don't have that energy. 1261 01:02:29,840 --> 01:02:32,080 So we are seeing this Internet disruption, right? 1262 01:02:32,080 --> 01:02:33,840 This dark talent that's hungrier. 1263 01:02:33,840 --> 01:02:37,640 I would always invest in that. I'd always back that because 1264 01:02:37,640 --> 01:02:39,280 it's hungrier and it wants it, right? 1265 01:02:39,560 --> 01:02:42,280 Whereas so I'm only seeing anti compounding. 1266 01:02:42,280 --> 01:02:43,400 I'm only seeing I guess. Yeah, no. 1267 01:02:43,400 --> 01:02:46,520 And I agree with all that, but I'm trying to get you to think 1268 01:02:46,520 --> 01:02:48,160 about the end state. OK, go, go. 1269 01:02:48,600 --> 01:02:53,240 Like I agree with everything you're saying, right? 1270 01:02:53,240 --> 01:02:55,840 But what I'm trying to say is, OK, consider the positive 1271 01:02:55,840 --> 01:02:59,280 feedback loop here, right. You've with AI now you've got 1272 01:02:59,280 --> 01:03:02,920 the ability to create more power, you know, and, and more 1273 01:03:02,920 --> 01:03:10,800 wealth and we're more connected. Like we could literally end up 1274 01:03:11,160 --> 01:03:15,320 with a global dictator, and we could literally end up with a 1275 01:03:15,360 --> 01:03:19,600 permanent underclass representing 99.99% of the 1276 01:03:19,600 --> 01:03:20,720 world. So let's talk about how we 1277 01:03:20,720 --> 01:03:22,680 prevent that, right? Because I, because this is 1278 01:03:22,680 --> 01:03:23,760 something I do think about, right? 1279 01:03:24,120 --> 01:03:28,960 So my view is, and you may, may or disagree with this or not, or 1280 01:03:28,960 --> 01:03:33,480 is that we got people got more left than they expected. 1281 01:03:33,480 --> 01:03:35,800 Now they're getting more right than they expect more MAGA and 1282 01:03:35,800 --> 01:03:37,680 then they're going to get more China than they expected. 1283 01:03:37,920 --> 01:03:40,440 Like basically, I think what's going to happen is China's 1284 01:03:40,440 --> 01:03:43,920 rolling up a lot of alliances like the EU is doing deals with 1285 01:03:43,920 --> 01:03:46,080 China. All its historical rivals in 1286 01:03:46,080 --> 01:03:48,160 Southeast Asia are now just all folding in. 1287 01:03:48,440 --> 01:03:50,840 So the whole global economy is re centralizing around China. 1288 01:03:50,840 --> 01:03:54,760 And America has not just become isolationist, they've isolated 1289 01:03:54,760 --> 01:03:57,760 itself from the world. And the most punishing, they've 1290 01:03:57,760 --> 01:04:00,000 sort of self-imposed, the most punishing sanctions of all time 1291 01:04:00,000 --> 01:04:02,520 on themselves. Like a rogue state, North Korea, 1292 01:04:02,520 --> 01:04:06,360 Iran would face this kind of embargo, but it was like 1293 01:04:06,360 --> 01:04:07,800 self-imposed. So they think it's going to make 1294 01:04:07,800 --> 01:04:09,840 them strong. It's really kind of crazy stuff, 1295 01:04:09,840 --> 01:04:13,040 MAGA Maoism or whatever, right? So as a consequence, I think a 1296 01:04:13,040 --> 01:04:14,920 lot of power gets centralized in China and. 1297 01:04:15,200 --> 01:04:17,160 And along with that, interestingly, you're seeing a 1298 01:04:17,240 --> 01:04:22,960 huge this kind of cultural isolationism happening in 1299 01:04:22,960 --> 01:04:27,320 America also like quite difficult to undo potentially. 1300 01:04:27,360 --> 01:04:28,520 Extremely difficult because they. 1301 01:04:28,520 --> 01:04:31,640 Don't want to end up like Japan pre the Maiji Restoration. 1302 01:04:31,680 --> 01:04:33,720 You know, they thought they're powerful, they thought they're 1303 01:04:33,720 --> 01:04:36,000 strong, but actually they separate themselves in society 1304 01:04:36,000 --> 01:04:37,840 and become weak. That's a good outcome. 