1 00:00:17,347 --> 00:00:21,067 S1: All right. Welcome to episode 483. Just a quick reminder. 2 00:00:21,067 --> 00:00:23,667 S1: I'm trying to get a lot more content out, and 3 00:00:23,707 --> 00:00:27,387 S1: I don't want to miss regular newsletter episodes. And I 4 00:00:27,387 --> 00:00:30,227 S1: want to actually get them out more closer to Monday 5 00:00:30,227 --> 00:00:35,387 S1: or Tuesday, ideally Monday, so that the podcast and any 6 00:00:35,387 --> 00:00:39,507 S1: videos and basically this newsletter and the podcast version of 7 00:00:39,507 --> 00:00:42,027 S1: the newsletter comes out as close to Monday as possible. 8 00:00:42,627 --> 00:00:44,507 S1: So I'm going to be doing a number of things 9 00:00:44,507 --> 00:00:46,147 S1: to try to speed that up. But one of those 10 00:00:46,147 --> 00:00:51,346 S1: things is actually having less editing on this podcast, especially 11 00:00:51,346 --> 00:00:53,627 S1: when I'm doing the newsletter. I mean, I'm talking straight 12 00:00:53,626 --> 00:00:56,147 S1: for 20, 30 minutes, and if I give this to 13 00:00:56,187 --> 00:00:58,107 S1: an editor, it's going to take them, you know, a 14 00:00:58,107 --> 00:01:01,387 S1: day or two to turn that around. Now, eventually, hopefully soon, 15 00:01:01,387 --> 00:01:04,307 S1: I will be able to find all my mistakes and 16 00:01:04,627 --> 00:01:07,867 S1: clean out the ums and ahs and the restatements of 17 00:01:07,867 --> 00:01:10,107 S1: a sentence. Because what I'll often do is like, make 18 00:01:10,107 --> 00:01:12,027 S1: a mistake in a sentence and then I'll say the 19 00:01:12,027 --> 00:01:15,627 S1: word edit, and then I will say the sentence over again. 20 00:01:15,747 --> 00:01:18,837 S1: So if you hear any of those or you hear me, 21 00:01:18,837 --> 00:01:22,717 S1: just restate the sentence. I know it's annoying. I apologize, 22 00:01:22,717 --> 00:01:24,797 S1: I do not like to hear that in podcasts I 23 00:01:24,837 --> 00:01:28,837 S1: listen to. But if the podcast is high enough signal, 24 00:01:28,837 --> 00:01:31,877 S1: then I'm okay. You know, taking a little bit of that. 25 00:01:31,877 --> 00:01:35,077 S1: And I'm saying, just bear with me because I think 26 00:01:35,077 --> 00:01:36,917 S1: it's more important that we get the content out, and 27 00:01:36,917 --> 00:01:39,797 S1: it's more important that we get the ideas out, and 28 00:01:39,797 --> 00:01:41,437 S1: we are just waiting for I to be able to 29 00:01:41,437 --> 00:01:43,837 S1: automatically clean that up. So I will just drag and drop. 30 00:01:44,037 --> 00:01:46,237 S1: It'll go clean all that up and we ship the 31 00:01:46,237 --> 00:01:49,037 S1: thing and there's no delay whatsoever. So I just wanted 32 00:01:49,037 --> 00:01:53,437 S1: to mention that and sort of preemptively apologize. All right. 33 00:01:53,477 --> 00:01:56,277 S1: First piece here released my new video on where I 34 00:01:56,277 --> 00:01:58,637 S1: think hacking is going. It's called The Future of Hacking 35 00:01:58,637 --> 00:02:01,357 S1: is Context, and it is on YouTube. And I recommend 36 00:02:01,357 --> 00:02:03,357 S1: you go check it out. Got a new essay on 37 00:02:03,357 --> 00:02:07,997 S1: how I see AI affecting education that's on the website. Uh, 38 00:02:07,997 --> 00:02:12,276 S1: new essay on AI job replacement timelines, and a new 39 00:02:12,276 --> 00:02:14,717 S1: essay on my two groups of cyber AI friends where 40 00:02:14,716 --> 00:02:18,257 S1: basically it's crazy. I've got one group of friends