1 00:00:22,363 --> 00:00:25,633 S1: Welcome to Unsupervised Learning, a security, AI and meaning focused 2 00:00:25,633 --> 00:00:28,513 S1: podcast that looks at how best to thrive as humans 3 00:00:28,513 --> 00:00:32,743 S1: in a post AI world. It combines original ideas, analysis, 4 00:00:32,743 --> 00:00:35,953 S1: and mental models to bring not just the news, but 5 00:00:35,953 --> 00:00:43,073 S1: why it matters and how to respond. All right. Welcome 6 00:00:43,073 --> 00:00:45,983 S1: to unsupervised learning. This is Daniel Missler. So we added 7 00:00:45,983 --> 00:00:48,503 S1: a ton of new patterns to fabric. This week. We 8 00:00:48,503 --> 00:00:51,653 S1: added crate threat model which basically you pipe in the 9 00:00:51,653 --> 00:00:54,593 S1: kind of situation that you're worried about in it outputs 10 00:00:54,593 --> 00:00:57,893 S1: you like a logical threat model. Based on an essay 11 00:00:57,893 --> 00:01:00,353 S1: that I did a while back. Find a hidden message. 12 00:01:00,353 --> 00:01:03,173 S1: This thing is so cool. You basically send in any 13 00:01:03,173 --> 00:01:06,263 S1: essay or article or whatever. You're not sure if you're 14 00:01:06,263 --> 00:01:08,513 S1: being lied to, if or if the person is being 15 00:01:08,513 --> 00:01:12,203 S1: disingenuous or whatever, and it'll actually tell you what they 16 00:01:12,203 --> 00:01:15,323 S1: claim to be trying to say versus what they actually 17 00:01:15,323 --> 00:01:18,773 S1: are saying. So that's a really cool piece of functionality. 18 00:01:19,403 --> 00:01:21,203 S1: And I put a bunch of stuff through it. It's 19 00:01:21,203 --> 00:01:24,143 S1: it's pretty accurate and have it even like fully tweaked 20 00:01:24,143 --> 00:01:28,223 S1: it yet create Ascii visualization. So this thing will actually 21 00:01:28,223 --> 00:01:32,813 S1: create a visualization, a very rudimentary Ascii visualization of basically 22 00:01:32,813 --> 00:01:35,513 S1: any concept. So you can send in like the meaning 23 00:01:35,513 --> 00:01:39,413 S1: of life or the Dunning-Kruger effect or really sort of 24 00:01:39,413 --> 00:01:43,283 S1: conceptual things, and it'll just draw a picture for you. 25 00:01:43,283 --> 00:01:46,013 S1: So you can like explain something to a kid or something. 26 00:01:46,013 --> 00:01:48,023 S1: So I thought that was pretty cool. You could do 27 00:01:48,023 --> 00:01:51,053 S1: the same thing with Mark map, which I didn't know 28 00:01:51,053 --> 00:01:54,263 S1: it was. It's basically like a markdown version of a 29 00:01:54,263 --> 00:01:57,233 S1: mind map. So you do the same thing. You take 30 00:01:57,233 --> 00:01:59,363 S1: any concept, you send it into this and it builds 31 00:01:59,363 --> 00:02:01,823 S1: you a mind map of those things and actually has 32 00:02:01,823 --> 00:02:05,273 S1: links and everything, and you can actually expand and contract 33 00:02:05,273 --> 00:02:09,743 S1: the mind map. It's really cool. And same for mermaid visualization. 34 00:02:09,743 --> 00:02:14,423 S1: This is a JavaScript based, also markdown, but JavaScript. And 35 00:02:14,423 --> 00:02:16,553 S1: you put it into one of the renderers and you 36 00:02:16,553 --> 00:02:20,063 S1: could basically see a mind map or a visualization of 37 00:02:20,063 --> 00:02:22,943 S1: the thing. It's more rich though, than some of the 38 00:02:22,943 --> 00:02:25,703 S1: other ones, so that was really cool. We also added 39 00:02:25,703 --> 00:02:31,913 S1: early crew eye integration. So many, many thanks to Jonathan Dunn, 40 00:02:31,913 --> 00:02:34,673 S1: because he's been working on so many of these pieces 41 00:02:34,673 --> 00:02:37,913 S1: of client functionality, and we now have our first one working. 42 00:02:37,913 --> 00:02:42,503 S1: It's like you run fabric agents, trip planner, trip underscore, planner, 43 00:02:42,503 --> 00:02:46,013 S1: and it will actually run a trip planner for you. 44 00:02:46,013 --> 00:02:48,113 S1: It'll ask you questions, where are you going? What do 45 00:02:48,113 --> 00:02:50,213 S1: you want to do while you're there? And it spins 46 00:02:50,213 --> 00:02:52,523 S1: up a team of agents and actually does that for you. 47 00:02:52,523 --> 00:02:56,453 S1: So that's now integrated and brand new. Just this morning. 