1 00:00:00,170 --> 00:00:02,930 S1: To zero day exploits and supply chain attacks keep you 2 00:00:02,930 --> 00:00:05,090 S1: up at night. Worry no more. You can harden your 3 00:00:05,090 --> 00:00:08,900 S1: security with threatlocker. Imagine taking a proactive deny by default 4 00:00:08,900 --> 00:00:13,160 S1: approach to cybersecurity, blocking every action, process and user unless 5 00:00:13,160 --> 00:00:16,430 S1: specifically authorized by your team. Threatlocker helps you do this 6 00:00:16,460 --> 00:00:19,130 S1: and provides a full audit of every action allowed or 7 00:00:19,130 --> 00:00:23,450 S1: blocked for risk management and compliance and onboarding and operation 8 00:00:23,450 --> 00:00:26,540 S1: is fully supported by their US based support team. Stop 9 00:00:26,540 --> 00:00:30,740 S1: the exploitation of trusted applications within your organization to keep 10 00:00:30,740 --> 00:00:35,269 S1: you running efficiently and secure. Protected from ransomware. Worldwide, companies 11 00:00:35,270 --> 00:00:38,630 S1: like JetBlue trust Threatlocker to secure their data and keep 12 00:00:38,630 --> 00:00:41,479 S1: their business operations flying high. To learn more about how 13 00:00:41,510 --> 00:00:47,089 S1: Threatlocker can mitigate unknown threats and ensure compliance for your organization, 14 00:00:47,090 --> 00:00:56,990 S1: visit threatlocker Kommerell. That's Threatlocker Kommerell. Welcome to Unsupervised Learning, 15 00:00:57,020 --> 00:01:00,020 S1: a security, AI, and meaning focused podcast that looks at 16 00:01:00,070 --> 00:01:03,100 S1: how best to thrive as humans in a post AI world. 17 00:01:03,520 --> 00:01:07,630 S1: It combines original ideas, analysis, and mental models to bring 18 00:01:07,630 --> 00:01:10,150 S1: not just the news, but why it matters and how 19 00:01:10,150 --> 00:01:17,980 S1: to respond. All right, welcome to unsupervised learning. This is 20 00:01:17,980 --> 00:01:21,699 S1: Daniel Miessler. All right. A few notes to start off. 21 00:01:21,700 --> 00:01:24,730 S1: All better from being sick. It was actually quite minor. 22 00:01:24,760 --> 00:01:29,290 S1: Definitely the most minor time I've been sick in recent history. 23 00:01:29,319 --> 00:01:31,720 S1: Would not even have known that I was sick if 24 00:01:31,720 --> 00:01:35,980 S1: it wasn't for testing. So yeah. Good, good that that's over. 25 00:01:35,980 --> 00:01:40,509 S1: We migrated fabric to go. So this is really major 26 00:01:40,510 --> 00:01:44,530 S1: massive shout out to Jonathan Dunn who helped massively with this. 27 00:01:44,560 --> 00:01:48,970 S1: And project still looks largely the same. But it's all 28 00:01:49,000 --> 00:01:52,750 S1: go now. And if you check this out look at 29 00:01:52,750 --> 00:01:56,110 S1: the installation. Now this is just insane. Watch this. You 30 00:01:56,110 --> 00:01:59,630 S1: just run this, that's all. You run okay? That's how 31 00:01:59,630 --> 00:02:02,660 S1: you install it and it just installs and it just works. 32 00:02:02,660 --> 00:02:06,980 S1: So no more Python, no more Python dependency. Hell, you 33 00:02:06,980 --> 00:02:09,380 S1: just run this and you run setup and it works. 34 00:02:09,410 --> 00:02:14,420 S1: Now you might have to add some environmental variables or 35 00:02:14,450 --> 00:02:17,450 S1: your go install might do it for you. It depends 36 00:02:17,450 --> 00:02:20,270 S1: how you install go, but check this out. If you're 37 00:02:20,270 --> 00:02:23,419 S1: going to migrate, you just do pip x uninstall fabric 38 00:02:23,419 --> 00:02:25,610 S1: and that will get rid of the Python version. And 39 00:02:25,610 --> 00:02:28,670 S1: then you want to check your bash RC and your 40 00:02:28,700 --> 00:02:32,000 S1: zshrc and basically clean those out. If you have any 41 00:02:32,030 --> 00:02:34,340 S1: aliases or anything like that in there. And then you 42 00:02:34,340 --> 00:02:36,380 S1: do this which is the same one as above, and 43 00:02:36,380 --> 00:02:38,180 S1: you do the setup. And again you could do the 44 00:02:38,180 --> 00:02:43,609 S1: environmental variables. But watch this. Let's say you haven't messed 45 00:02:43,610 --> 00:02:46,370 S1: with it in a while, which would be blasphemy. But 46 00:02:46,400 --> 00:02:48,440 S1: let's say you haven't messed with it in a while 47 00:02:48,440 --> 00:02:50,570 S1: and you come back, you went on vacation. Whatever, you 48 00:02:50,600 --> 00:02:52,910 S1: come back three weeks later and you want to install 49 00:02:52,910 --> 00:02:55,190 S1: this thing and you want to get the latest version, 50 00:02:55,190 --> 00:02:58,960 S1: how do you upgrade? You just run the go install again. 51 00:02:58,960 --> 00:03:01,149 S1: That's all you do. You just run the go install 52 00:03:01,150 --> 00:03:06,820 S1: again and you're completely updated and upgraded. And yeah, we 53 00:03:06,820 --> 00:03:10,630 S1: kind of kind of simplified up the the Readme here. 54 00:03:10,660 --> 00:03:14,679 S1: It should be pretty clean. And yeah, just really, really 55 00:03:14,680 --> 00:03:17,980 S1: happy with this. And it is so much faster. Yeah, 56 00:03:17,980 --> 00:03:21,550 S1: it is so much faster now. Yeah. I don't know 57 00:03:21,550 --> 00:03:25,840 S1: what um, I could show you here. All right. Yeah. 58 00:03:25,840 --> 00:03:29,500 S1: That's basically, uh, getting some aphorisms. So I think I 59 00:03:29,500 --> 00:03:31,360 S1: might have used one of these. Yeah, I used one 60 00:03:31,360 --> 00:03:35,110 S1: of these for the newsletter this week. Oh, actually. And 61 00:03:35,110 --> 00:03:37,840 S1: I'm recording the newsletter right now. Of course. So we'll 62 00:03:37,870 --> 00:03:40,390 S1: actually see this. This is the command that I used 63 00:03:40,390 --> 00:03:44,230 S1: to actually come up with some ideas for aphorisms. So 64 00:03:44,230 --> 00:03:48,280 S1: I just echo Red Queen concept from Ridley's book. So 65 00:03:48,280 --> 00:03:52,330 S1: I knew I wanted aphorisms around that. And um, look 66 00:03:52,360 --> 00:03:55,900 S1: at this. I get back some really, really cool different 67 00:03:55,930 --> 00:03:58,850 S1: aphorisms that I could potentially Potential use and depending on 68 00:03:58,850 --> 00:04:01,850 S1: the model, I was using one model and it only 69 00:04:01,850 --> 00:04:05,240 S1: gave me. Matt Ridley quotes. So that was pretty cool. 70 00:04:05,240 --> 00:04:08,150 S1: So really depends on the model that you use. But. 71 00:04:08,180 --> 00:04:10,610 S1: I basically use fabric all the time and the go 72 00:04:10,610 --> 00:04:15,440 S1: version is just unbelievably good. So that's that. Um, highly 73 00:04:15,440 --> 00:04:19,010 S1: recommend you check out this conversation between Peter Thiel and 74 00:04:19,010 --> 00:04:24,109 S1: Joe Rogan. Um, Joe's. An interesting character. He he had 75 00:04:24,110 --> 00:04:26,750 S1: Katt Williams on, and I watched as much as I 76 00:04:26,750 --> 00:04:29,150 S1: could of. That episode. And it was just like these 77 00:04:29,150 --> 00:04:34,490 S1: crazy conspiracy theories. And Katt Williams was quite serious, and 78 00:04:34,490 --> 00:04:37,400 S1: Joe was kind of pretending to be serious. But you 79 00:04:37,400 --> 00:04:39,380 S1: also don't know with Joe, like, what he believes and 80 00:04:39,380 --> 00:04:41,720 S1: what he doesn't believe sometimes when he's going off on 81 00:04:41,720 --> 00:04:44,180 S1: those things. But then you see him with someone like 82 00:04:44,180 --> 00:04:47,779 S1: Peter Thiel and he's way more balanced. He's like, he's 83 00:04:47,779 --> 00:04:50,420 S1: not coming up with a bunch of crazy stuff. He's 84 00:04:50,420 --> 00:04:53,990 S1: just having a normal conversation with someone he knows is smart. 85 00:04:53,990 --> 00:04:58,420 S1: And here's what's crazy. I actually thought his his description 86 00:04:58,420 --> 00:05:02,260 S1: of the future of humanity was better than Peter Thiel's. Now, 87 00:05:02,290 --> 00:05:06,039 S1: maybe just me and Joe are wrong, but Thiel seemed 88 00:05:06,040 --> 00:05:09,190 S1: unwilling to sort of go in the direction that the 89 00:05:09,220 --> 00:05:12,039 S1: Joe was taking it, which happens to be the exact 90 00:05:12,040 --> 00:05:14,830 S1: same direction that I always take it in, which is like, 91 00:05:14,860 --> 00:05:18,130 S1: this is inevitable. We have to merge with them. Basically, 92 00:05:18,130 --> 00:05:20,440 S1: it's like AI is going to get better and better, 93 00:05:20,440 --> 00:05:22,630 S1: and either we're going to merge with it or we're 94 00:05:22,630 --> 00:05:25,780 S1: going to be completely replaced and it would be better 95 00:05:25,779 --> 00:05:29,560 S1: to merge. And that just becomes the new us. And 96 00:05:29,560 --> 00:05:32,410 S1: that was Joe's perspective on it. And it's always been 97 00:05:32,410 --> 00:05:35,980 S1: my perspective on it. And I thought that was really interesting. 