who 41 00:02:18,257 --> 00:02:21,497 S1: are like, I can't believe this whole thing is a fad. 42 00:02:21,496 --> 00:02:25,337 S1: This is nothing but billionaires trying to get rich. Like 43 00:02:25,376 --> 00:02:28,537 S1: it's an NFT crypto scheme and it's just complete and 44 00:02:28,536 --> 00:02:31,976 S1: absolute garbage. And you know, I've got one friend, Marcus Hutchins, 45 00:02:31,977 --> 00:02:35,656 S1: who doesn't even believe that AI is intelligence. He just 46 00:02:35,656 --> 00:02:40,537 S1: believes it's a complete scam. The entire industry is broken. 47 00:02:41,936 --> 00:02:46,496 S1: It's not real intelligence. It's not real AI. And it 48 00:02:46,496 --> 00:02:50,337 S1: just completely down on it. And he published a really 49 00:02:50,337 --> 00:02:53,657 S1: cool essay, I think, which I linked to, which talks 50 00:02:53,656 --> 00:02:55,936 S1: about why it's not real intelligence. And in the short 51 00:02:55,936 --> 00:02:59,377 S1: answer or the short summary of it, is the fact 52 00:02:59,376 --> 00:03:01,617 S1: that it knows so much, but it still can't invent 53 00:03:01,656 --> 00:03:07,537 S1: new things. And my concise answer to this, and by 54 00:03:07,536 --> 00:03:11,177 S1: the way, I'm about to debate him in public in 55 00:03:11,177 --> 00:03:15,216 S1: a recorded podcast on both of our shows, and we're 56 00:03:15,216 --> 00:03:18,267 S1: going to go through this in depth. But the short 57 00:03:18,267 --> 00:03:24,067 S1: answer to that is your rights. It's not producing real 58 00:03:24,066 --> 00:03:28,267 S1: innovation yet. But my argument is that real innovation, real 59 00:03:28,306 --> 00:03:32,746 S1: net new ideas is actually a tiny fraction of the 60 00:03:32,746 --> 00:03:38,027 S1: intelligence that humans use every day. You know, 99.99% of 61 00:03:38,027 --> 00:03:42,587 S1: intelligence that we use every day is not inventing net 62 00:03:42,627 --> 00:03:46,907 S1: new things. It's applying existing knowledge that we have existing 63 00:03:46,907 --> 00:03:50,067 S1: experience that we have, using our intelligence to apply that 64 00:03:50,067 --> 00:03:52,547 S1: to a new problem. And the new problem is also 65 00:03:52,547 --> 00:03:58,067 S1: not a net new 99.99% problem. It's just a slightly 66 00:03:58,067 --> 00:04:00,747 S1: different version of that problem. Right. If you think about 67 00:04:01,267 --> 00:04:03,747 S1: an average white collar worker, if you think about an 68 00:04:03,747 --> 00:04:11,387 S1: average admin or secretary person or a an average coder even. Right. 69 00:04:11,427 --> 00:04:14,266 S1: Usually the problems are like very similar to problems we've 70 00:04:14,267 --> 00:04:18,207 S1: solved before, right? And there are some exceptions with with 71 00:04:18,207 --> 00:04:21,447 S1: coding and with high end engineering and of course, with, 72 00:04:21,487 --> 00:04:25,327 S1: you know, pinnacle level science. There's lots of exceptions, but 73 00:04:25,727 --> 00:04:28,166 S1: the vast majority of the intelligence that we use every 74 00:04:28,167 --> 00:04:31,167 S1: single day, it's all the same stuff with different names 75 00:04:31,167 --> 00:04:35,007 S1: and slightly different configurations. It's all the same stuff, right? 