48 00:02:56,453 --> 00:03:01,433 S1: It's not even in the newsletter yet is local model support. 49 00:03:01,433 --> 00:03:05,333 S1: So you can now run local model. You actually do 50 00:03:05,363 --> 00:03:08,843 S1: capital L and then if you do like llama too 51 00:03:08,843 --> 00:03:11,813 S1: it'll run. You have to install a llama and get 52 00:03:11,813 --> 00:03:14,573 S1: it running. But that's like one command. And once it's 53 00:03:14,573 --> 00:03:17,903 S1: done and you have the model actually downloaded, it'll give 54 00:03:17,903 --> 00:03:19,853 S1: you the result from a local one. So you're not 55 00:03:19,853 --> 00:03:23,333 S1: even being charged any tokens. And this thing totally works. 56 00:03:23,333 --> 00:03:26,243 S1: Just got that running in. The documentation is going as well. 57 00:03:26,243 --> 00:03:29,393 S1: So thanks again to Jonathan Dunn for that work. And 58 00:03:29,393 --> 00:03:31,943 S1: you want to just yeah, you want to rerun setup 59 00:03:31,943 --> 00:03:34,703 S1: and restart your shell and then it'll work. All right. 60 00:03:34,703 --> 00:03:38,183 S1: So my work this week wrote this essay called To 61 00:03:38,183 --> 00:03:42,173 S1: Survive I we must become creators. Highly recommend this one. 62 00:03:42,173 --> 00:03:45,233 S1: Especially for anyone who is in school or has a 63 00:03:45,233 --> 00:03:47,783 S1: loved one in school. It was kind of spawned by 64 00:03:47,783 --> 00:03:50,843 S1: a post that I saw from somebody on Hacker News. 65 00:03:50,843 --> 00:03:53,633 S1: In fact, I'm just going to click into this thing. Yeah, 66 00:03:53,633 --> 00:03:57,293 S1: this post from Hacker News click into that as well. Yeah. 67 00:03:57,293 --> 00:03:59,603 S1: So they're basically saying I'm about to finish my CSS 68 00:03:59,603 --> 00:04:02,423 S1: degree in May, and I'm already sick of the tech industry. 69 00:04:02,423 --> 00:04:04,673 S1: I'm at a loss. I know the job market could 70 00:04:04,673 --> 00:04:07,823 S1: get better soon, but there's no guarantee that it will. 71 00:04:07,823 --> 00:04:11,213 S1: And things are looking bad for us with less experience. 72 00:04:11,213 --> 00:04:15,083 S1: And basically just it's kind of sad. It's kind of 73 00:04:15,083 --> 00:04:17,723 S1: a real situation that's going on. So I wanted to 74 00:04:17,723 --> 00:04:20,693 S1: address that, and I think this is the way that 75 00:04:20,693 --> 00:04:22,973 S1: people should be thinking about it. In my opinion, this 76 00:04:22,973 --> 00:04:27,713 S1: is just my opinion. But I think that coders and 77 00:04:27,713 --> 00:04:30,053 S1: lots of people like that are simply not going to 78 00:04:30,053 --> 00:04:35,063 S1: have jobs, right? And some things will go faster than others. 79 00:04:35,063 --> 00:04:38,123 S1: But do a quick little walkthrough through here. You want 80 00:04:38,123 --> 00:04:41,243 S1: to read the whole thing, but basically what is a 81 00:04:41,243 --> 00:04:45,653 S1: creator versus a worker? So on one hand it's pretty simple, right? 82 00:04:45,653 --> 00:04:49,013 S1: So its creators come up with new solutions to human problems. 83 00:04:49,013 --> 00:04:52,013 S1: Workers are the people who execute on building them and 84 00:04:52,013 --> 00:04:55,043 S1: building those solutions. And then like a corporate setting, it 85 00:04:55,043 --> 00:04:58,103 S1: would be like creators are the people who determine what 86 00:04:58,103 --> 00:05:00,293 S1: to build, how to build it, if it's a new thing, 87 00:05:00,293 --> 00:05:02,093 S1: and then how to sell it in the current market. 