98 00:05:35,980 --> 00:05:39,940 S1: So basically you see this kind of elevated really smart 99 00:05:39,940 --> 00:05:43,180 S1: Joe on here. And then you kind of see the 100 00:05:43,180 --> 00:05:45,940 S1: opposite when he's talking to someone else. He kind of 101 00:05:45,970 --> 00:05:48,910 S1: just basically adapts to whoever he's talking to. If he's 102 00:05:48,910 --> 00:05:51,610 S1: talking to some fighter, he'll just talk about fights and 103 00:05:51,610 --> 00:05:53,890 S1: he'll just vibe out on that. If he's talking to 104 00:05:53,920 --> 00:05:57,110 S1: some hunter, he'll talk about hunting like he just blends 105 00:05:57,110 --> 00:06:01,099 S1: in with whoever he kind of is empathic with whoever 106 00:06:01,100 --> 00:06:03,710 S1: he's talking to. And you might want to say, well, 107 00:06:03,710 --> 00:06:07,340 S1: that's just a gimmick and he's just trash and blah, blah, blah. 108 00:06:07,370 --> 00:06:11,960 S1: He did start podcasting, basically. He did make podcasting a 109 00:06:11,960 --> 00:06:15,950 S1: thing by doing the thing that he's doing. So if 110 00:06:15,980 --> 00:06:18,349 S1: you don't like his politics or whatever, you still have 111 00:06:18,350 --> 00:06:22,250 S1: to objectively recognize that he is good at doing what 112 00:06:22,250 --> 00:06:25,849 S1: he does. And I believe it's because of this empathy thing. 113 00:06:25,850 --> 00:06:29,089 S1: So anyway, recommend checking it out. Most of the cool 114 00:06:29,089 --> 00:06:32,330 S1: thoughts were coming from teal one because he's super smart 115 00:06:32,330 --> 00:06:34,940 S1: and two because he was the guest. So that's Joe's 116 00:06:34,940 --> 00:06:37,940 S1: job is to extract that out of the guest. But 117 00:06:38,270 --> 00:06:43,910 S1: I'm I'm still trying to solve Peter Thiel because I 118 00:06:44,029 --> 00:06:47,450 S1: have always thought he was kind of like this right wing, uh, 119 00:06:47,450 --> 00:06:51,770 S1: crazy person. Uh, I really, really was mad at him 120 00:06:51,770 --> 00:06:56,440 S1: because he was supporting Trump back in what, 2020 or 2016? 121 00:06:56,440 --> 00:06:58,960 S1: I can't remember. And I was like, yeah, this is 122 00:06:58,990 --> 00:07:01,539 S1: I just don't like this person because he's supporting Trump 123 00:07:01,540 --> 00:07:03,310 S1: and how could you possibly do that? So I was 124 00:07:03,310 --> 00:07:06,580 S1: very triggered by that. He has since been like, yeah, 125 00:07:06,580 --> 00:07:09,430 S1: that was a horrible mistake. Like he's way more chaotic 126 00:07:09,460 --> 00:07:11,440 S1: than I thought he was going to be. And I 127 00:07:11,440 --> 00:07:15,010 S1: think he's supporting Trump again. But in a very sort 128 00:07:15,040 --> 00:07:17,620 S1: of distant kind of way, almost kind of like Elon 129 00:07:17,620 --> 00:07:20,530 S1: seems to be doing for the purposes of same with 130 00:07:20,530 --> 00:07:23,950 S1: Andreessen and same with Horowitz for the purpose of saying, 131 00:07:23,950 --> 00:07:27,280 S1: I think the other one is worse. You're not getting 132 00:07:27,310 --> 00:07:31,060 S1: like strong endorsements from any of these. I would call 133 00:07:31,060 --> 00:07:34,270 S1: centre right people. You're not getting strong endorsements of like, 134 00:07:34,300 --> 00:07:38,110 S1: oh yeah, Trump is great, right? And I don't mean 135 00:07:38,110 --> 00:07:41,170 S1: to be talking about politics here, but I'm actually not 136 00:07:41,170 --> 00:07:43,750 S1: talking about politics. I'm talking about what it means to 137 00:07:43,780 --> 00:07:50,230 S1: be an intellectual and have opinions that are counter to Orthodoxy. Right. 138 00:07:50,260 --> 00:07:53,050 S1: So essentially what I'm saying here is I'm trying to 139 00:07:53,050 --> 00:07:56,930 S1: figure teal out And the thing that I'm realizing about 140 00:07:56,960 --> 00:08:00,170 S1: teal is, and what I'm really liking about teal is 141 00:08:00,170 --> 00:08:03,350 S1: he and sort of dislike in some ways, it's like 142 00:08:03,350 --> 00:08:06,140 S1: he's not really committing to anything. He's not. I don't 143 00:08:06,140 --> 00:08:08,630 S1: see him saying, I think we should do this, which 144 00:08:08,630 --> 00:08:11,630 S1: I would really appreciate. Um, it's something that I try 145 00:08:11,660 --> 00:08:14,570 S1: to do. I try to do the objective thing, but 146 00:08:14,570 --> 00:08:17,300 S1: then come up with a direction to move in. I 147 00:08:17,300 --> 00:08:18,830 S1: feel like that's what you have to do if you 148 00:08:18,830 --> 00:08:21,650 S1: want to try to lead or to motivate or inspire. 149 00:08:21,650 --> 00:08:24,050 S1: But I feel like what teal does is just he's 150 00:08:24,050 --> 00:08:27,590 S1: just got this giant table full of hats, and he's 151 00:08:27,590 --> 00:08:30,260 S1: just always putting these different hats on. He's like, well, 152 00:08:30,260 --> 00:08:34,850 S1: let's let's steel man that let's beat that up. Let's 153 00:08:34,850 --> 00:08:37,010 S1: steel man that, let's beat that up. So he's just 154 00:08:37,010 --> 00:08:40,069 S1: like constantly looking at pros and cons of different ways 155 00:08:40,070 --> 00:08:43,400 S1: of thinking. And I think he has some general tendencies right. 156 00:08:43,429 --> 00:08:47,210 S1: He's obviously generally libertarian but he'll just straight up tell 157 00:08:47,210 --> 00:08:49,819 S1: you where he's not libertarian. And so what I like 158 00:08:49,850 --> 00:08:52,970 S1: about him is he's not fitting into any sort of 159 00:08:52,970 --> 00:08:56,140 S1: right wing bolt. Right. A lot of right people, right 160 00:08:56,140 --> 00:08:58,540 S1: wing people probably think he is. But if you actually 161 00:08:58,540 --> 00:09:01,750 S1: listen to him, he's not. He's actually going down his 162 00:09:01,750 --> 00:09:06,100 S1: own path from first principles. From all this reading, if 163 00:09:06,100 --> 00:09:09,490 S1: you watch his conversation with Tyler Cowen. Oh my God, 164 00:09:09,520 --> 00:09:11,980 S1: I mean, it's like the Joe Rogan one times 1000 165 00:09:11,980 --> 00:09:14,979 S1: because they are just riffing on all this history and 166 00:09:14,980 --> 00:09:17,469 S1: all this different stuff that they're talking about, and they're 167 00:09:17,470 --> 00:09:20,020 S1: just jumping from one topic to another. And so he's 168 00:09:20,020 --> 00:09:23,590 S1: using all that knowledge to walk his own path using 169 00:09:23,620 --> 00:09:27,340 S1: first principles. And sometimes that takes him like center right. 170 00:09:27,370 --> 00:09:30,520 S1: Sometimes that takes him center left. I imagine sometimes he 171 00:09:30,520 --> 00:09:33,970 S1: goes more extreme left and more extreme right. But I 172 00:09:33,970 --> 00:09:36,819 S1: don't think he honestly cares about left and right. I 173 00:09:36,820 --> 00:09:40,390 S1: think what he cares about is first principles and and 174 00:09:40,390 --> 00:09:43,599 S1: studying and learning and trying to just constantly improve his 175 00:09:43,600 --> 00:09:46,809 S1: models so that he can have all these different hats. 176 00:09:46,840 --> 00:09:49,390 S1: Because ultimately, I think he's an investor, and I think 177 00:09:49,420 --> 00:09:51,790 S1: ultimately his whole game is trying to figure out how 178 00:09:51,790 --> 00:09:55,630 S1: to predict the future to some degree, some limited degree. Oh, 179 00:09:55,630 --> 00:09:58,990 S1: that's the other thing. He's extremely humble. He's extremely cautious. 180 00:09:59,020 --> 00:10:00,580 S1: They'll be like, hey, what do you think of this? 181 00:10:00,610 --> 00:10:03,459 S1: What's the future going to be? He's like, yeah, nobody knows. 182 00:10:03,460 --> 00:10:06,460 S1: It's a dumb question. So it's like the opposite of 183 00:10:06,460 --> 00:10:09,670 S1: the right. Okay. You get a typical right wing pundit. 184 00:10:09,670 --> 00:10:12,670 S1: So let me give you a perfect example. Peter Thiel 185 00:10:12,670 --> 00:10:17,830 S1: is the absolute opposite from Tucker Carlson. Tucker Carlson is 186 00:10:17,830 --> 00:10:22,059 S1: extremely sure about things and extremely wrong. And Peter Thiel 187 00:10:22,059 --> 00:10:25,960 S1: is extremely unsure about things and tends to be very right. 188 00:10:25,960 --> 00:10:29,560 S1: And he's very cautious and he's just constantly consuming and learning. 