1305 01:04:37,840 --> 01:04:40,000 I actually I think it's quite. That's a good outcome. 1306 01:04:40,120 --> 01:04:43,320 Yeah, fair enough. I think, I mean, because that's 1307 01:04:43,320 --> 01:04:46,160 actually something where they give up the Empire, but they're 1308 01:04:46,160 --> 01:04:47,920 just like, you know, a country or. 1309 01:04:47,920 --> 01:04:49,440 What? Stay isolationistic, except 1310 01:04:49,560 --> 01:04:51,480 they've got nuclear weapons. Well, that's. 1311 01:04:51,680 --> 01:04:53,880 That's a problem. And The thing is, I think, you 1312 01:04:53,880 --> 01:04:56,280 know, there's a lot of people who'll say like, actually both 1313 01:04:56,280 --> 01:04:59,320 on the left and the right, who'll say we need to, you know, 1314 01:04:59,760 --> 01:05:02,880 a Republic, non empire or we need to shut down, you know, And 1315 01:05:02,880 --> 01:05:05,880 the problem is that first of all, maybe you'll agree with 1316 01:05:05,880 --> 01:05:07,520 these things. I'll give a view and then maybe 1317 01:05:07,520 --> 01:05:10,200 you shoot at it, right. I think the first thing at least 1318 01:05:10,200 --> 01:05:12,760 that I start with is American Empire is real. 1319 01:05:12,760 --> 01:05:16,360 And it was spectacular in the sense of arguably for all it's 1320 01:05:16,360 --> 01:05:17,960 false, one of the greatest of all time. 1321 01:05:18,400 --> 01:05:21,440 It did have capitalism, democracy, World Peace in many 1322 01:05:21,440 --> 01:05:24,360 ways, then lost its way, especially recently. 1323 01:05:24,360 --> 01:05:28,360 And and now you've got a very common kind of thing where the 1324 01:05:28,360 --> 01:05:30,960 folks on the left think, oh, the US is bombing lots of countries 1325 01:05:30,960 --> 01:05:34,400 that should stop doing that. Folks on the right think the US 1326 01:05:34,400 --> 01:05:37,400 is, is, is being exploited by all these foreigners abroad. 1327 01:05:37,400 --> 01:05:39,240 It's it's being cheated. We've deindustrialized. 1328 01:05:39,240 --> 01:05:40,680 We need to stop all that, bring all those jobs. 1329 01:05:40,720 --> 01:05:43,640 OK, fine. So this group thinks the US is 1330 01:05:43,760 --> 01:05:45,800 harming the world. This group thinks the world is 1331 01:05:45,880 --> 01:05:47,800 harming the US. Both of them think they want to 1332 01:05:47,800 --> 01:05:50,280 shut down the empire, bring the troops home, you know, and so 1333 01:05:50,320 --> 01:05:50,760 on. OK. 1334 01:05:51,080 --> 01:05:54,120 Remember also like it during that heyday of the 50s, you 1335 01:05:54,120 --> 01:05:57,320 know, the American top match world tax rate was like 80%, 1336 01:05:57,320 --> 01:05:58,760 like 90%. Ninety. 1337 01:05:58,760 --> 01:06:01,240 Yeah, there's like they're working very hard to avoid this 1338 01:06:02,680 --> 01:06:04,440 positive feedback. I loop I mentioned, you know, 1339 01:06:04,440 --> 01:06:09,240 redistricting the wealthiest. So on that point, just to talk 1340 01:06:09,280 --> 01:06:12,200 about that, the at that time though, power was completely 1341 01:06:12,200 --> 01:06:13,800 centralized in the US government, right? 1342 01:06:13,960 --> 01:06:17,560 So you almost have like a toothpaste tube squeezing, where 1343 01:06:17,800 --> 01:06:20,800 like if you avoid centralization on one axis, you often get it in 1344 01:06:20,800 --> 01:06:22,160 another kind of thing. It's kind. 1345 01:06:22,520 --> 01:06:26,240 Of right so cause. 'Cause the people who want power 1346 01:06:26,240 --> 01:06:29,080 will find ways to get it. Yeah, you can have total 1347 01:06:29,080 --> 01:06:30,800 centralization of government power or you can have 1348 01:06:30,800 --> 01:06:33,520 totalization of corporate power or maybe military power. 