88 00:05:02,093 --> 00:05:05,693 S1: And workers are the people who make that happen. Right now, 89 00:05:05,693 --> 00:05:08,423 S1: there are situations like we talk about this draw the 90 00:05:08,423 --> 00:05:11,783 S1: rest of the owl. There are some situations where it's 91 00:05:11,783 --> 00:05:14,573 S1: not quite clear. If it's if you're doing a creator 92 00:05:14,573 --> 00:05:18,803 S1: job or an execution job, right? Because if you're making 93 00:05:18,803 --> 00:05:22,253 S1: something that's never been built, technically you're executing, but you're 94 00:05:22,253 --> 00:05:25,823 S1: making something that's brand new, you're kind of creating, especially 95 00:05:25,823 --> 00:05:27,353 S1: if you're doing that and you have your own company, 96 00:05:27,353 --> 00:05:30,773 S1: you're definitely creating. But the vice versa is also true. 97 00:05:30,773 --> 00:05:33,383 S1: So it's like if you used to be a creator, 98 00:05:33,383 --> 00:05:35,903 S1: but now you're creating things that kind of are kind 99 00:05:35,903 --> 00:05:38,363 S1: of a commodity at this point. Well, now you're more 100 00:05:38,363 --> 00:05:40,883 S1: of an executor than a creator. So it goes both ways. 101 00:05:41,183 --> 00:05:43,313 S1: And here's the way to look at it. Basically, I 102 00:05:43,313 --> 00:05:45,833 S1: say as the amount of creation goes up, the amount 103 00:05:45,833 --> 00:05:49,433 S1: of execution will balloon massively, but it's going to be 104 00:05:49,433 --> 00:05:52,703 S1: AI that are actually filling those jobs, right? It's not 105 00:05:52,703 --> 00:05:55,703 S1: going to be humans that are doing all that execution. 106 00:05:55,703 --> 00:05:58,523 S1: Once AI comes online is good enough to do a 107 00:05:58,523 --> 00:06:00,653 S1: lot of this stuff, which I don't think is going 108 00:06:00,653 --> 00:06:04,823 S1: to be long. Humans are basically hard to train and retrain, 109 00:06:04,823 --> 00:06:09,953 S1: and human competence is fixed, right? Our IQs are not 110 00:06:09,953 --> 00:06:11,753 S1: going to go up that fast. I mean, we might 111 00:06:11,753 --> 00:06:15,413 S1: get some improvement on our IQs from, you know, some 112 00:06:15,413 --> 00:06:18,983 S1: AI or some enhancements or nutrition. There's lots of different 113 00:06:18,983 --> 00:06:22,493 S1: things can improve our our IQs as humans, but it's 114 00:06:22,493 --> 00:06:25,043 S1: not going to go up nearly as fast as AI is. 115 00:06:25,043 --> 00:06:29,543 S1: And ultimately humans are expensive, sensitive, and need constant replacement. 116 00:06:29,543 --> 00:06:31,973 S1: And I won't have these problems, right? It's going to 117 00:06:31,973 --> 00:06:34,433 S1: scale with creation. We can make as many AIS as 118 00:06:34,433 --> 00:06:37,283 S1: we need to do a particular job. When you upgrade 119 00:06:37,283 --> 00:06:39,953 S1: an AI, it's pretty easy to roll it out, right? 120 00:06:39,953 --> 00:06:43,253 S1: And they aren't conscious, they don't get tired, they don't complain, 121 00:06:43,253 --> 00:06:46,103 S1: they don't go to HR, you know, to start investigation 122 00:06:46,103 --> 00:06:49,433 S1: and they don't quit. And they're getting smarter at an 123 00:06:49,433 --> 00:06:53,273 S1: insane rate. So what do you tangibly do? Right. You know, 124 00:06:53,273 --> 00:06:55,763 S1: what fields do you go into? How do I steer 125 00:06:55,763 --> 00:06:59,363 S1: my kids towards something or away from something? And should 126 00:06:59,363 --> 00:07:01,853 S1: I still go to college? So my advice is to 127 00:07:01,853 --> 00:07:05,873 S1: focus on creation, on ideas, on making new things, focus 128 00:07:05,873 --> 00:07:07,733 S1: on problems that exist in the world that need to 129 00:07:07,733 --> 00:07:10,013 S1: be solved, and start thinking about what you could do 130 00:07:10,013 --> 00:07:14,093 S1: to solve those problems, what you can build that solves 131 00:07:14,093 --> 00:07:18,803 S1: those problems, right? And basically creating new things to solve 132 00:07:18,803 --> 00:07:22,523 S1: human problems is like the immortal job skill. It's the 133 00:07:22,523 --> 00:07:25,613 S1: best job skill that you can have. And then you 134 00:07:25,613 --> 00:07:29,963 S1: want to think of tech skill and specifically programming in 135 00:07:29,963 --> 00:07:34,133 S1: AI as the new reading and writing. It's essential. Meaning 136 00:07:34,133 --> 00:07:36,863 S1: if you're not good at them, then you're probably not 137 00:07:36,863 --> 00:07:41,523 S1: going to succeed. You're probably going to be like an executor. And. 138 00:07:42,193 --> 00:07:46,393 S1: Even as a creator who's actually giving these this work 139 00:07:46,393 --> 00:07:50,563 S1: to eyes, you still need to be decently fluid, fluent 140 00:07:50,563 --> 00:07:55,063 S1: in programming in AI because they are the language of creation. 