189 00:10:29,559 --> 00:10:32,890 S1: And it's just yeah, so I'm still trying to lock 190 00:10:32,890 --> 00:10:35,439 S1: in on Peter Thiel. I'm still trying to look for 191 00:10:35,440 --> 00:10:38,800 S1: skeletons or look for biases, but I feel like he's 192 00:10:38,800 --> 00:10:42,100 S1: a pretty good model of how to learn about the world, 193 00:10:42,130 --> 00:10:43,809 S1: as far as I can tell. So far, I've got 194 00:10:43,840 --> 00:10:47,320 S1: like a 75% read. I would love to see if 195 00:10:47,320 --> 00:10:50,710 S1: anyone has a different read, or has knowledge that can 196 00:10:50,740 --> 00:10:54,370 S1: sort of shape my model of him based on this, 197 00:10:54,370 --> 00:10:56,559 S1: and I bought one of those mini libraries to put 198 00:10:56,559 --> 00:10:58,300 S1: up in my neighborhood. I don't know if you've ever 199 00:10:58,300 --> 00:11:02,140 S1: seen these. I've seen them all over, definitely in Palo Alto. 200 00:11:02,140 --> 00:11:04,600 S1: So if you drive through one of these neighborhoods in 201 00:11:04,600 --> 00:11:08,110 S1: Palo Alto, like kind of near Stanford, um, not Atherton, 202 00:11:08,110 --> 00:11:11,110 S1: but one of the nicer neighborhoods in Palo Alto, you'll see, 203 00:11:11,140 --> 00:11:13,780 S1: like these little it looks like a birdhouse, kind of. 204 00:11:13,809 --> 00:11:16,870 S1: But it's like on the curb, like right against the street. 205 00:11:16,870 --> 00:11:19,750 S1: And it says like mini library or community library or 206 00:11:19,750 --> 00:11:22,510 S1: whatever on it. And it's just like this little birdhouse thing, 207 00:11:22,510 --> 00:11:24,590 S1: except for when you open it up. It's like 1 208 00:11:24,590 --> 00:11:28,090 S1: or 2 shelves of books. And the idea is you 209 00:11:28,090 --> 00:11:30,309 S1: take a book, you just walk away. You just walk 210 00:11:30,309 --> 00:11:32,319 S1: away with it. Maybe you never bring it back, but 211 00:11:32,320 --> 00:11:35,890 S1: the idea is community karma. So yes, you do bring 212 00:11:35,890 --> 00:11:37,750 S1: it back. Or maybe you bring back an extra book. 213 00:11:37,750 --> 00:11:40,179 S1: So it's just like this. It's a combination of community 214 00:11:40,179 --> 00:11:44,829 S1: with books. Love it. Obviously. All right. So my work 215 00:11:44,860 --> 00:11:48,309 S1: got this new piece on the four components of what's 216 00:11:48,309 --> 00:11:52,350 S1: going to make models. When I models win and it's 217 00:11:52,350 --> 00:11:55,980 S1: basically I'll go into it here. It's basically the model itself, 218 00:11:55,980 --> 00:12:00,210 S1: post-training internal tooling and then agents. And I'm actually going 219 00:12:00,240 --> 00:12:02,910 S1: to add some extra stuff here. But these are the four. 220 00:12:02,940 --> 00:12:05,100 S1: And then the summary is just the four of those. 221 00:12:05,100 --> 00:12:08,640 S1: So we should start thinking about top models as model 222 00:12:08,640 --> 00:12:11,820 S1: ecosystems instead of just models. Because it's not just the 223 00:12:11,820 --> 00:12:14,910 S1: neural net weights doing the work. There are four main components. 224 00:12:14,910 --> 00:12:18,480 S1: And the company that wins the AI model wars will 225 00:12:18,480 --> 00:12:20,790 S1: need to excel at all four of these. Not just 226 00:12:20,790 --> 00:12:23,160 S1: spending lots of money to have the neural net with 227 00:12:23,160 --> 00:12:26,130 S1: the most parameters. So that's that piece. Next one here 228 00:12:26,130 --> 00:12:29,160 S1: is the link between free will and LLM denial. That 229 00:12:29,160 --> 00:12:32,820 S1: one's a bit deep and esoteric. And if you don't 230 00:12:32,820 --> 00:12:35,220 S1: like free will it won't interest you. So I'm going 231 00:12:35,250 --> 00:12:38,280 S1: to skip that one. Okay. Security. Microsoft just released a 232 00:12:38,280 --> 00:12:42,929 S1: bunch of patches, uh, 90 security flaws, including ten zero days, 233 00:12:42,929 --> 00:12:47,010 S1: six of which being actively exploited. I think the biggest one. 234 00:12:47,010 --> 00:12:51,100 S1: If I'm not incorrect about this is Wasn't there a 235 00:12:51,100 --> 00:12:54,670 S1: really nasty IPv6 one? Just pull this up real quick. 236 00:12:54,700 --> 00:13:01,870 S1: IPv6 Microsoft repeatedly sent IPv6 packets include specially crafted packets 237 00:13:01,870 --> 00:13:07,510 S1: to windows machine enable remote code execution. Yeah, so pretty nasty. 238 00:13:07,510 --> 00:13:10,360 S1: And I think it's before the firewall. So somebody was 239 00:13:10,360 --> 00:13:13,630 S1: saying it might have been Marcus Hutchins saying this, who 240 00:13:13,630 --> 00:13:15,699 S1: he got to meet in person for the first time 241 00:13:16,000 --> 00:13:19,750 S1: after talking to him for a number of years just online, 242 00:13:19,750 --> 00:13:22,630 S1: but got to meet him at this crater meetup that 243 00:13:22,630 --> 00:13:25,809 S1: we do in Vegas. Probably got Covid there. Lots of 244 00:13:25,809 --> 00:13:28,030 S1: people probably gave me Covid there. Lots of people probably 245 00:13:28,030 --> 00:13:30,640 S1: got Covid there. Anyway. Got to meet him in person 246 00:13:30,640 --> 00:13:33,940 S1: there in Vegas. That was cool. But I think he 247 00:13:33,940 --> 00:13:37,960 S1: might have been saying somewhere that basically the IPv6 packets 248 00:13:37,990 --> 00:13:40,720 S1: come in and don't even make it to the firewall. 249 00:13:40,720 --> 00:13:44,980 S1: It just goes straight through and gets parsed and produces 250 00:13:44,980 --> 00:13:49,330 S1: the ability to create RC. So that's great. Uh Russian 251 00:13:49,350 --> 00:13:54,420 S1: cyber spies from the FSB. Cold War. Trail. Cold war's trail. 252 00:13:54,420 --> 00:13:57,870 S1: Cold wastrel. Who knows how to say that? Been running 253 00:13:57,870 --> 00:14:01,620 S1: a massive phishing campaign dubbed River of fish, targeting U.S. 254 00:14:01,620 --> 00:14:07,170 S1: and European entities since 2022, going after high risk individuals, NGOs, 255 00:14:07,200 --> 00:14:10,980 S1: media outlets, government people. And the Pentagon is talking about 256 00:14:10,980 --> 00:14:13,890 S1: their plan to flood the Taiwan Strait with thousands of 257 00:14:13,890 --> 00:14:17,040 S1: drones in the event of a Chinese invasion. So this guy, 258 00:14:17,070 --> 00:14:21,750 S1: Admiral Sam Samuel Paparo, described the strategy as creating an 259 00:14:21,750 --> 00:14:26,490 S1: unmanned hellscape to delay Chinese forces. Why are we talking 260 00:14:26,490 --> 00:14:29,730 S1: about this? Unless this is part of the strategy to 261 00:14:29,760 --> 00:14:31,739 S1: make them think we're doing this, and then they try 262 00:14:31,770 --> 00:14:34,740 S1: to adjust for that. But that's not actually what we're doing. Like, 263 00:14:34,770 --> 00:14:37,830 S1: are we playing games here? Like what exactly. Why are 264 00:14:37,830 --> 00:14:41,370 S1: we talking about this? Evidently, it's not meant to be 265 00:14:41,370 --> 00:14:45,930 S1: useful as a surprise because it's no longer a surprise. Whatever. 266 00:14:45,960 --> 00:14:47,910 S1: That'd be cool if it was like, yeah, we're not 267 00:14:47,910 --> 00:14:49,720 S1: really going to do that. We're just going to make 268 00:14:49,720 --> 00:14:52,990 S1: them do a bunch of anti-drone stuff. And actually our 269 00:14:52,990 --> 00:14:55,900 S1: strategy is going to be completely different. Whatever. We'd like 270 00:14:55,900 --> 00:14:58,990 S1: to know more. Someone published a timeline of his research 271 00:14:59,020 --> 00:15:04,600 S1: on offensive AI agents, three distinct types of AI offensive systems. 272 00:15:04,600 --> 00:15:10,570 S1: SolarWinds patched a critical Deserialization vulnerability 9.8 on the Richter scale. 273 00:15:10,570 --> 00:15:13,480 S1: Iranian banks have been hit by a massive cyber attack, 274 00:15:13,480 --> 00:15:16,900 S1: probably tied to Israel. I'm just guessing Trump shared a 275 00:15:16,900 --> 00:15:20,530 S1: fake image. Oh man, I didn't put the link. I 276 00:15:20,530 --> 00:15:22,930 S1: just got an email about this. Didn't put the link. 277 00:15:22,960 --> 00:15:26,860 S1: Oh man. Whatever fake image of Harris speaking at a 278 00:15:26,860 --> 00:15:29,710 S1: Communist event. All right, let me let me find this thing. 279 00:15:29,740 --> 00:15:35,920 S1: Trump share. Yeah, look at that. Look at that. I mean, 280 00:15:35,950 --> 00:15:38,200 S1: does anyone think she would actually go to an event 281 00:15:38,200 --> 00:15:40,869 S1: like that? Well, first of all, yeah, this this is 282 00:15:40,870 --> 00:15:43,600 S1: what I said actually in the piece. Right. Um, one 283 00:15:43,600 --> 00:15:46,660 S1: lots of people will believe it's real. And two, current 284 00:15:46,690 --> 00:15:49,650 S1: tech can already make more believable ones than this. So 285 00:15:49,650 --> 00:15:52,380 S1: this isn't the actual problem. The actual problem is when 286 00:15:52,380 --> 00:15:55,710 S1: it gets much worse. Oh, here's the other one that 287 00:15:55,710 --> 00:16:01,560 S1: they were sharing. Watch this one. Trump. Share. Taylor Swift 288 00:16:01,590 --> 00:16:05,100 S1: should be enough, right? No links. Yeah. Look at this. 289 00:16:05,130 --> 00:16:07,680 S1: Look at this. So we've been talking about for years. 