76 00:04:35,047 --> 00:04:39,967 S1: And more importantly, teaching kids answering their questions, being a tutor, 77 00:04:40,287 --> 00:04:44,167 S1: being a teacher, being a helper, you know, being able 78 00:04:44,167 --> 00:04:47,487 S1: to help out inside of companies and be an assistant and, 79 00:04:47,807 --> 00:04:51,487 S1: you know, make meetings happen and coordinate and summarize things 80 00:04:51,647 --> 00:04:54,647 S1: and keep everyone up to date. This is what the 81 00:04:54,647 --> 00:04:59,527 S1: vast majority of work is, right? And I would argue 82 00:04:59,567 --> 00:05:04,567 S1: that's absolutely intelligence. It's absolutely human intelligence. And we know 83 00:05:04,567 --> 00:05:08,167 S1: this because we can't automate it. That's the reason it's 84 00:05:08,167 --> 00:05:13,247 S1: AI is because. If if we could automate it away, 85 00:05:13,247 --> 00:05:15,287 S1: it would have already been scripted and we wouldn't actually 86 00:05:15,287 --> 00:05:18,897 S1: have human employees doing that work. So if you think 87 00:05:18,897 --> 00:05:21,937 S1: about the average job that someone does in an intellectual 88 00:05:21,937 --> 00:05:27,337 S1: or cognitive role, like knowledge work, 99.9% of it, and 89 00:05:27,337 --> 00:05:30,177 S1: I'm just making up that number to, you know, to 90 00:05:30,217 --> 00:05:35,137 S1: be a little bit hyperbolic there. 99.99% of that work 91 00:05:35,337 --> 00:05:40,377 S1: is extremely repetitive. It's just slightly changing the work that 92 00:05:40,377 --> 00:05:44,137 S1: we do, right? It's slightly different scenarios, slightly different users, 93 00:05:44,137 --> 00:05:50,537 S1: slightly different use cases. But ultimately what this comes down 94 00:05:50,537 --> 00:05:54,297 S1: to is can I replace that? And I think we're 95 00:05:54,297 --> 00:05:56,657 S1: already seeing that. So I think that's empirically being shown 96 00:05:56,657 --> 00:05:59,897 S1: to be the case. Now Marcus's argument here is well, 97 00:05:59,897 --> 00:06:04,097 S1: that's not real intelligence. That's not real AI because it's 98 00:06:04,097 --> 00:06:06,377 S1: just doing the same thing over and over. And my 99 00:06:06,377 --> 00:06:09,337 S1: point is, that's what millions of jobs are actually doing. 100 00:06:09,697 --> 00:06:12,617 S1: That's why millions of jobs are actually going to get replaced, 101 00:06:12,857 --> 00:06:15,457 S1: is because AI is going to be able to do that. 102 00:06:16,077 --> 00:06:21,117 S1: 0.9 9% better and faster and much cheaper than most humans. 103 00:06:21,117 --> 00:06:24,277 S1: So my whole thing with AI is that it matters 104 00:06:24,277 --> 00:06:29,317 S1: because of how it impacts humans, right? Getting rid of 105 00:06:29,317 --> 00:06:33,357 S1: millions of jobs affects humans. That's why I define AGI 106 00:06:33,397 --> 00:06:37,757 S1: as an AI that can replace an average knowledge worker, 107 00:06:38,037 --> 00:06:41,197 S1: because I don't care about AI in a theoretical sense. 108 00:06:41,477 --> 00:06:43,277 S1: I don't care about it in a technical sense. I 109 00:06:43,317 --> 00:06:47,277 S1: care about it mostly in how it impacts humans and 110 00:06:47,277 --> 00:06:51,277 S1: human society and human meaning. So that's what our debate 111 00:06:51,277 --> 00:06:54,717 S1: is going to come around. And anyway, the way I 112 00:06:54,717 --> 00:06:56,957 S1: got on to this is that that's one view. It's 113 00:06:56,957 --> 00:07:00,957 S1: like this very anti-ai, very negative view of like, we 114 00:07:00,957 --> 00:07:03,397 S1: don't even have it. It's not real. It's all hype. 115 00:07:03,397 --> 00:07:05,517 S1: It's just like crypto. It's just a bunch of billionaires 116 00:07:05,517 --> 00:07:09,476 S1: trying to get rich. All these stats and benchmarks, everything 117 00:07:09,477 --> 00:07:14,517 S1: being said is basically marketing from executives for these lame, stupid, 118 00:07:14,557 --> 00:07:18,447 S1: temporary AI companies that are going to be gone immediately. 