141 00:07:55,063 --> 00:07:57,433 S1: You need to be able to explain yourself to AI, 142 00:07:57,433 --> 00:08:00,073 S1: and that that largely is going to be in the 143 00:08:00,073 --> 00:08:04,453 S1: language of like prompting and speaking and thinking and writing. 144 00:08:04,453 --> 00:08:07,333 S1: And some of that is going to involve programming, right? 145 00:08:07,513 --> 00:08:09,673 S1: It'll be less and less as time goes on, but 146 00:08:09,673 --> 00:08:11,713 S1: you should just learn that stuff. The other thing is 147 00:08:11,713 --> 00:08:13,603 S1: you want to get trained on this stuff, and I 148 00:08:13,603 --> 00:08:16,093 S1: think you want to get trained in two primary areas. 149 00:08:16,093 --> 00:08:19,033 S1: And I do think college actually still matters, because it's 150 00:08:19,033 --> 00:08:21,943 S1: a way of filtering the quality of a person when 151 00:08:21,943 --> 00:08:24,343 S1: they don't know anything else about you. So I think 152 00:08:24,343 --> 00:08:26,773 S1: college is still going to matter for a little while longer, 153 00:08:26,773 --> 00:08:30,163 S1: maybe like a decade or something. But when you do study, 154 00:08:30,163 --> 00:08:33,283 S1: whether it's in college or self-study, you want to be 155 00:08:33,283 --> 00:08:36,943 S1: studying two branches simultaneously. One is a hard skill that's 156 00:08:36,943 --> 00:08:41,353 S1: valuable in the market, like AI or something. AI or 157 00:08:41,353 --> 00:08:45,313 S1: development or something like that. And number two is training 158 00:08:45,313 --> 00:08:48,883 S1: on how to think. This one is even more important, 159 00:08:48,883 --> 00:08:51,133 S1: but you don't want to do it at the expense 160 00:08:51,133 --> 00:08:52,903 S1: of the first one. You want to have something, a 161 00:08:52,903 --> 00:08:56,743 S1: hard skill to go along with it. But as time 162 00:08:56,743 --> 00:09:00,223 S1: goes on, being able to think especially about problems and 163 00:09:00,223 --> 00:09:08,263 S1: solutions is like the most important thing. So liberal arts, education, philosophy, logic, dialectic, 164 00:09:08,263 --> 00:09:11,773 S1: all these things become way, way more important. They they 165 00:09:11,773 --> 00:09:14,563 S1: become the things that separate the creators and like the 166 00:09:14,563 --> 00:09:18,973 S1: owners and the, the top people from everyone else. So 167 00:09:18,973 --> 00:09:21,433 S1: you really want to focus on both of those things 168 00:09:21,433 --> 00:09:23,473 S1: in your education. You don't want to think so much 169 00:09:23,473 --> 00:09:25,963 S1: about fields or companies. You want to think more about 170 00:09:25,963 --> 00:09:29,863 S1: problems and problem spaces, right? What are the problems that 171 00:09:29,863 --> 00:09:32,203 S1: will go away versus what are the problems that are 172 00:09:32,203 --> 00:09:34,243 S1: going to stay with us all the time? And you 173 00:09:34,243 --> 00:09:37,543 S1: want to go to where the problems are that interest you, 174 00:09:37,543 --> 00:09:39,643 S1: and you'd be good at solving. And most of all, 175 00:09:39,643 --> 00:09:41,323 S1: you want to get out of the mindset of being 176 00:09:41,323 --> 00:09:43,993 S1: a worker and enter the mindset of being a creator 177 00:09:43,993 --> 00:09:46,273 S1: or a builder. So you don't want to be the 178 00:09:46,273 --> 00:09:48,883 S1: person building other people's ideas. You want to be the 179 00:09:48,883 --> 00:09:52,933 S1: person creating ideas, and in finding the problems and building 180 00:09:52,933 --> 00:09:55,543 S1: things that solve those problems. And that's the summary there, 181 00:09:55,543 --> 00:09:58,363 S1: and that's essentially the post. So I didn't think we 182 00:09:58,363 --> 00:09:59,683 S1: were going to do the whole thing, but that was 183 00:09:59,683 --> 00:10:03,103 S1: essentially it. All right. So that was that one also 184 00:10:03,103 --> 00:10:07,663 S1: posted our sponsored interview with BlackBerry. The second one, BlackBerry 185 00:10:07,663 --> 00:10:13,243 S1: silence with Ismael Valenzuela, and he is the Threat research 186 00:10:13,243 --> 00:10:17,173 S1: and intelligence VP at BlackBerry silence. And we talked about 187 00:10:17,173 --> 00:10:19,213 S1: a whole bunch of cool stuff there. So definitely go 188 00:10:19,213 --> 00:10:22,693 S1: check that one out. Going into security, researchers created a 189 00:10:22,693 --> 00:10:27,403 S1: worm that exploits generative AI and spreads via prompt injection. 