290 00:16:07,710 --> 00:16:09,660 S1: I went looking for this post where I first started 291 00:16:09,690 --> 00:16:12,480 S1: talking about this. I couldn't find it. So. Beehive, fix 292 00:16:12,480 --> 00:16:15,870 S1: your search, please. Uh, couldn't find it with Google either, so. 293 00:16:16,140 --> 00:16:19,800 S1: Kind of annoying. Anyway, I wrote about basically deep fakes 294 00:16:19,800 --> 00:16:22,109 S1: becoming an issue because we're not going to be able 295 00:16:22,110 --> 00:16:24,960 S1: to believe things. It's so obvious now. It's not even 296 00:16:24,960 --> 00:16:28,020 S1: a good post, but it was very early. This was 297 00:16:28,020 --> 00:16:31,650 S1: like two years ago. Um, I was talking about this. Um, 298 00:16:31,650 --> 00:16:36,870 S1: but anyway, everyone knows now, uh, and here is the actual. 299 00:16:36,930 --> 00:16:40,530 S1: It's just now starting to matter. Okay, look at this picture. 300 00:16:40,530 --> 00:16:43,800 S1: The top right one that looks like an actual, like event. 301 00:16:43,800 --> 00:16:46,080 S1: This person's got his phone. He's got his foot sticking 302 00:16:46,080 --> 00:16:50,380 S1: out That looks super real. Super real Swifties for Trump. 303 00:16:50,410 --> 00:16:53,080 S1: Now you don't be funny is if that was if 304 00:16:53,080 --> 00:16:56,620 S1: that was actually real like some. Okay the bottom left 305 00:16:56,620 --> 00:17:00,520 S1: definitely not real. Okay. They're probably none of these are 306 00:17:00,520 --> 00:17:03,250 S1: probably real. Just to be very clear about, you know, 307 00:17:03,280 --> 00:17:07,240 S1: disinformation or whatever. But, um, the point is you can 308 00:17:07,240 --> 00:17:09,490 S1: make these. Oh, yeah. This bottom right one also looks 309 00:17:09,490 --> 00:17:12,940 S1: super real. The bottom left one does not because that 310 00:17:12,940 --> 00:17:17,560 S1: Swifties for Trump looks like two clear like the white 311 00:17:17,590 --> 00:17:20,679 S1: of the text just looks a little bit too artificial. 312 00:17:20,680 --> 00:17:24,220 S1: But this one Swifties for Trump. It's got shadow on it. 313 00:17:24,220 --> 00:17:27,160 S1: The T is hidden because it's on the other side 314 00:17:27,160 --> 00:17:30,010 S1: of her body, so that one looks real. It's actually 315 00:17:30,010 --> 00:17:33,640 S1: the same person as this. Honestly, these two look really, 316 00:17:33,640 --> 00:17:37,480 S1: really good that that would trick basically anybody, including me 317 00:17:37,510 --> 00:17:40,660 S1: right now. So the point is, these are getting so 318 00:17:40,660 --> 00:17:43,929 S1: good that you can't tell the difference between reality and 319 00:17:43,960 --> 00:17:47,820 S1: doesn't really matter for this. Right now. This one maybe 320 00:17:47,820 --> 00:17:50,910 S1: kind of does. The question is what happens when it's 321 00:17:50,910 --> 00:17:54,870 S1: a thing that really matters? Like did a crime actually occur? Um, 322 00:17:54,869 --> 00:17:59,580 S1: who is the assailant? Evidence of something bad happening. Where 323 00:17:59,609 --> 00:18:03,240 S1: where we're going to use pictures or video as part 324 00:18:03,240 --> 00:18:05,580 S1: of the evidence for that crime. That's when it starts 325 00:18:05,580 --> 00:18:09,600 S1: to really, really matter. And also for politics as well. Um, 326 00:18:09,600 --> 00:18:14,129 S1: the worst possible one is, is like the absolute worst 327 00:18:14,130 --> 00:18:19,980 S1: one is where you have, um, a video of Trump saying, 328 00:18:20,010 --> 00:18:22,439 S1: you know, it looks like we're going to have to 329 00:18:22,440 --> 00:18:26,820 S1: attack Russia, pre-emptive nuclear strike. And it's very believable. And 330 00:18:26,820 --> 00:18:29,880 S1: it freaks somebody out and that person freaks out Putin 331 00:18:29,880 --> 00:18:32,640 S1: and something goes off or something similar to that. Maybe 332 00:18:32,640 --> 00:18:35,700 S1: it's not a nuclear war, but something where there's an 333 00:18:35,700 --> 00:18:39,090 S1: actual threat being made and it causes the other person 334 00:18:39,090 --> 00:18:43,440 S1: to respond. So it's a trigger point for kinetic action 335 00:18:43,440 --> 00:18:48,350 S1: that that's pretty serious and we have the tech now 336 00:18:48,470 --> 00:18:51,469 S1: to be able to do that. Not everyone could do it, 337 00:18:51,470 --> 00:18:54,800 S1: but it's getting pretty close, and it's much easier to 338 00:18:54,800 --> 00:18:57,170 S1: do with images than it is for video. But the 339 00:18:57,170 --> 00:19:00,410 S1: point is, this is getting very, very real. No longer 340 00:19:00,410 --> 00:19:04,580 S1: theoretical Iranian hacking group Abt 42 has targeted both Trump 341 00:19:04,580 --> 00:19:10,460 S1: and Biden campaigns, according to Google's Tag group. They look 342 00:19:10,490 --> 00:19:14,479 S1: like they work for Iran's Revolutionary Guard Corps, which is 343 00:19:14,510 --> 00:19:16,820 S1: like their elite group. And it looks like they targeted 344 00:19:16,820 --> 00:19:21,830 S1: both campaigns. But only Trump's campaign had sensitive files leaked. 345 00:19:22,130 --> 00:19:24,920 S1: Kind of interesting. I mean, maybe there was nothing there 346 00:19:24,920 --> 00:19:27,949 S1: on the Democratic side. Maybe, I don't know, why are 347 00:19:27,950 --> 00:19:33,619 S1: they why are they anti-Trump? Um, oh, maybe. Maybe because 348 00:19:33,650 --> 00:19:38,270 S1: Trump is about to massively support Israel and massively crush Iran. 349 00:19:38,390 --> 00:19:41,360 S1: That would be a good explanation. Trump corroborated this by 350 00:19:41,390 --> 00:19:44,910 S1: pointing the finger at Iran for hacking his campaign, praising 351 00:19:44,910 --> 00:19:48,120 S1: the FBI's investigation into the breach. This is one thing 352 00:19:48,119 --> 00:19:51,300 S1: that I like decently about Trump when he does it, 353 00:19:51,300 --> 00:19:53,700 S1: and it's only because the bar is so low. But 354 00:19:53,700 --> 00:19:56,400 S1: if he gets on a phone call with the FBI 355 00:19:56,400 --> 00:19:59,190 S1: and the FBI is like, yes sir, no sir, blah, blah, blah, 356 00:19:59,220 --> 00:20:01,409 S1: you know, we did this or we did that. If 357 00:20:01,410 --> 00:20:03,990 S1: he just got done railing against the FBI for like 358 00:20:03,990 --> 00:20:06,629 S1: the last six months and he has that call, he'll 359 00:20:06,630 --> 00:20:08,310 S1: go and talk to the press and be like, yeah, 360 00:20:08,310 --> 00:20:11,040 S1: I just had a great call with the FBI. Very professional, 361 00:20:11,040 --> 00:20:13,890 S1: very cool. It's like the thing that happened most recently 362 00:20:13,890 --> 00:20:16,350 S1: is like his view of reality, but at least he's 363 00:20:16,350 --> 00:20:19,440 S1: willing to, like, be positive if a positive thing happened. 364 00:20:19,470 --> 00:20:21,449 S1: I mean, I can't believe I'm saying that as a 365 00:20:21,450 --> 00:20:26,400 S1: positive thing, since that's called normal. China linked cyber spies 366 00:20:26,400 --> 00:20:30,780 S1: have infected dozens of Russian government and IT sector computers 367 00:20:30,780 --> 00:20:35,040 S1: with backdoors and trojans since late July, according to Kaspersky. So, 368 00:20:35,070 --> 00:20:38,610 S1: China v Russia scammers are targeting young Chinese job seekers 369 00:20:38,609 --> 00:20:42,720 S1: in a tough economy. AI and tech grok chatbot now 370 00:20:42,720 --> 00:20:45,830 S1: lets users create images from text prompts and publish them 371 00:20:45,830 --> 00:20:49,490 S1: to X, leading to chaotic results like Barack Obama doing 372 00:20:49,490 --> 00:20:52,640 S1: cocaine and Donald Trump in a Nazi uniform. Yeah, the 373 00:20:52,640 --> 00:20:56,389 S1: one that I saw. Yeah, I wrote this right here. Um, 374 00:20:56,390 --> 00:20:59,450 S1: the one I saw was Trump basically had his arm 375 00:20:59,450 --> 00:21:02,870 S1: around Ellen, and Ellen was pregnant. And Ellen posted that 376 00:21:02,869 --> 00:21:04,640 S1: and he was like, live by the sword, die by 377 00:21:04,640 --> 00:21:09,350 S1: the sword. Yeah. Uh, okay. Got someone who's on a 378 00:21:09,350 --> 00:21:12,800 S1: mission to make you love reading again. So he's using 379 00:21:12,830 --> 00:21:17,090 