119 00:07:18,447 --> 00:07:21,447 S1: And basically it's not going to have much job impact 120 00:07:21,447 --> 00:07:23,967 S1: because it can't actually replace a human. So that's kind 121 00:07:23,967 --> 00:07:26,207 S1: of their argument. I'm not lumping everything I just said 122 00:07:26,207 --> 00:07:29,287 S1: into Marcuse's argument. I'll let him make his own arguments 123 00:07:29,287 --> 00:07:34,247 S1: when we have that discussion. But that's one group generally now. 124 00:07:34,247 --> 00:07:37,527 S1: The other group is like myself and Joseph Thacker and 125 00:07:37,807 --> 00:07:42,047 S1: my friend Jason Haddox and my friend Sasha Geller. Like, 126 00:07:42,047 --> 00:07:46,007 S1: we're all thinking, you know, we kind of saw where 127 00:07:46,007 --> 00:07:49,567 S1: this was going, especially myself and Joseph Thacker and Sasha. 128 00:07:50,247 --> 00:07:55,407 S1: It's like end of 22. We're like, oh my goodness. Right. 129 00:07:55,447 --> 00:07:57,767 S1: So all the stuff I've been talking about for the last, 130 00:07:58,087 --> 00:08:02,927 S1: you know, two and a half, three years is we 131 00:08:02,967 --> 00:08:06,287 S1: I kind of saw the glimpses of that already happening 132 00:08:06,287 --> 00:08:09,167 S1: at the end of 2022. It was kind of inevitable. 133 00:08:09,167 --> 00:08:10,927 S1: And the more the more the AI started to happen, 134 00:08:10,927 --> 00:08:13,047 S1: the more obvious it gets or whatever. And of course, 135 00:08:13,047 --> 00:08:15,047 S1: it's part of the book. That was in 2016 as well. 136 00:08:15,047 --> 00:08:19,897 S1: But the point is, one group of people just still 137 00:08:19,937 --> 00:08:24,337 S1: today in mid 2025 doesn't accept it. Other another group 138 00:08:24,337 --> 00:08:26,737 S1: of people like myself and a bunch of other people 139 00:08:26,737 --> 00:08:30,297 S1: just like me or similar to me in mindset. They 140 00:08:30,297 --> 00:08:33,937 S1: got it instantly. Or if I'm right, they got it instantly. 141 00:08:33,937 --> 00:08:36,257 S1: If I'm wrong, then they were also instantly wrong, just 142 00:08:36,256 --> 00:08:39,656 S1: like me. And what's fascinating to me is these are 143 00:08:39,657 --> 00:08:44,617 S1: all security experts. These are all deeply technical, deeply experienced, 144 00:08:44,617 --> 00:08:49,697 S1: multiple decades of experience in cybersecurity, and we have drastically 145 00:08:49,697 --> 00:08:54,337 S1: different views on this. That's what's so fascinating to me, right? Why? 146 00:08:55,136 --> 00:08:57,217 S1: Why do they think that way? And why do we 147 00:08:57,217 --> 00:09:00,577 S1: think this way? What makes the difference between being in 148 00:09:00,577 --> 00:09:02,456 S1: one group or the other? And I think the answer 149 00:09:02,457 --> 00:09:06,857 S1: comes down to just how they overall view the world. Um, 150 00:09:06,857 --> 00:09:11,016 S1: Marcus in particular is very, I want to say socialist, 151 00:09:11,016 --> 00:09:13,457 S1: but not not socialist in a, in a bad way, 152 00:09:13,497 --> 00:09:18,837 S1: socialist in a human uplifting way, which I am as well, right? 153 00:09:18,877 --> 00:09:23,597 S1: I just I'm also a capitalist. So it, uh. Complex topic. 