1349 01:06:33,520 --> 01:06:35,320 And so or you can have checks and balances. 1350 01:06:35,320 --> 01:06:38,440 And where where I think the world is going to go is a 1351 01:06:38,440 --> 01:06:41,080 billion person Chinese superstate. 1352 01:06:41,400 --> 01:06:44,800 And then eventually like 1000 million person network states, 1353 01:06:45,160 --> 01:06:47,600 like, and then I think India is going to be in the middle. 1354 01:06:47,600 --> 01:06:49,400 I think there's other countries are going to be in the middle 1355 01:06:49,400 --> 01:06:51,680 and so on and so forth. But that's that's where I think 1356 01:06:51,680 --> 01:06:54,040 things go by like 2040 or so, right. 1357 01:06:54,480 --> 01:06:57,760 And and so hopefully that gives I I'm not saying they're all a 1358 01:06:57,760 --> 01:06:59,720 million person network states. Some could might be bigger, some 1359 01:06:59,720 --> 01:07:02,960 might be smaller, but I but I do think that we'll have a lot of 1360 01:07:02,960 --> 01:07:05,840 choice of jurisdictions. I mean that that that would be 1361 01:07:05,840 --> 01:07:07,560 nice. That that's at least a hope. 1362 01:07:07,920 --> 01:07:08,480 Yeah. Go ahead. 1363 01:07:10,080 --> 01:07:12,520 I just got to say, keep thinking about the positive feedback 1364 01:07:12,520 --> 01:07:16,200 problem because I think it still has it, you know, it, it feels, 1365 01:07:16,360 --> 01:07:21,680 you know, rosy to the level of being like, well, that's that 1366 01:07:21,680 --> 01:07:25,320 seems not in line with how power dynamics. 1367 01:07:25,440 --> 01:07:28,280 I guess, I guess my biggest argument against that is 1368 01:07:28,560 --> 01:07:32,200 arbitrage because it's very difficult to get or let me give 1369 01:07:32,200 --> 01:07:35,600 a game theoretic argument, right, which is going back to 1370 01:07:35,600 --> 01:07:40,280 your sales example, right? If you have two people, you have 1371 01:07:40,480 --> 01:07:42,160 4 possible outcomes and a win, lose thing. 1372 01:07:42,160 --> 01:07:44,320 You can have win, win, win, lose, lose, win, lose, lose, 1373 01:07:44,320 --> 01:07:45,840 right? If you have three people, you 1374 01:07:45,840 --> 01:07:49,200 have two to the third, so 88 possible grams win, win, win, 1375 01:07:49,200 --> 01:07:51,920 win, win, lose, right? And you have K people, you have 1376 01:07:52,000 --> 01:07:55,960 two to the K possible outcomes where you know any of them can 1377 01:07:55,960 --> 01:07:58,920 win and n -, K can lose and so on for any value of N&K. 1378 01:07:58,920 --> 01:08:03,360 OK, So this is how I think about like managing a startup, right? 1379 01:08:03,760 --> 01:08:06,600 The startup, if you have 100 people, what you don't want is 1380 01:08:06,600 --> 01:08:09,960 political behavior where some subset of them loses and the 1381 01:08:09,960 --> 01:08:12,800 other subset wins. You want to have a single thing 1382 01:08:12,800 --> 01:08:15,560 which aligns everybody, and that's like equity, and that's 1383 01:08:15,560 --> 01:08:18,040 like the exit. So they all know if I work 1384 01:08:18,040 --> 01:08:20,960 together, we all get the maximum payoff when it's all win, win, 1385 01:08:20,960 --> 01:08:22,439 win, win across the board, right? 1386 01:08:23,560 --> 01:08:26,840 However, there's limits to the to to how large you can make 1387 01:08:26,840 --> 01:08:29,000 that, right? You might make that 100 people 1388 01:08:29,000 --> 01:08:30,880 might make that 1000. You might make it even a million 1389 01:08:30,880 --> 01:08:33,279 people like cryptocurrencies of getting it to 10s or hundreds of 1390 01:08:33,279 --> 01:08:35,960 millions of people, right? But I don't think you can get to 1391 01:08:35,960 --> 01:08:39,200 everybody. And the reason you can't get to 1392 01:08:39,200 --> 01:08:43,040 everybody is at some point there is an incentive to breakaway to 1393 01:08:43,040 --> 01:08:45,279 dis line is what I call network defect, right? 