S1: AI tools like ChatGPT to recommend books, understand deeper themes, 380 00:21:17,090 --> 00:21:22,580 S1: and novels like Hermann Hesse's Siddhartha, and create actionable strategies 381 00:21:22,580 --> 00:21:26,780 S1: from business books like Alex Hormozi $100 million Offers. I've 382 00:21:26,780 --> 00:21:29,660 S1: already got that built into fabric. Um, I've got one 383 00:21:29,660 --> 00:21:33,860 S1: called create $100 million offer. You give it any product, 384 00:21:33,859 --> 00:21:37,790 S1: concept or idea and it'll build you a hormozi offer 385 00:21:37,790 --> 00:21:40,580 S1: based on it. Comedians are increasingly using AI to help 386 00:21:40,580 --> 00:21:44,609 S1: write jokes and brainstorm ideas with mixed results. I think 387 00:21:44,609 --> 00:21:48,240 S1: this is as big or bigger, not bigger as big, 388 00:21:48,240 --> 00:21:50,820 S1: maybe as the Turing test in terms of the importance 389 00:21:50,820 --> 00:21:54,510 S1: of AI progress. Because if I can write a full 390 00:21:54,510 --> 00:21:59,460 S1: comedy set and make humans laugh, that is massively huge. 391 00:21:59,460 --> 00:22:03,270 S1: Even better if it can be deepfaked performed right. So 392 00:22:03,270 --> 00:22:06,900 S1: it's actually AI writing it and AI performing it, and 393 00:22:06,900 --> 00:22:10,920 S1: all the human did was tell an agent to write 394 00:22:10,950 --> 00:22:15,659 S1: a comedy set around a topic. Okay, that is this. 395 00:22:15,660 --> 00:22:20,370 S1: This is an unbelievable metric, okay, a milestone metric. So basically, 396 00:22:20,400 --> 00:22:24,300 S1: here's the test. Can I give an agent or any model? 397 00:22:24,300 --> 00:22:28,020 S1: Let's say, for example, I hand this to I write 398 00:22:28,050 --> 00:22:33,330 S1: a comedy sketch skit or set around, um, good coffee 399 00:22:33,359 --> 00:22:37,679 S1: versus bad coffee or how baristas don't care about you 400 00:22:37,680 --> 00:22:40,889 S1: or how customer services people don't care about you. Do 401 00:22:40,920 --> 00:22:44,030 S1: write a 20 minute set just about this topic. Okay. 402 00:22:44,060 --> 00:22:46,700 S1: So now I want to see on a screen either 403 00:22:46,700 --> 00:22:50,030 S1: a stand up view or a face view. Yeah. A 404 00:22:50,030 --> 00:22:51,800 S1: stand up view would be better. Actually, I want to 405 00:22:51,800 --> 00:22:53,689 S1: see a stage. I want to see a microphone. I 406 00:22:53,690 --> 00:22:56,750 S1: want to see a person standing there talking. And I 407 00:22:56,750 --> 00:22:58,490 S1: want to see them moving. I want to see them 408 00:22:58,490 --> 00:23:01,550 S1: moving their hands. I want to see them gesturing. A full, 409 00:23:01,550 --> 00:23:05,060 S1: complete deepfake of the content that they wrote. And this 410 00:23:05,060 --> 00:23:07,160 S1: has to be funny. Okay. And it doesn't even have 411 00:23:07,160 --> 00:23:09,139 S1: to be, like, the funniest thing. It doesn't have to 412 00:23:09,140 --> 00:23:12,770 S1: be Dave Chappelle. It has to be decently funny. Like 413 00:23:12,770 --> 00:23:16,160 S1: good enough to get gigs on a comedy tour. And 414 00:23:16,160 --> 00:23:18,350 S1: if you've been on a comedy, tours like The Bar 415 00:23:18,350 --> 00:23:21,980 S1: is not super high for the average comic that I 416 00:23:21,980 --> 00:23:24,230 S1: don't know what I skeptics are going to say to that. 417 00:23:24,230 --> 00:23:26,149 S1: And I would say I'm going to I'm going to 418 00:23:26,150 --> 00:23:29,000 S1: call it right now. We are in the middle of 2024. 419 00:23:29,030 --> 00:23:34,580 S1: I'm going to call that as 2025, 2026 at the latest. 420 00:23:34,580 --> 00:23:40,129 S1: Full deepfake integration 2026 at the latest. So both 2026, 421 00:23:40,130 --> 00:23:42,840 S1: both in 2026. That's two years from now, a year 422 00:23:42,840 --> 00:23:44,880 S1: and a half from now, we will have the ability 423 00:23:44,880 --> 00:23:48,240 S1: to have a Deepfaked comic, right? A funny set and 424 00:23:48,240 --> 00:23:50,370 S1: perform it in a way that if a regular person 425 00:23:50,369 --> 00:23:52,920 S1: who enjoys comedy watches it, they'll be like, yeah, that's 426 00:23:52,920 --> 00:23:56,220 S1: pretty funny. That's decently funny. Now, could they get their 427 00:23:56,220 --> 00:24:00,150 S1: own Netflix Netflix special based on that content? That would 428 00:24:00,150 --> 00:24:02,370 S1: be like the top, top tier. But keep in mind 429 00:24:02,369 --> 00:24:06,510 S1: how few comics actually have Netflix specials. So yeah. Interesting. 430 00:24:06,540 --> 00:24:09,419 S1: What did I say? 27. No, I said 26. Yeah. 431 00:24:09,450 --> 00:24:11,909 S1: All right. I'm holding myself to that. San Francisco is 432 00:24:11,910 --> 00:24:14,490 S1: looking to ban software that critics claim is being used 433 00:24:14,490 --> 00:24:20,280 S1: to artificially inflate rates, so it allegedly helps landlords coordinate 434 00:24:20,280 --> 00:24:24,060 S1: rent increases. You might be over using vim visual mode. 435 00:24:24,060 --> 00:24:26,399 S1: I've been worried about this for quite some time, actually. 436 00:24:26,400 --> 00:24:29,129 S1: Oh yeah, instead of doing GG takes you to the 437 00:24:29,130 --> 00:24:33,690 S1: top V which switched to visual mode, and then capital G, 438 00:24:33,869 --> 00:24:37,050 S1: which goes to the bottom of the file. Instead you 439 00:24:37,080 --> 00:24:41,300 S1: do GG quote plus y Why capital? That's a lot 440 00:24:41,330 --> 00:24:43,580 S1: of characters, man. Anyway, I'm going to try to learn 441 00:24:43,580 --> 00:24:45,649 S1: how to do this and do it more with normal 442 00:24:45,650 --> 00:24:48,889 S1: commands instead of visual. All right humans. We're going to 443 00:24:48,890 --> 00:24:53,840 S1: have driver's licenses in the Apple Wallet before too long, 444 00:24:53,869 --> 00:24:56,359 S1: and you'll be able to go through LAX and SFO. 445 00:24:56,390 --> 00:25:00,260 S1: China's manufacturers are facing a financial crisis, with many going 446 00:25:00,260 --> 00:25:03,679 S1: bankrupt due to a combination of weak demand, rising costs 447 00:25:03,710 --> 00:25:07,460 S1: and increased competition. Scientists at Fermilab have detected the first 448 00:25:07,460 --> 00:25:13,189 S1: neutrinos using a prototype detector. Deep Underground Neutrino Experiment. Dune. 449 00:25:13,190 --> 00:25:15,560 S1: You got to land. You got to land. The acronym. 450 00:25:15,560 --> 00:25:18,830 S1: They landed it. Venture capitalists aren't looking for nice founders. 451 00:25:18,830 --> 00:25:22,760 S1: They want risk takers. Nate Silver highlights that 70% of 452 00:25:22,760 --> 00:25:29,720 S1: the billionaires on the 2023 Forbes 400 list are self-made, 70% self-made, 453 00:25:29,720 --> 00:25:33,710 S1: modest backgrounds. So I saw Scott Galloway talking to Nate 454 00:25:33,710 --> 00:25:37,220 S1: Silver about this. I really love this idea of basically 455 00:25:37,220 --> 00:25:39,050 S1: Nate Silver was talking about this thing I've been talking 456 00:25:39,050 --> 00:25:42,580 S1: about for a while, which is they want crazy religious 457 00:25:42,580 --> 00:25:45,640 S1: people who are just like, yep, I have no life whatsoever. 458 00:25:45,640 --> 00:25:48,040 S1: I will sleep under my desk. That is what people 459 00:25:48,040 --> 00:25:50,859 S1: are looking for. They're looking for a crazy vision. And 460 00:25:50,859 --> 00:25:54,040 S1: when I say people, I'm talking about VCs and also 461 00:25:54,040 --> 00:25:57,429 S1: startup founders who are hiring people. And I would argue 462 00:25:57,430 --> 00:25:59,919 S1: this is the future of all work. The future of 463 00:25:59,920 --> 00:26:02,710 S1: all work is you only hire these people. You only 464 00:26:02,710 --> 00:26:07,449 S1: hire the most talented and the most crazy dedicated because 465 00:26:07,450 --> 00:26:10,870 S1: to hire a worker who just does normal work, they 466 00:26:10,869 --> 00:26:14,320 S1: show up 9 to 5 and they do average level work. 467 00:26:14,320 --> 00:26:17,590 S1: That's called AI except for it doesn't go home, it 468 00:26:17,590 --> 00:26:21,070 S1: doesn't commute, it doesn't draw a salary, and you can 469 00:26:21,070 --> 00:26:24,939 S1: upgrade it instantly. Okay, so 9 to 5 workers who 470 00:26:24,940 --> 00:26:28,930 S1: just do average work that gets replaced. The only people 471 00:26:28,930 --> 00:26:31,480 S1: I think that people are going to want going forward. 