190 00:10:27,403 --> 00:10:29,713 S1: So there's kind of a and they named it Morris two, 191 00:10:29,743 --> 00:10:32,593 S1: which I thought was pretty cool. GitHub is now blocking 192 00:10:32,593 --> 00:10:37,273 S1: commits with secrets and public repos. That's really cool. Biden 193 00:10:37,273 --> 00:10:40,633 S1: is viewing Chinese connected cars as a national security threat, 194 00:10:40,633 --> 00:10:43,993 S1: which I'm very happy about. Got to sponsor here. Collide. 195 00:10:43,993 --> 00:10:48,253 S1: Thank you to collide for sponsoring us. Military's Project Maven 196 00:10:48,253 --> 00:10:52,393 S1: is now actively using AI to identify and strike targets, 197 00:10:52,393 --> 00:10:57,163 S1: marking a significant shift from skepticism around doing this to 198 00:10:57,163 --> 00:10:59,923 S1: actually relying on it. And they've already done this in Yemen, 199 00:10:59,923 --> 00:11:04,663 S1: the Red sea, Iraq and Syria. Shot spotter who renamed 200 00:11:04,663 --> 00:11:08,773 S1: themselves probably to get around some PR drama, is now 201 00:11:08,773 --> 00:11:12,793 S1: called sound Thinking and uses hidden sensors for gunfire detection. 202 00:11:12,793 --> 00:11:17,473 S1: And it basically reveals exact locations of those sensors. And 203 00:11:17,473 --> 00:11:21,853 S1: this data was previously unknown. Researchers found over 200 hacking 204 00:11:21,853 --> 00:11:27,103 S1: services using AI. Things like bots like bad GPT. Cryptocurrency 205 00:11:27,103 --> 00:11:30,763 S1: enthusiasts are being targeted with Mac malware. A group of hackers, 206 00:11:30,763 --> 00:11:34,933 S1: including Rizzo my buddy Joseph Thakur, found some vulnerabilities in 207 00:11:34,933 --> 00:11:37,573 S1: Google's AI and cloud systems, and they got 50 grand 208 00:11:37,573 --> 00:11:40,903 S1: for that and a new vulnerability in hugging faces Safe 209 00:11:40,903 --> 00:11:44,923 S1: Tensors conversion service could lead to supply chain attacks in 210 00:11:44,923 --> 00:11:49,363 S1: AI models. All right. Technology. Yeah. Nvidia's CEO thinks I 211 00:11:49,363 --> 00:11:52,663 S1: will soon make coding obsolete. Yeah, like we talked about above. 212 00:11:52,663 --> 00:11:55,003 S1: I think that's true. And we just talked about this 213 00:11:55,003 --> 00:11:59,473 S1: creators and executors essay here. Waymo is expanding to more 214 00:11:59,473 --> 00:12:02,203 S1: highways in LA in the Bay area. I can't wait 215 00:12:02,203 --> 00:12:05,083 S1: till they come to my area. I absolutely loved my 216 00:12:05,083 --> 00:12:08,533 S1: experience with Waymo and Apple canceled the car project. I 217 00:12:08,533 --> 00:12:10,753 S1: think everyone knows about this at this point, but they 218 00:12:10,753 --> 00:12:14,563 S1: moved over 2000 employees to AI initiatives, which I'm super 219 00:12:14,563 --> 00:12:18,013 S1: excited about. I really cannot wait until this June when 220 00:12:18,013 --> 00:12:20,593 S1: they do the new announcements. It's going to be. I'm 221 00:12:20,593 --> 00:12:24,523 S1: so hyped for those announcements. Basically, good AI that's close 222 00:12:24,523 --> 00:12:26,743 S1: to you and actually on your phone is so much 223 00:12:26,743 --> 00:12:29,623 S1: better than amazing. I that's stuck in like an IDE 224 00:12:29,623 --> 00:12:33,913 S1: somewhere in 2023, public tech companies out of 2.4 trillion 225 00:12:33,913 --> 00:12:37,963 S1: to their market caps. Musk is suing OpenAI, saying that 226 00:12:37,963 --> 00:12:41,383 S1: they basically got away from their altruistic nature and are just. 227 00:12:41,493 --> 00:12:43,983 S1: Going after profits, something like that. I'm not sure if 228 00:12:43,983 --> 00:12:47,073 S1: that's his exact claim, but it's something like that. Basically, 229 00:12:47,073 --> 00:12:50,793 S1: OpenAI says the New York Times paid someone to hack 230 00:12:50,793 --> 00:12:55,323 S1: its products to produce content matching the newspaper's articles. This 231 00:12:55,323 --> 00:12:58,803 S1: is wild. I haven't fully dove into, like, their exact 232 00:12:58,803 --> 00:13:03,153 S1: claims in the in the lawsuit or whatever they're doing, 233 00:13:03,153 --> 00:13:07,173 S1: but interesting plot point there. DocuSign has been using customer 234 00:13:07,173 --> 00:13:10,113 S1: data to train their AI. People are freaking out about it. 235 00:13:10,113 --> 00:13:12,543 S1: I think this is kind of like the wrong question. 