154 00:09:23,597 --> 00:09:25,597 S1: I'm not going to go into that, but the point 155 00:09:25,597 --> 00:09:29,157 S1: is he is very sort of anti billionaire anti-big business. 156 00:09:29,156 --> 00:09:32,036 S1: He sees all this as a giant scam, a big, uh, 157 00:09:32,396 --> 00:09:36,277 S1: you know, a big push to try to just, uh, 158 00:09:36,317 --> 00:09:39,396 S1: remove jobs or trick people into removing jobs when it's 159 00:09:39,396 --> 00:09:42,357 S1: not even real. He's just very cynical about the entire 160 00:09:42,357 --> 00:09:47,797 S1: thing based on, you know, kind of being anti-capitalist, anti, 161 00:09:47,997 --> 00:09:51,436 S1: you know, big business anti billionaire for sure. So I 162 00:09:51,437 --> 00:09:54,157 S1: think people who have that view or I think the 163 00:09:54,156 --> 00:09:57,037 S1: other view that that puts people in this category often 164 00:09:57,036 --> 00:10:00,477 S1: is just being anti-change. And a lot of people in 165 00:10:00,957 --> 00:10:04,957 S1: information security are actually anti-change. I didn't realize this until 166 00:10:04,957 --> 00:10:07,157 S1: a few years ago, but it's like there are many 167 00:10:07,156 --> 00:10:09,836 S1: people who are just like, oh, don't change that. You know, 168 00:10:09,877 --> 00:10:11,877 S1: when I was growing up in the 80s or 90s, 169 00:10:11,916 --> 00:10:14,117 S1: you know, the protocol worked like this and that's how 170 00:10:14,117 --> 00:10:17,286 S1: it should always work and can't stand these these crazy 171 00:10:17,286 --> 00:10:19,686 S1: kids with all their JavaScript and all their dynamic web 172 00:10:19,727 --> 00:10:23,367 S1: apps and, you know, all the crazy APIs. Like, I 173 00:10:23,406 --> 00:10:26,207 S1: just see a lot of get off my lawn type mentality. 174 00:10:26,406 --> 00:10:29,006 S1: And people who have that mentality tend to be very 175 00:10:29,247 --> 00:10:32,487 S1: anti-ai in this way. And then my other group of 176 00:10:32,487 --> 00:10:35,167 S1: friends are like, yeah, tech changes all the time, no 177 00:10:35,166 --> 00:10:38,287 S1: big deal. And they kind of have this mentality that 178 00:10:38,286 --> 00:10:41,087 S1: I have, which is shepherding. So I see myself as 179 00:10:41,087 --> 00:10:44,367 S1: a shepherd, you know, with like the stick and the robes. 180 00:10:44,766 --> 00:10:47,407 S1: And I don't want to use sheep here because I 181 00:10:47,406 --> 00:10:51,247 S1: don't consider all of humanity to be sheep, but whatever, 182 00:10:51,247 --> 00:10:54,086 S1: we'll use sheep. It's not actually pertinent in this. The 183 00:10:54,087 --> 00:10:57,847 S1: idea is you. It's your job to walk up ahead 184 00:10:58,247 --> 00:11:01,247 S1: and see if there's danger and to, like, help steer people. 185 00:11:01,567 --> 00:11:05,927 S1: Because in this metaphor, it's not superiority of the human 186 00:11:05,926 --> 00:11:10,406 S1: to the inferiority of the sheep. It's more like superiority 187 00:11:10,406 --> 00:11:14,327 S1: of your knowledge of the risks versus inferiority, of the 188 00:11:14,367 --> 00:11:17,987 S1: knowledge of the risks. So it's my job. Switching to 189 00:11:18,026 --> 00:11:19,826 S1: a military metaphor, it's my job to go out there 190 00:11:19,827 --> 00:11:22,026 S1: and step on mines. It's my job to go out 191 00:11:22,026 --> 00:11:24,947 S1: there and detect all the mines, and then spray paint 192 00:11:24,987 --> 00:11:28,347 S1: a path on this minefield so people can walk out, 193 00:11:28,386 --> 00:11:31,427 S1: walk on it without getting hurt. Right. That's the way 194 00:11:31,426 --> 00:11:34,186 S1: I see it. And I see the this minefield is 195 00:11:34,187 --> 00:11:37,227 S1: constantly changing because the tech is constantly changing, and we 196 00:11:37,227 --> 00:11:40,426 S1: should just accept that and everything's fine, right? Change is constant. 