472 00:26:31,480 --> 00:26:34,000 S1: And this is an exaggeration. There's still going to be 473 00:26:34,000 --> 00:26:36,670 S1: a need for some some of the other type. But 474 00:26:36,670 --> 00:26:40,280 S1: I would say in general, the motion towards is towards 475 00:26:40,280 --> 00:26:44,420 S1: this dedicated zealot crazy person which happens to be young, 476 00:26:44,450 --> 00:26:48,649 S1: happens to be male, usually um, and just obsessed with tech, 477 00:26:48,680 --> 00:26:52,040 S1: obsessed with building things, obsessed with like. And you got 478 00:26:52,040 --> 00:26:55,130 S1: all sorts of hangups around this. You got God complexes, 479 00:26:55,130 --> 00:26:59,180 S1: you got narcissism, you got like, just ambition. Just all 480 00:26:59,180 --> 00:27:02,990 S1: sorts of crazy testosterone powered things going on there. And 481 00:27:02,990 --> 00:27:07,040 S1: the upside is taking risk. They're like, oh yeah, just 482 00:27:07,040 --> 00:27:09,170 S1: I'll take all my money and put it into this 483 00:27:09,170 --> 00:27:12,320 S1: idea and sleep under my desk for four years. And 484 00:27:12,320 --> 00:27:14,540 S1: you don't see many older people doing that. You don't 485 00:27:14,540 --> 00:27:17,810 S1: see many women doing that. There are some, uh, a 486 00:27:17,810 --> 00:27:20,270 S1: decent amount, right? If you look at the entire world, 487 00:27:20,270 --> 00:27:23,929 S1: but in general, it starts to look like one particular demographic. 488 00:27:23,930 --> 00:27:27,770 S1: And silver is basically saying, yeah, that's what gets it done. 489 00:27:27,800 --> 00:27:31,129 S1: That's what VCs are looking for and not people who 490 00:27:31,130 --> 00:27:35,660 S1: are cautious. And I find that really interesting. This negativity, 491 00:27:35,660 --> 00:27:39,880 S1: this risk acceptance. It feels very Atlas Shrugged to me, 492 00:27:39,880 --> 00:27:42,700 S1: which is kind of awesome and kind of gross for 493 00:27:42,760 --> 00:27:46,600 S1: different reasons, but I think that's one of the premises 494 00:27:46,600 --> 00:27:50,109 S1: of Atlas Shrugged is that the risk takers are the 495 00:27:50,109 --> 00:27:52,840 S1: ones who get the benefits, like the awesome people are 496 00:27:52,840 --> 00:27:55,120 S1: the ones who are actually building. And I feel like 497 00:27:55,119 --> 00:27:58,240 S1: Silicon Valley is based around this entire concept, is that 498 00:27:58,240 --> 00:28:00,369 S1: those people who sleep under the desk, they are the 499 00:28:00,369 --> 00:28:03,639 S1: special people, and of course, very few of them actually succeed, 500 00:28:03,820 --> 00:28:07,510 S1: you know, especially the first time. But anyway, really cool 501 00:28:07,510 --> 00:28:10,060 S1: concept coming out of Nate Silver. And by the way, 502 00:28:10,060 --> 00:28:13,180 S1: he's the one who did FiveThirtyEight, although he does not 503 00:28:13,180 --> 00:28:16,150 S1: run it anymore, someone else is running it and they're 504 00:28:16,150 --> 00:28:20,890 S1: actually competing using Nate Silver's brand of FiveThirtyEight. So do 505 00:28:20,890 --> 00:28:23,410 S1: not tie yourself to a brand and then sell it, 506 00:28:23,410 --> 00:28:25,750 S1: because everyone will think you still have it. Growing trend 507 00:28:25,780 --> 00:28:29,500 S1: of Gen Z men becoming NEETs not in employment, education 508 00:28:29,500 --> 00:28:33,100 S1: or training. 1 in 5 young men under 25 unemployed 509 00:28:33,100 --> 00:28:36,910 S1: and not actively looking for work, 1 in 5 unemployed 510 00:28:36,910 --> 00:28:40,070 S1: and not looking for work. What the f is going on? 511 00:28:40,100 --> 00:28:43,130 S1: Slow is smooth. Smooth is fast. I've always loved this. 512 00:28:43,370 --> 00:28:46,700 S1: I'm ex-military myself. I've heard this a million times. It's 513 00:28:46,700 --> 00:28:49,550 S1: not just Navy Seals. Lots of military talk about this, 514 00:28:49,550 --> 00:28:52,790 S1: but cool idea. No one wants kids anymore. And it's 515 00:28:52,790 --> 00:28:56,690 S1: not you. Cool video talks about declining birth rates, touching 516 00:28:56,690 --> 00:29:01,430 S1: on economic pressures, changing societal values, and personal choices. Imposter 517 00:29:01,430 --> 00:29:05,720 S1: syndrome often stems from systemic biases, not just self-doubt, and 518 00:29:05,720 --> 00:29:10,760 S1: highlights that many women experience this due to real exclusionary practices, 519 00:29:10,760 --> 00:29:13,670 S1: which is 100% true. Guy got fired and replaced by 520 00:29:13,670 --> 00:29:17,330 S1: AI at cosmos magazine, and the management didn't tell anyone. 521 00:29:17,330 --> 00:29:21,080 S1: They're using generative AI to write articles, possibly trained on 522 00:29:21,110 --> 00:29:24,590 S1: their own authors work, and basically firing all the humans 523 00:29:24,590 --> 00:29:27,650 S1: who could have possibly predicted this. I give my kids 524 00:29:27,680 --> 00:29:31,160 S1: a summer like mine. In the 1980s, this parent decided 525 00:29:31,160 --> 00:29:34,040 S1: to give her ten and five year old daughters a 526 00:29:34,040 --> 00:29:38,230 S1: taste of a 1980s summer holiday where the bedroom where 527 00:29:38,260 --> 00:29:44,920 S1: boredom was common and self entertainment was key. 1980. Summer holiday. 528 00:29:44,920 --> 00:29:46,390 S1: This is when I grew up. I grew up in 529 00:29:46,390 --> 00:29:51,880 S1: the 80s where boredom was common and self entertainment was key. Boredom. 530 00:29:51,880 --> 00:29:54,220 S1: You know how awesome it is. I'm thinking about this 531 00:29:54,220 --> 00:29:58,450 S1: right now to be a kid outside on a weekend. 532 00:29:58,480 --> 00:30:01,210 S1: You had some sort of breakfast which your parents cooked 533 00:30:01,210 --> 00:30:03,940 S1: for you. You are now out playing. That's what it's called. 534 00:30:03,940 --> 00:30:06,520 S1: You're out playing. I don't know what age this is. 535 00:30:06,520 --> 00:30:10,090 S1: Let's call it 12 or something like that, or 14. 536 00:30:10,090 --> 00:30:13,690 S1: You then get together with your buddies and obviously I'm 537 00:30:13,690 --> 00:30:16,780 S1: a boy, so I'm going to have a boy perspective. 538 00:30:16,780 --> 00:30:19,420 S1: But you would get together with your guy friends all 539 00:30:19,420 --> 00:30:21,340 S1: the same age and you're just like, what are we 540 00:30:21,370 --> 00:30:23,140 S1: going to do? I don't know, I don't know, what 541 00:30:23,140 --> 00:30:24,820 S1: are you going to do? And you're just like, maybe 542 00:30:24,850 --> 00:30:26,890 S1: we should go over here. No. That's stupid. We just 543 00:30:26,890 --> 00:30:29,080 S1: did that. Uh, maybe we should go over here. Hey, 544 00:30:29,080 --> 00:30:30,580 S1: you want to, like, see if we can lift up 545 00:30:30,580 --> 00:30:33,280 S1: these manhole covers and, like, roll them around? Yeah, that 546 00:30:33,280 --> 00:30:35,840 S1: sounds cool. So you go over, you try to stick 547 00:30:35,870 --> 00:30:38,209 S1: your finger in the hole. You try to like, wedge 548 00:30:38,210 --> 00:30:40,640 S1: some sort of piece of metal in there and yank 549 00:30:40,670 --> 00:30:42,530 S1: it up. Turns out the metal was sharp and you 550 00:30:42,530 --> 00:30:45,380 S1: just gouged your hand open. You're bleeding all over the place, 551 00:30:45,380 --> 00:30:47,600 S1: so someone, like, rips the sleeve off of their shirt 552 00:30:47,600 --> 00:30:49,700 S1: and wraps it around your hand, and then you just 553 00:30:49,700 --> 00:30:52,940 S1: keep doing it, and then you just, like, roll this thing. 554 00:30:52,940 --> 00:30:55,640 S1: Then you realize, well, someone walks by here and falls 555 00:30:55,640 --> 00:30:58,700 S1: in that hole like you're super screwed. You know you're 556 00:30:58,700 --> 00:31:00,800 S1: going to hurt yourself. Maybe one of your friends actually 557 00:31:00,800 --> 00:31:03,320 S1: do it. Maybe you decide to climb down there because 558 00:31:03,320 --> 00:31:05,690 S1: that's stupid also, but also kind of fun. Do you 559 00:31:05,690 --> 00:31:09,200 S1: know how many hundreds of these moments that I've had 560 00:31:09,200 --> 00:31:11,960 S1: that Gen X people have had that are just magical? 561 00:31:11,960 --> 00:31:14,060 S1: I was talking to my girl about this as well. 562 00:31:14,060 --> 00:31:17,330 S1: She loves this. The same exact thing. She would get 563 00:31:17,360 --> 00:31:20,750 S1: on her bike and go riding all over the frickin town, 564 00:31:20,750 --> 00:31:23,660 S1: riding all over. Just go visit her friends and hang out. 565 00:31:23,660 --> 00:31:26,210 S1: And she'd have her bike with a banana seat and 566 00:31:26,210 --> 00:31:28,790 S1: just like, take it everywhere. And then she would start 567 00:31:28,790 --> 00:31:31,430 S1: getting hungry. The sun would start going down. And if 568 00:31:31,430 --> 00:31:34,130 S1: she didn't get home once that happened, that was an 569 00:31:34,130 --> 00:31:39,460 S1: ass whoopin and just everyone understood that you are free, 1,000% 570 00:31:39,460 --> 00:31:41,950 S1: free until the street lights come on. Then you got 571 00:31:41,950 --> 00:31:44,770 S1: to get home. Then there's dinner there for you. You 572 00:31:44,770 --> 00:31:48,250 S1: have dinner and it's just magical. There is no boredom now. 573 00:31:48,280 --> 00:31:52,120 S1: Kids do not get bored because TikTok or Facebook or 574 00:31:52,120 --> 00:31:56,260 S1: Instagram or whatever it is. Texting is so compelling and 575 00:31:56,260 --> 00:32:00,700 S1: arguably more compelling than manhole covers, and that's a problem. 