236 00:13:12,543 --> 00:13:16,413 S1: It's not about if someone is using customer data to 237 00:13:16,413 --> 00:13:18,993 S1: train their AI. Of course they are. Everyone should be 238 00:13:18,993 --> 00:13:21,123 S1: doing that. The question is, are you training on like 239 00:13:21,123 --> 00:13:24,723 S1: personal data, on sensitive data, on privacy related data in 240 00:13:24,723 --> 00:13:27,483 S1: a way that customers would object to? Customers are not 241 00:13:27,483 --> 00:13:31,413 S1: going to object to you understanding, like their behavior and stuff. 242 00:13:31,413 --> 00:13:33,633 S1: It's like it's going to come down to like, are 243 00:13:33,633 --> 00:13:37,713 S1: you targeting somebody or are you doing something that's unexpected 244 00:13:37,713 --> 00:13:43,383 S1: and gross that that's the distinction. SpaceX just hit 17mbps 245 00:13:43,383 --> 00:13:47,223 S1: downloading directly to an Android phone. No. Now, this is 246 00:13:47,223 --> 00:13:49,683 S1: like a new cell tower, essentially, but it's a cell 247 00:13:49,683 --> 00:13:53,343 S1: tower that's actually a satellite. 17 megs download to a 248 00:13:53,343 --> 00:13:57,123 S1: stock Android phone. Wendy's is looking to test dynamic surge 249 00:13:57,123 --> 00:14:01,173 S1: pricing for food in 2025. Seems a little early to 250 00:14:01,173 --> 00:14:06,363 S1: announce a strategy that you haven't implemented yet. Like, isn't 251 00:14:06,363 --> 00:14:10,233 S1: that giving competitors too long to adjust? Yeah, right. What 252 00:14:10,233 --> 00:14:13,863 S1: do I know about fast food strategy? Though? February and 253 00:14:13,863 --> 00:14:18,693 S1: January saw a resurgence in tech job cuts. Yes it did. 254 00:14:18,693 --> 00:14:21,093 S1: They are continuing. I didn't think that was going to happen. 255 00:14:21,183 --> 00:14:26,163 S1: And the 50 over 90 by Nvidia is going to 256 00:14:26,163 --> 00:14:30,423 S1: be like 70% faster than the 4090. Makes me sad. 257 00:14:30,423 --> 00:14:35,913 S1: I got two 4090, but whatever, 192 streaming Multiprocessors and 258 00:14:35,913 --> 00:14:40,563 S1: 24,000 Cuda cores, that is insane. And a new study 259 00:14:40,563 --> 00:14:44,103 S1: of 13,000 people showed. And this is in the human 260 00:14:44,103 --> 00:14:48,663 S1: section that long Covid. And going into the human section, 261 00:14:48,663 --> 00:14:53,013 S1: a new study of 13,000 people showed with long Covid, 262 00:14:53,043 --> 00:14:56,343 S1: they lost about six IQ points, according to the study. 263 00:14:56,343 --> 00:15:00,903 S1: Political extremism is now one of America's top concerns. Edging 264 00:15:00,903 --> 00:15:04,833 S1: out the economy and immigration organ is reversing its drug 265 00:15:04,833 --> 00:15:10,203 S1: decriminalization policy due to overdose deaths and public concern. I 266 00:15:10,203 --> 00:15:12,963 S1: feel like 2024 is the year of like, the pendulum 267 00:15:12,963 --> 00:15:16,563 S1: swinging back against a whole bunch of hyper liberal policies 268 00:15:16,563 --> 00:15:19,053 S1: and attitudes. I just hope it doesn't swing too far 269 00:15:19,053 --> 00:15:22,623 S1: and then, you know, then just keeps violent swings of 270 00:15:22,623 --> 00:15:25,533 S1: the pendulum. It's just no good for anybody. California is 271 00:15:25,533 --> 00:15:29,373 S1: proposing a bill to ban the homeless encampments near public spaces. 