1394 01:08:45,800 --> 01:08:48,160 And so that is the counterweight to kind of I think what you're 1395 01:08:48,160 --> 01:08:49,640 saying about infinite compounding is? 1396 01:08:50,080 --> 01:08:52,880 Actually, if you're allowed to go ahead, if you're allowed to, 1397 01:08:52,880 --> 01:08:55,359 like, I mean, like, yeah, it's like, oh, you know, the people 1398 01:08:55,359 --> 01:08:58,160 in Wessex could have left or whatever. 1399 01:08:58,160 --> 01:09:00,160 It's like, no, they're in a futile state and they would have 1400 01:09:00,160 --> 01:09:02,120 got killed. And there's violence and, and 1401 01:09:02,120 --> 01:09:05,359 like if you had AI in the mix, then you can have like absolute 1402 01:09:05,359 --> 01:09:08,560 global surveillance and power and total control. 1403 01:09:08,640 --> 01:09:09,319 Right. So now OK. 1404 01:09:09,520 --> 01:09:12,040 So here it's it's fine. In theory, you could go and do 1405 01:09:12,040 --> 01:09:14,439 something else. In practice, if you even talk 1406 01:09:14,439 --> 01:09:16,720 about it, you get shot in the face. 1407 01:09:16,800 --> 01:09:20,200 Yeah, so, so the right so the practical way where I do agree 1408 01:09:20,200 --> 01:09:24,800 with you is the Chinese drone Armada, right? 1409 01:09:25,040 --> 01:09:29,880 Because they can manufacture huge numbers of robots and those 1410 01:09:29,880 --> 01:09:33,560 robots are they're no longer like human beings who can 1411 01:09:33,560 --> 01:09:35,880 defect, right, Because they can't defect. 1412 01:09:36,200 --> 01:09:38,200 All this concepts I've been talking about the game, the 1413 01:09:38,200 --> 01:09:40,279 principal agent problem goes away and just one guy pushing a 1414 01:09:40,279 --> 01:09:43,319 button and it's like a machine that just enacts our action 1415 01:09:43,319 --> 01:09:46,040 around the world, right. That is definitely something 1416 01:09:46,240 --> 01:09:50,040 which changes these dynamics. That is actually something where 1417 01:09:50,040 --> 01:09:52,920 you could have centralization and power for a long time, and 1418 01:09:52,920 --> 01:09:55,040 that is actually something we should think of as the most 1419 01:09:55,040 --> 01:09:58,000 important thing to build counterweights to going 35. 1420 01:09:58,000 --> 01:10:00,520 So I think your network states idea can hit that too. 1421 01:10:00,720 --> 01:10:04,840 So fast AI, you've got this Practical Deep Learning for 1422 01:10:04,840 --> 01:10:06,360 Coders Part 1. Part 2. 1423 01:10:06,360 --> 01:10:10,920 We've done a new course called How to Solve it with Code and 1424 01:10:10,920 --> 01:10:14,920 we've got a whole new platform for it which we basically he 1425 01:10:14,920 --> 01:10:18,400 beta tested it. We opened up sign ups for 24 1426 01:10:18,400 --> 01:10:22,240 hours kind of reasonably quietly, 1000 people signed up 1427 01:10:22,240 --> 01:10:24,120 within 24 hours. So then we closed it. 1428 01:10:24,440 --> 01:10:30,960 We did that and the reactions we got were amazing. 1429 01:10:30,960 --> 01:10:33,160 Like we've had hundreds of people come back and say, this 1430 01:10:33,160 --> 01:10:36,920 changed my life. I've got a new job. 1431 01:10:37,720 --> 01:10:40,040 Well, it's not open for everybody, but it's, it's solve 1432 01:10:40,040 --> 01:10:42,560 it dot fast dot AI. So we're trying to figure out 1433 01:10:42,560 --> 01:10:44,640 how to mail make the most of this because we've come up we've 1434 01:10:44,720 --> 01:10:48,280 we've, we've created something clearly extraordinary. 