576 00:32:00,700 --> 00:32:02,530 S1: This is what I think a big problem is. I 577 00:32:02,530 --> 00:32:05,200 S1: don't know why I'm fucking rambling about this shit, and 578 00:32:05,200 --> 00:32:08,350 S1: I'm supposed to be supposed to be delivering this podcast. 579 00:32:08,410 --> 00:32:11,410 S1: I'm just going to do whatever, um, when stimuli is 580 00:32:11,410 --> 00:32:15,550 S1: too high, the dopamine level is too high, and anything 581 00:32:15,550 --> 00:32:18,670 S1: that drops the dopamine level down, let's say that currently 582 00:32:18,670 --> 00:32:22,420 S1: in 2024, things are so awesome in terms of stimuli 583 00:32:22,420 --> 00:32:26,800 S1: and inputs. The dopamine levels are like an 85, let's 584 00:32:26,800 --> 00:32:30,730 S1: say in the 1980s. As an average kid, they used 585 00:32:30,730 --> 00:32:32,890 S1: to be like a 35, okay. And now they're an 586 00:32:32,890 --> 00:32:36,890 S1: 85 because of like all this tech. Okay, better games, 587 00:32:36,890 --> 00:32:43,310 S1: better game consoles, better screens, better audio experiences, more video games. Okay, 588 00:32:43,340 --> 00:32:47,960 S1: more access to porn. All the social media, TikTok. I mean, 589 00:32:47,960 --> 00:32:50,930 S1: it's not even 85. It's like 285. And it used 590 00:32:50,930 --> 00:32:54,200 S1: to be a 35. Okay, now try to take somebody 591 00:32:54,200 --> 00:32:56,990 S1: like that and put him outside there like, are you 592 00:32:57,020 --> 00:32:59,840 S1: kidding me? It's dirty out here. You want me to 593 00:32:59,870 --> 00:33:02,390 S1: ride my bike? You want me to, like, what am 594 00:33:02,390 --> 00:33:05,989 S1: I supposed to do out here? Nothing compares. Nothing in 595 00:33:06,020 --> 00:33:11,510 S1: that world compares to TikTok. TikTok is crack. It's crack cocaine. 596 00:33:11,510 --> 00:33:15,500 S1: And that's dangerous. Now what everyone seems to know and 597 00:33:15,500 --> 00:33:17,750 S1: the experts seem to know. And I think you can 598 00:33:17,750 --> 00:33:20,390 S1: learn this just by doing it, is you can actually 599 00:33:20,390 --> 00:33:24,860 S1: reset your dopamine level very quickly. So what this person, 600 00:33:24,860 --> 00:33:28,880 S1: I think did this parents decided to give her ten 601 00:33:28,910 --> 00:33:31,459 S1: and five year old daughters a taste of the 1980s 602 00:33:31,490 --> 00:33:35,150 S1: summer holiday where boredom was common and self entertainment was key. 603 00:33:35,180 --> 00:33:37,610 S1: So what that means is you have to take the 604 00:33:37,610 --> 00:33:40,370 S1: things away and throw them together and just be like, 605 00:33:40,400 --> 00:33:43,040 S1: figure out something to do and they'll be like, oh, well, 606 00:33:43,040 --> 00:33:45,140 S1: let's take this blanket and put it at an angle. 607 00:33:45,140 --> 00:33:47,570 S1: And now we have a fort, right? And pretty soon 608 00:33:47,570 --> 00:33:51,260 S1: you're making forts because you were bored. And that enjoyment 609 00:33:51,260 --> 00:33:54,500 S1: of coming up with something from boredom and getting your 610 00:33:54,500 --> 00:33:59,270 S1: dopamine level from a five to a 25 feels amazing. 611 00:33:59,270 --> 00:34:03,830 S1: And it involves more creativity, more innovation. I think it's 612 00:34:03,830 --> 00:34:06,830 S1: just healthier. It's just healthier. It's more natural, it's more healthy. 613 00:34:06,830 --> 00:34:09,470 S1: And I don't think there's a problem with all the 614 00:34:09,469 --> 00:34:14,570 S1: tech and hyper, you know, blasting with dopamine from a 615 00:34:14,570 --> 00:34:17,660 S1: TikTok or whatever. But it's you got to regulate it. 616 00:34:17,660 --> 00:34:20,270 S1: You got to regulate it and go back to this 617 00:34:20,270 --> 00:34:24,530 S1: dopamine fasting and, you know, lower and slower anyway. All right. 618 00:34:24,560 --> 00:34:28,400 S1: Few ideas I've had recently. The ultimate privilege, I think 619 00:34:28,400 --> 00:34:31,250 S1: the ultimate privilege might be growing up in a stable 620 00:34:31,250 --> 00:34:35,010 S1: household with two parents who give you a strong work ethic. 621 00:34:35,040 --> 00:34:37,200 S1: It trips me out how simple this is and how 622 00:34:37,200 --> 00:34:39,960 S1: the best advice is often like this. It's the exact 623 00:34:39,960 --> 00:34:43,830 S1: same with diet, exercise, relationships, and a million other things. 624 00:34:43,860 --> 00:34:48,840 S1: The best advice is concise, wise, and generally hard to do, 625 00:34:48,840 --> 00:34:52,020 S1: but it is not a mystery. I think the US 626 00:34:52,020 --> 00:34:55,440 S1: and the world should lock in on this one thing. 627 00:34:55,440 --> 00:34:59,430 S1: Stable two parent households that imbue a strong work ethic 628 00:34:59,430 --> 00:35:03,360 S1: and focus a lot of energy on getting to 100% 629 00:35:03,360 --> 00:35:06,480 S1: on that metric. All right, next idea. The biggest market 630 00:35:06,480 --> 00:35:09,600 S1: opening right now is for a product platform that validates 631 00:35:09,600 --> 00:35:13,890 S1: the authenticity of content coming from a creator or publisher. Yeah, 632 00:35:13,890 --> 00:35:16,680 S1: that's a tweet. I'm not going to do the whole thread. 633 00:35:16,710 --> 00:35:18,839 S1: I used to think there was a big difference between 634 00:35:18,840 --> 00:35:21,479 S1: somebody being weak and somebody being evil. I now treat 635 00:35:21,480 --> 00:35:24,360 S1: them mostly the same because the outcomes they manifest are 636 00:35:24,360 --> 00:35:26,490 S1: mostly the same. The only difference is that with a 637 00:35:26,489 --> 00:35:29,190 S1: weak person, I could try to make them strong. Someone 638 00:35:29,190 --> 00:35:32,719 S1: responded and said, don't you mean cowardice and not weakness? 639 00:35:32,750 --> 00:35:36,140 S1: I'm kind of equivocating, those two. But I do see 640 00:35:36,140 --> 00:35:39,530 S1: their point. I'm not talking about, like, little children. I'm 641 00:35:39,560 --> 00:35:42,380 S1: talking about. I'm not talking about somebody who's supposed to 642 00:35:42,380 --> 00:35:47,089 S1: be weak because they're disabled or they're super young. They're, 643 00:35:47,120 --> 00:35:49,640 S1: you know, one and a half years old. Stop being weak. 644 00:35:49,670 --> 00:35:51,920 S1: That's not the point. The point is somebody who is 645 00:35:51,920 --> 00:35:55,700 S1: weak due to a flaw in character and who is 646 00:35:55,700 --> 00:36:00,439 S1: causing damage to people around them and themselves as a result. 647 00:36:00,440 --> 00:36:04,970 S1: And importantly, somebody, ideally, who could potentially fix themselves. That's 648 00:36:04,969 --> 00:36:08,000 S1: what I'm talking about. All right. Discovery fabric plus raycast. 649 00:36:08,000 --> 00:36:11,060 S1: So this guy Wil Chen did a YouTube video on 650 00:36:11,060 --> 00:36:14,810 S1: how to integrate fabric into raycast. And guess what. I 651 00:36:14,840 --> 00:36:18,020 S1: am now using raycast. Look at that activate trial. I 652 00:36:18,020 --> 00:36:22,250 S1: need to do that. This is. Yeah, I switched back 653 00:36:22,250 --> 00:36:25,280 S1: to spotlight because I like to use native tools rather 654 00:36:25,280 --> 00:36:27,620 S1: than third party. I was on Alfred for years and 655 00:36:27,620 --> 00:36:30,469 S1: years and years and years. Then I switched back to 656 00:36:30,510 --> 00:36:33,509 S1: spotlight because there was some OS upgrade. I was like, oh, 657 00:36:33,540 --> 00:36:36,000 S1: let's see how good spotlight is. I prefer to use 658 00:36:36,000 --> 00:36:38,820 S1: the native tools anyway, unless there's a compelling reason to 659 00:36:38,820 --> 00:36:41,130 S1: use a third party tool, which is why I'm on 660 00:36:41,130 --> 00:36:44,190 S1: Safari instead of Chrome. Even though Chrome is way faster. 