197 00:11:40,426 --> 00:11:42,427 S1: That's fine. Oh, we're going to lose all these jobs. Cool. 198 00:11:42,426 --> 00:11:45,627 S1: What's next? Um, AI is going to replace a whole 199 00:11:45,627 --> 00:11:48,427 S1: bunch of stuff. Cool. What's next? Let's talk about the upsides. Oh, 200 00:11:48,467 --> 00:11:50,906 S1: we're all going to become way more creative. Okay, cool. 201 00:11:50,906 --> 00:11:53,146 S1: Let's talk about that. Let's figure out how to uplevel 202 00:11:53,187 --> 00:11:56,587 S1: and upskill. So I think that mentality versus the other 203 00:11:56,587 --> 00:11:59,546 S1: anti-change mentality is the primary thing that's guiding whether or 204 00:11:59,546 --> 00:12:03,386 S1: not someone is in one of these camps versus the other. So, um, 205 00:12:04,227 --> 00:12:07,107 S1: that's that's essentially what that essay is about. And what 206 00:12:07,107 --> 00:12:09,146 S1: I just said is like 20 times longer than the 207 00:12:09,146 --> 00:12:13,026 S1: actual essay, which is like a one minute read. But 208 00:12:13,026 --> 00:12:17,156 S1: that's the advantage of doing a long form podcast. All right. 209 00:12:18,276 --> 00:12:23,516 S1: Let's see here. Gukesh Indian current world champion beat Magnus 210 00:12:23,516 --> 00:12:26,556 S1: Carlsen in the for the first time in classical chess. 211 00:12:26,557 --> 00:12:30,717 S1: And Magnus like hammer fisted the table. Crore. Super super loud. 212 00:12:31,117 --> 00:12:33,036 S1: Gukesh didn't jump, so I gave him credit for that. 213 00:12:33,036 --> 00:12:38,036 S1: But that was cool. Got a video of that. All right. Cybersecurity. 214 00:12:38,156 --> 00:12:41,636 S1: Google patches new Chrome zero day bug in attacks Microsoft 215 00:12:41,636 --> 00:12:45,877 S1: and CrowdStrike create a shared threat actor dictionary. This one's 216 00:12:45,877 --> 00:12:48,517 S1: cool because some threat actors have like 15 different names 217 00:12:48,516 --> 00:12:52,836 S1: for them, so they unified that. That's, I think, progress. 218 00:12:53,276 --> 00:12:56,877 S1: OpenAI's O3 discovers Linux kernel zero day vulnerability I talked 219 00:12:56,877 --> 00:12:59,877 S1: about this last week, but, um, yeah, this article is 220 00:12:59,877 --> 00:13:03,117 S1: pretty good here. Meta plans to automate product risk assessments 221 00:13:03,117 --> 00:13:06,437 S1: with AI. This is really cool. 90% of app updates 222 00:13:06,557 --> 00:13:09,517 S1: they're going to do the assessments using AI. This has 223 00:13:09,516 --> 00:13:11,477 S1: always been one of my favorite things. I did this 224 00:13:11,477 --> 00:13:17,647 S1: very early, um, when I first started to pop, um, 225 00:13:17,646 --> 00:13:21,326 S1: a filtering system for a big problem with security teams 226 00:13:21,327 --> 00:13:23,647 S1: is they are the bottleneck between engineering and being able 227 00:13:23,646 --> 00:13:25,487 S1: to ship and not being able to ship. So they 228 00:13:25,487 --> 00:13:27,886 S1: get really mad at security about that. I love the 229 00:13:27,886 --> 00:13:29,886 S1: idea of having all the context of the app that 230 00:13:29,886 --> 00:13:33,647 S1: they're trying to push through and sorting it into, like buckets. 