1435 01:10:48,400 --> 01:10:50,960 Basically the, the the fundamental idea. 1436 01:10:50,960 --> 01:10:52,960 I don't know how familiar with the Polya book, but it's 1437 01:10:52,960 --> 01:10:55,600 basically like. It's a bag of tricks for solving 1438 01:10:55,600 --> 01:10:57,400 math problems. Yeah, but it's more than a bag 1439 01:10:57,400 --> 01:10:59,000 of tricks. It's actually a fundamental 1440 01:10:59,000 --> 01:11:04,160 idea, which is that to do things iteratively, step by step. 1441 01:11:04,360 --> 01:11:07,960 And when you apply that idea to coding and then you bring AI 1442 01:11:07,960 --> 01:11:12,040 into the mix as well, you can. We, we've kind of come up with 1443 01:11:12,040 --> 01:11:16,720 this way of solving problems with code and AI where you're 1444 01:11:16,720 --> 01:11:19,680 constantly in control of the AI. You never get into that 1445 01:11:19,680 --> 01:11:21,960 situation where the AI is kind of controlling you. 1446 01:11:22,160 --> 01:11:23,000 Yeah. So, yeah. 1447 01:11:23,000 --> 01:11:25,680 So we've, we, we, like I said, we, we this was from months ago. 1448 01:11:25,680 --> 01:11:28,480 We haven't let anybody use it for months because we've been 1449 01:11:28,480 --> 01:11:30,600 running it and testing it. Yeah. 1450 01:11:30,600 --> 01:11:33,160 So it's a bit of a long story, but basically it's it's a whole 1451 01:11:33,160 --> 01:11:38,320 different way of thinking about problem solving, which is the 1452 01:11:38,400 --> 01:11:41,080 exact opposite of the whole vibe coding kind of. 1453 01:11:41,080 --> 01:11:43,360 It's like, let's think step by step for humans. 1454 01:11:43,600 --> 01:11:47,280 Yeah, let's think step by step for human plus AI together. 1455 01:11:47,280 --> 01:11:49,240 The AI sees all of your thinking. 1456 01:11:49,240 --> 01:11:52,000 You see the AI is thinking. You write code. 1457 01:11:52,000 --> 01:11:55,280 The AI write code. You're constantly focused on 1458 01:11:55,280 --> 01:12:00,640 learning and iteratively improving, you know, vibe 1459 01:12:00,640 --> 01:12:02,640 coding. It's just like one shot thing 1460 01:12:02,760 --> 01:12:05,680 where you don't learn anything. You get up more and more 1461 01:12:05,680 --> 01:12:08,080 technical debt. So it's actually it's 1462 01:12:08,080 --> 01:12:12,120 interesting like my Co founder Eric Grease has this lean 1463 01:12:12,120 --> 01:12:15,560 startup approach, which it turns out is really similar to the 1464 01:12:15,560 --> 01:12:17,720 Polya approach. Again, it's like highly 1465 01:12:17,720 --> 01:12:21,160 iterative learning based. So we're hoping that through 1466 01:12:21,160 --> 01:12:23,600 this Solvit course that we're going to eventually build 1467 01:12:24,360 --> 01:12:28,560 something like your startup engineering, but start using 1468 01:12:28,560 --> 01:12:32,920 this solvent approach and with with the help of AI to allow and 1469 01:12:32,920 --> 01:12:37,360 then to create like 1000 new startups from that course. 1470 01:12:37,360 --> 01:12:41,040 And then work with investors to give each of them, you know, a 1471 01:12:41,040 --> 01:12:43,920 start financially and maybe hopefully build the next 1472 01:12:43,920 --> 01:12:46,440 generation of of founders. Amazing. 1473 01:12:46,480 --> 01:12:48,320 And I think you know would be good. 1474 01:12:48,600 --> 01:12:50,640 I want to actually talk about the network school fellowship 1475 01:12:50,960 --> 01:12:53,360 with your fast AI folks because I think a lot of them could 1476 01:12:53,360 --> 01:12:54,800 benefit from applying or whatever. 1477 01:12:54,840 --> 01:12:57,480 So, OK, awesome. Thank you very much, Jeremy. 1478 01:12:57,880 --> 01:12:58,320 Thank you, Sir.