661 00:36:44,190 --> 00:36:49,140 S1: I like the inbuilt stuff in Safari. Anyway, fabric integration 662 00:36:49,140 --> 00:36:52,739 S1: into raycast. That is a big enough reason for me 663 00:36:52,739 --> 00:36:54,989 S1: to switch back. So guess what? Now I'm about to 664 00:36:55,020 --> 00:36:59,100 S1: become a raycast guru and look at this. I might 665 00:36:59,100 --> 00:37:01,860 S1: integrate it more deeply by hosting a set of these 666 00:37:01,860 --> 00:37:05,310 S1: scripts within fabric. So basically these are Python scripts that 667 00:37:05,310 --> 00:37:09,630 S1: raycast reads and that's how raycast executes these things. So 668 00:37:09,630 --> 00:37:12,330 S1: I'm thinking about putting a whole directory of these inside 669 00:37:12,330 --> 00:37:14,489 S1: of fabric. So you could just point raycast to that 670 00:37:14,489 --> 00:37:17,280 S1: directory and boom you're ready to go. And so the 671 00:37:17,280 --> 00:37:19,440 S1: way he described this you know what. Frick it. Just 672 00:37:19,440 --> 00:37:20,550 S1: open it. Watch this. 673 00:37:20,550 --> 00:37:22,470 S2: Uh, okay. So, um. 674 00:37:22,500 --> 00:37:23,609 S1: I don't know if you can hear that, but. 675 00:37:23,610 --> 00:37:27,090 S2: So watch this. I, uh, look at this, like going back. 676 00:37:27,120 --> 00:37:31,360 S2: Going back through these, uh, to these various patterns and fabric. Um, 677 00:37:31,390 --> 00:37:33,819 S2: like if you're if you're like if you have some 678 00:37:33,850 --> 00:37:36,370 S2: text in your selection, like if you, if you have 679 00:37:36,400 --> 00:37:38,230 S2: something you want to select or you want to quick 680 00:37:38,230 --> 00:37:40,299 S2: Paul Graham essay, I can just go ahead and just 681 00:37:40,300 --> 00:37:43,570 S2: type in extract wisdom. Right. And let's say I can 682 00:37:43,570 --> 00:37:44,560 S2: grab this one. 683 00:37:44,560 --> 00:37:45,550 S1: So he copied. 684 00:37:45,580 --> 00:37:47,170 S2: Or I can I can grab the whole thing. Grabs 685 00:37:47,170 --> 00:37:49,989 S2: the whole wisdom right. Extract wisdom. And then just type. 686 00:37:49,989 --> 00:37:51,490 S1: X and then paste it and. 687 00:37:51,489 --> 00:37:52,239 S2: It will run. And look at. 688 00:37:52,239 --> 00:37:53,140 S1: This. Here comes fabric. 689 00:37:53,200 --> 00:37:56,560 S2: That fabric prompt I use it to run it with 690 00:37:56,680 --> 00:38:00,070 S2: Claude Sonnet 3.5. And it's insane. 691 00:38:00,190 --> 00:38:04,120 S1: So yeah, I'm about to rig everything up in that way. 692 00:38:04,150 --> 00:38:07,120 S1: Eric Schmidt of Google did a crazy, honest interview at Stanford, 693 00:38:07,120 --> 00:38:09,310 S1: and it was so spicy that Stanford took it down. 694 00:38:09,310 --> 00:38:11,710 S1: But guess what? I got the link to the video, 695 00:38:11,710 --> 00:38:14,049 S1: and I pulled out the transcript and I made a 696 00:38:14,050 --> 00:38:17,109 S1: fabric summary of it. So there you go. Ideal founding 697 00:38:17,140 --> 00:38:20,110 S1: team Ben Horowitz lays out the perfect founding team in 698 00:38:20,110 --> 00:38:23,049 S1: the clearest way I've ever seen. It's just it's really good. 699 00:38:23,050 --> 00:38:27,279 S1: Scrape it now. New CLI tool for web scraping grok two. 700 00:38:27,310 --> 00:38:29,549 S1: This is the one that just does all sorts of 701 00:38:29,580 --> 00:38:33,690 S1: crazy stuff. Prompt caching with Claude Flux AI by Black 702 00:38:33,690 --> 00:38:36,390 S1: Forest Labs. This one getting a lot of talk. I 703 00:38:36,390 --> 00:38:38,609 S1: haven't used it yet because I'm not doing a lot 704 00:38:38,610 --> 00:38:41,550 S1: of AI image stuff. When I need an AI image, 705 00:38:41,550 --> 00:38:44,939 S1: I usually use Midjourney. Still graphic info. New website lets 706 00:38:44,940 --> 00:38:48,690 S1: you generate infographics to make your articles more engaging. Agile 707 00:38:48,719 --> 00:38:51,360 S1: is for losers, a rant about the author's decade long 708 00:38:51,390 --> 00:38:55,860 S1: frustrations with the agile methodology infiltrating digital agencies, and the 709 00:38:55,860 --> 00:38:59,759 S1: recommendation of the week. Stop accepting it when your loved ones, 710 00:38:59,760 --> 00:39:02,940 S1: especially the young people, are not AI literate. Here's the 711 00:39:02,940 --> 00:39:04,980 S1: way I recommend you think about this. Imagine that the 712 00:39:04,980 --> 00:39:08,460 S1: competition level for getting top jobs or mates or whatever 713 00:39:08,460 --> 00:39:12,690 S1: was 100. In 2022, the average person was at like 714 00:39:12,719 --> 00:39:17,480 S1: an 80. Okay. AI is augmentation technology. It adds 20 715 00:39:17,480 --> 00:39:20,040 S1: to 50 points to people who get good at it. 716 00:39:20,040 --> 00:39:23,100 S1: So now the person with an 85 learns AI and 717 00:39:23,100 --> 00:39:26,850 S1: they're like a 125. So the new standard is reset 718 00:39:26,850 --> 00:39:30,340 S1: from 80 to 120. So if you were a 90 719 00:39:30,370 --> 00:39:33,910 S1: before or a 110, you are now behind. Don't let 720 00:39:33,910 --> 00:39:38,110 S1: your people get left behind. I is the new reading. 721 00:39:38,110 --> 00:39:40,990 S1: It's the new high school diploma. It's the new degree. 722 00:39:40,989 --> 00:39:44,709 S1: Make sure your people, the people that you love, have it. 723 00:39:44,710 --> 00:39:47,350 S1: And just to show you how real this thing is. Okay, 724 00:39:47,380 --> 00:39:50,200 S1: this is the best. This is the absolute best. And 725 00:39:50,200 --> 00:39:53,620 S1: to get you motivated, here is an 18. No an 726 00:39:53,650 --> 00:39:57,370 S1: eight year old. Here's an eight year old doing live coding. 727 00:39:57,370 --> 00:40:01,780 S3: Watch this thing every one. And today I'm well I'm. 728 00:40:01,780 --> 00:40:09,460 S3: Yeah I'm going to show you how to code with cursor. 729 00:40:09,489 --> 00:40:12,850 S3: I'm gonna go to terminal and I'm going to say 730 00:40:12,880 --> 00:40:15,730 S3: new terminal and new terminal. 731 00:40:15,760 --> 00:40:16,030 S1: Look at. 732 00:40:16,030 --> 00:40:16,330 S3: This. 733 00:40:16,810 --> 00:40:18,070 S1: She's she's running. 734 00:40:18,250 --> 00:40:22,120 S3: Pom pom run npm. 735 00:40:22,120 --> 00:40:22,780 S1: Run dev. 736 00:40:22,810 --> 00:40:26,770 S3: There I have that. And then I go to be 737 00:40:26,800 --> 00:40:30,450 S3: and look, I made it. Say hello. I want this 738 00:40:30,450 --> 00:40:33,720 S3: to say. Say hey. 739 00:40:33,750 --> 00:40:34,649 S1: Okay, so we're gonna. 740 00:40:34,950 --> 00:40:35,279 S3: We're gonna. 741 00:40:35,310 --> 00:40:36,720 S1: Jump ahead here. 742 00:40:38,040 --> 00:40:42,600 S3: So, um, I wanted to say chat with Harry and 743 00:40:43,200 --> 00:40:46,620 S3: and then recent cloudfare workers AI. And then it will 744 00:40:46,620 --> 00:40:50,820 S3: do all of the stuff. So you can't do it 745 00:40:50,820 --> 00:40:54,390 S3: if it's not finished, because then it just won't work. 746 00:40:54,480 --> 00:41:01,200 S3: So now I'm gonna save it. So now I'm going 747 00:41:01,230 --> 00:41:06,240 S3: to say hi. Wicked. Hi there. I'm Harry James Potter. 748 00:41:08,580 --> 00:41:09,690 S3: Oh my goodness. 749 00:41:09,690 --> 00:41:12,930 S1: She just built a chat bot to talk to Harry Potter. 750 00:41:13,290 --> 00:41:14,100 S1: She's eight years. 751 00:41:14,100 --> 00:41:15,299 S3: Old, right? 752 00:41:15,330 --> 00:41:18,240 S1: Okay, I'm telling you, that is an eight year old 753 00:41:18,239 --> 00:41:21,509 S1: who just built a chat bot to talk to Harry Potter. Okay. 754 00:41:21,540 --> 00:41:24,060 S1: And you and I have a lot of loved ones 755 00:41:24,060 --> 00:41:26,680 S1: who were like, oh, ChatGPT. Is that still a thing? 756 00:41:26,710 --> 00:41:29,680 S1: How is that still a thing? I thought that blew out, like, 757 00:41:29,710 --> 00:41:32,620 S1: a year ago. I thought it got really important when 758 00:41:32,620 --> 00:41:35,169 S1: people start. Stopped talking about it, and now I don't 759 00:41:35,170 --> 00:41:38,740 S1: care anymore. This is what you're competing with. Eight year olds, 760 00:41:38,739 --> 00:41:41,410 S1: building chat bots to talk to Harry Potter. All right. 761 00:41:41,410 --> 00:41:44,410 S1: And the aphorism of the week standing still in evolution 762 00:41:44,410 --> 00:41:48,610 S1: is equivalent to moving backwards. Standing still in evolution is 763 00:41:48,610 --> 00:41:54,580 S1: equivalent to moving backwards Matt Ridley. Unsupervised learning is produced 764 00:41:54,580 --> 00:41:57,609 S1: and edited by Daniel Miessler on a Neumann U87 AI 765 00:41:57,610 --> 00:42:01,870 S1: microphone using Hindenburg. Intro and outro music is by zombie 766 00:42:01,870 --> 00:42:04,870 S1: with a Y. And to get the text and links 767 00:42:04,870 --> 00:42:07,120 S1: from this episode, sign up for the newsletter version of 768 00:42:07,120 --> 00:42:12,730 S1: the show at Daniel missler.com/newsletter. We'll see you next time.