231 00:13:33,646 --> 00:13:35,487 S1: You could actually use lots of different buckets. You could 232 00:13:35,487 --> 00:13:37,806 S1: use like 2 or 3, but you might be able 233 00:13:37,807 --> 00:13:41,367 S1: to sort it into like 15 different filtering buckets where 234 00:13:41,607 --> 00:13:44,886 S1: your pipeline of security checks is vastly different based on okay, 235 00:13:44,926 --> 00:13:48,686 S1: is it privacy focused? Is it customer data security focused? 236 00:13:48,687 --> 00:13:52,927 S1: Is it proprietary, data focused? Like what are what's the 237 00:13:52,926 --> 00:13:55,927 S1: threat model, which it could automatically do as well. It 238 00:13:55,926 --> 00:13:58,367 S1: could understand your risks of the company. It can understand 239 00:13:58,367 --> 00:14:00,967 S1: worst case scenarios for the company. And let's say it's 240 00:14:01,087 --> 00:14:06,087 S1: sorting into 16 different granular security check types. Hopefully most 241 00:14:06,087 --> 00:14:08,406 S1: of those being automated. But if it's a super high 242 00:14:08,406 --> 00:14:10,687 S1: priority thing it goes into the top one. It goes 243 00:14:10,687 --> 00:14:13,527 S1: into number 16. And that one is actually a manual 244 00:14:13,526 --> 00:14:16,666 S1: red team pen test, which is associated with automated scans 245 00:14:16,666 --> 00:14:19,467 S1: that already happened or whatever. But imagine being able to 246 00:14:19,467 --> 00:14:26,107 S1: speed up 85%, 95%, 98% of projects that engineering is 247 00:14:26,107 --> 00:14:30,027 S1: trying to push out for the company and have only 2% 248 00:14:30,027 --> 00:14:34,907 S1: actually involve manual testing, um, or whatever. I think that's really, 249 00:14:34,907 --> 00:14:37,667 S1: really powerful use case for this. And anyway, so meta 250 00:14:37,667 --> 00:14:40,667 S1: is doing that or something like that for 90% of 251 00:14:40,667 --> 00:14:46,547 S1: app updates using AI. Massive Asus router botnet uses persistent backdoors. 252 00:14:46,587 --> 00:14:50,587 S1: Over 8000 Asus routers affected by this, and it survives updates. 253 00:14:50,587 --> 00:14:52,947 S1: So if you reboot the router, the thing is still there. 254 00:14:53,427 --> 00:14:57,307 S1: Russian market becomes top destination for stolen credentials. So this 255 00:14:57,307 --> 00:15:01,187 S1: is the Russian market cybercrime platform. That's the name of it. 256 00:15:01,467 --> 00:15:04,987 S1: And it's filling the gap left by Genesis Markets takedown 257 00:15:05,547 --> 00:15:08,627 S1: DOJ takes down four major services using that were used 258 00:15:08,627 --> 00:15:13,547 S1: by cyber criminals. So it's a basically like URL hiding 259 00:15:13,627 --> 00:15:18,477 S1: uh infrastructure China linked hackers exploit SAP and SQL server 260 00:15:18,477 --> 00:15:21,397 S1: flaws in attacks across Asia and Brazil. And this actor 261 00:15:21,397 --> 00:15:27,117 S1: is called Earth. Lamia. Lamia and they're hitting multiple organizations 262 00:15:27,157 --> 00:15:33,637 S1: since 2023. National security. Ukraine hides explosive drones in wooden 263 00:15:33,637 --> 00:15:37,237 S1: sheds to hit parked Russian bombers. This one is absolutely insane. 264 00:15:37,237 --> 00:15:40,157 S1: I'm sure you've seen stuff on it already. It's. It 265 00:15:40,157 --> 00:15:42,997 S1: reminds me of what actually happened with the Israeli attacks 266 00:15:42,997 --> 00:15:46,877 S1: on Hezbollah, with this ingenious, sort of like personalized little 267 00:15:46,877 --> 00:15:50,077 S1: mini bombs or whatever that went after leadership. So it 268 00:15:50,077 --> 00:15:53,117 S1: wasn't like giant drone strikes. It wasn't like giant bombing raids. 