272 00:15:29,373 --> 00:15:34,023 S1: Florida is experiencing a number of outbreaks of already beaten diseases. Basically, 273 00:15:34,023 --> 00:15:36,003 S1: you've got people on the left and the right who 274 00:15:36,003 --> 00:15:39,753 S1: are vaccine skeptics. They're not vaccinating their kids. And we're 275 00:15:39,753 --> 00:15:43,143 S1: not hitting the numbers for herd immunity that are required 276 00:15:43,143 --> 00:15:47,073 S1: for herd immunity. And people are getting diseases like the 277 00:15:47,073 --> 00:15:50,283 S1: measles that that have been, like, already knocked out for 278 00:15:50,283 --> 00:15:54,213 S1: decades or hundreds of years or whatever. And alcohol related 279 00:15:54,213 --> 00:15:58,353 S1: deaths in the US jumped by nearly 30%, almost 500 daily. 280 00:15:58,353 --> 00:16:02,253 S1: And a neurosurgeon is using ultrasound to tackle Alzheimer's and 281 00:16:02,253 --> 00:16:06,273 S1: addiction with some really exciting results. Had a crazy idea 282 00:16:06,273 --> 00:16:10,563 S1: in the idea section here. What if AGI powered robots? 283 00:16:10,563 --> 00:16:14,553 S1: What if we get those before we get autonomous cars? 284 00:16:14,553 --> 00:16:18,753 S1: What if it's actually easier to do AGI that is 285 00:16:18,753 --> 00:16:22,863 S1: capable of like evaluating a situation? Maybe not. Maybe. Maybe 286 00:16:22,863 --> 00:16:27,363 S1: they are the same thing, because the AGI that we 287 00:16:27,363 --> 00:16:30,903 S1: will have built wouldn't be trained on car stuff, and 288 00:16:30,903 --> 00:16:34,563 S1: Tesla stuff would still be the best ones anyway. It's 289 00:16:34,563 --> 00:16:37,293 S1: just a possibility. It random thought. In other words, maybe 290 00:16:37,293 --> 00:16:40,833 S1: it's easier to put a local GPT six level AI 291 00:16:40,833 --> 00:16:43,083 S1: in a car and have it learn a whole bunch 292 00:16:43,083 --> 00:16:47,583 S1: of stuff than to actually try to make the car autonomous. 293 00:16:47,583 --> 00:16:50,013 S1: I don't know, like I said, it's a random thought 294 00:16:50,013 --> 00:16:53,973 S1: and Sunday was 30 years with my love two piece 295 00:16:53,973 --> 00:16:57,813 S1: Combee and Dune two was insanely good. I've already seen 296 00:16:57,813 --> 00:17:00,243 S1: it twice. Going to see it again. All right. Discovery. 297 00:17:00,243 --> 00:17:03,273 S1: My buddy Jason Haddix put out a sick episode of 298 00:17:03,273 --> 00:17:07,623 S1: his newsletter called Executive Offense, and lots of prompt injection 299 00:17:07,623 --> 00:17:10,413 S1: and other resources about hacking. I definitely want to go 300 00:17:10,413 --> 00:17:14,133 S1: check that out and subscribe to his newsletter. Do literally anything. 301 00:17:14,133 --> 00:17:19,353 S1: Great essay here. Caltrans has public data in CCTV, data 302 00:17:19,353 --> 00:17:24,003 S1: in CSV, Json, text, XML formats. Got a deep dive 303 00:17:24,003 --> 00:17:27,093 S1: into a git configuration here. How to get nmap to 304 00:17:27,093 --> 00:17:30,663 S1: detect new services. How did this person decides if your 305 00:17:30,663 --> 00:17:34,263 S1: website is worth a revisit? The internet feels fake now. 306 00:17:34,263 --> 00:17:39,423 S1: Tyler Cohen shares his personal strategy around listening to music. 307 00:17:39,423 --> 00:17:41,753 S1: I didn't agree with it, but I. I actually really 308 00:17:41,753 --> 00:17:45,923 S1: enjoyed reading it. Apple's releasing Neuromancer as a ten episode 309 00:17:45,923 --> 00:17:49,553 S1: series on Apple TV+. I definitely agree with somebody who 310 00:17:49,553 --> 00:17:51,893 S1: said this. I can't remember if it was Scott Galloway, 311 00:17:51,893 --> 00:17:55,943 S1: but Apple TV is basically basically become the new HBO. 312 00:17:55,943 --> 00:17:58,523 S1: HBO got knocked out because they're calling it Macs now 313 00:17:58,523 --> 00:18:03,413 S1: and nobody cares. And so now Apple TV is like fire, 314 00:18:03,413 --> 00:18:06,413 S1: like it's doing so well, and it's basically become the 315 00:18:06,413 --> 00:18:09,413 S1: new HBO. It's like if a show comes on to 316 00:18:09,413 --> 00:18:13,703 S1: Apple TV, it's probably pretty good. Bad therapy argues modern 317 00:18:13,703 --> 00:18:17,903 S1: therapeutic parenting is failing, leaving kids anxious and unprepared for life. 318 00:18:17,903 --> 00:18:21,263 S1: Algorithmic thinking. I've ordered this thing. I've got the physical 319 00:18:21,263 --> 00:18:24,143 S1: copy coming. I read the first one. Absolutely loved it. 320 00:18:24,143 --> 00:18:26,303 S1: I was working at Apple. I actually remember I was 321 00:18:26,633 --> 00:18:29,693 S1: talking to someone about it taking a walk, but absolutely 322 00:18:29,693 --> 00:18:32,123 S1: love this book. And now it's version two or a 323 00:18:32,123 --> 00:18:34,583 S1: second edition and it's got new chapters. Can't wait to 324 00:18:34,583 --> 00:18:37,913 S1: read it. It's on the way, and spending ten minutes 325 00:18:37,913 --> 00:18:40,733 S1: on something is 1% of your day. And I think 326 00:18:40,733 --> 00:18:44,393 S1: that's assuming, like I forget how many hours, 16 hours 327 00:18:44,393 --> 00:18:47,363 S1: a week. They have some assumptions in there and the 328 00:18:47,363 --> 00:18:51,743 S1: recommendation of the week. Ask yourself if you're primarily a nurturer, 329 00:18:52,133 --> 00:18:55,313 S1: a creator or a worker, because it's my belief that 330 00:18:55,313 --> 00:18:58,493 S1: creators are nurturers. And this goes back to the essay, um, 331 00:18:58,493 --> 00:19:02,063 S1: people that help nurtures are basically people that help creators 332 00:19:02,063 --> 00:19:06,233 S1: or nurtures grow up and and become, uh, fruitful. So 333 00:19:06,233 --> 00:19:11,303 S1: it's basically parents or nurturers. Um, artists are creators, entrepreneurs 334 00:19:11,303 --> 00:19:15,953 S1: are creators, therapists are nurturers, etc. and I think that 335 00:19:15,953 --> 00:19:19,313 S1: creators are nurturers are the roles that will be most 336 00:19:19,313 --> 00:19:22,223 S1: resilient to AI, and they're also the most human roles 337 00:19:22,223 --> 00:19:24,893 S1: that they're what we should be doing anyway. We shouldn't 338 00:19:24,893 --> 00:19:28,973 S1: be like moving paperwork around and tilling fields and all 339 00:19:28,973 --> 00:19:32,033 S1: this crap. Try to get this is my recommendation. Try 340 00:19:32,033 --> 00:19:35,663 S1: to get out of the worker mentality. My family is Lutheran, right? 341 00:19:36,053 --> 00:19:39,203 S1: Hard work. My dad always instilled this into me. Like 342 00:19:39,203 --> 00:19:41,663 S1: your reputation is based on the work that you do. 343 00:19:41,663 --> 00:19:44,063 S1: So like I get this and I feel it deeply 344 00:19:44,063 --> 00:19:49,133 S1: and I think hard work is noble and honorable 100%. 345 00:19:49,133 --> 00:19:53,453 S1: But we were handed arbitrary jobs just because of the 346 00:19:53,453 --> 00:19:56,753 S1: the escalation of tech through our lives. Right? When there's 347 00:19:56,753 --> 00:19:59,213 S1: no clean water, you know, a lot of the work 348 00:19:59,213 --> 00:20:01,193 S1: is trying to get clean water when there's no food. 349 00:20:01,193 --> 00:20:02,873 S1: A lot of the work is trying to get food. 350 00:20:02,873 --> 00:20:05,033 S1: Like all those things get phased out as the tech 351 00:20:05,033 --> 00:20:07,763 S1: gets better. And that's what we're about to go through. Again, 352 00:20:07,763 --> 00:20:10,763 S1: an AI is going to be doing most of the 353 00:20:10,763 --> 00:20:14,063 S1: old worker style jobs better, so we have to start 354 00:20:14,063 --> 00:20:19,733 S1: planning our migration to a creator and a nurturer style economy. 355 00:20:19,733 --> 00:20:21,833 S1: And we're all going to be hybrids for for a 356 00:20:21,833 --> 00:20:24,653 S1: long time, but you want to try to migrate to 357 00:20:24,653 --> 00:20:26,903 S1: that as soon as possible. And until the people that 358 00:20:26,903 --> 00:20:28,673 S1: you care about to try to do the same and 359 00:20:28,673 --> 00:20:31,583 S1: the optimism for the week along those lines, the end 360 00:20:31,583 --> 00:20:36,613 S1: of labor is to gain leisure. Aristotle. Unsupervised Learning is 361 00:20:36,613 --> 00:20:39,763 S1: produced and edited by Daniel Miller on a Norman 87 362 00:20:39,763 --> 00:20:43,843 S1: AI microphone using Hindenburg. Intro and outro music is by 363 00:20:43,843 --> 00:20:47,053 S1: zombie with the Y, and to get the text and 364 00:20:47,053 --> 00:20:49,423 S1: links from this episode, sign up for the newsletter version 365 00:20:49,423 --> 00:20:54,993 S1: of the show at Daniel missler.com/newsletter. We'll see you next time.