269 00:15:53,117 --> 00:15:57,357 S1: It was these personalized little things against leadership, which, uh, 270 00:15:57,797 --> 00:16:00,836 S1: a lot of people are saying, you know, tactical win here. Uh, 271 00:16:00,837 --> 00:16:03,437 S1: and it was pretty effective. And I think this is 272 00:16:03,437 --> 00:16:09,837 S1: similar in the sense that it's asymmetrical use of force. Um, 273 00:16:10,517 --> 00:16:14,956 S1: you have Ukraine, which is out outmanned, outgunned. And they 274 00:16:14,997 --> 00:16:20,696 S1: snuck these drones, uh, over a thousand miles inside of Russia. 275 00:16:20,737 --> 00:16:23,777 S1: I think maybe even close to 2000. 3000 miles in 276 00:16:23,777 --> 00:16:30,056 S1: some cases. Um, snuck them in, and, um, basically, they're 277 00:16:30,057 --> 00:16:32,857 S1: inside of Russia. They they get all the way to 278 00:16:32,897 --> 00:16:37,417 S1: the airfield. They remotely open them. They're now sitting on Russian, uh, 279 00:16:37,457 --> 00:16:40,977 S1: cell networks, uh, which is what's used to activate them 280 00:16:40,977 --> 00:16:45,177 S1: and control them. And they, um, they hit these Russian bombers, 281 00:16:45,177 --> 00:16:48,337 S1: and these Russian bombers cannot be reproduced. Russia does not 282 00:16:48,337 --> 00:16:51,417 S1: have the ability to make these bombers again. And what's 283 00:16:51,417 --> 00:16:56,457 S1: interesting about this is they didn't diminish Russian bombing against Ukraine. 284 00:16:56,457 --> 00:17:03,257 S1: They diminished Russian bombing capabilities completely strategically as compared to 285 00:17:03,257 --> 00:17:06,936 S1: the United States as compared to China. So they significantly 286 00:17:06,937 --> 00:17:11,216 S1: damaged Russia's capabilities, right? Not to mention, you know, the 287 00:17:11,337 --> 00:17:15,297 S1: hundreds of thousands of actual troops they've lost. Um, but 288 00:17:15,387 --> 00:17:21,067 S1: this this war is seriously, seriously harming Russia's overall strategic 289 00:17:21,067 --> 00:17:25,267 S1: military capabilities. And, uh, the thing to call out here 290 00:17:25,267 --> 00:17:31,306 S1: is just how much Ukraine has changed warfare using drones. 291 00:17:31,787 --> 00:17:35,667 S1: And again, I implore you to go read Kill Decision 292 00:17:35,667 --> 00:17:40,307 S1: by Daniel Suarez, which talks about specifically autonomous drones, which 293 00:17:40,307 --> 00:17:43,907 S1: can't be taken out by GPS blockers or Em blockers 294 00:17:44,387 --> 00:17:48,147 S1: because they could do everything, uh, self-sufficient inside the drone itself. 295 00:17:50,027 --> 00:17:53,867 S1: China's deep network penetration signals war preparations. So former Trump 296 00:17:53,867 --> 00:17:57,506 S1: adviser H.R. McMaster told lawmakers that China's extensive hacking of 297 00:17:57,547 --> 00:18:02,466 S1: US infrastructure is actually them preparing for kinetic war. FBI 298 00:18:02,466 --> 00:18:06,427 S1: arrests defense intelligence IT worker for Pak drop espionage. So 299 00:18:06,547 --> 00:18:08,907 S1: he basically went to go do one of these, you know, 300 00:18:08,946 --> 00:18:13,506 S1: spy versus spy dead drops in a park in Virginia. And, uh, 301 00:18:13,867 --> 00:18:18,847 S1: the guy was actually part of DIA's internal threat division. So? 302 00:18:18,887 --> 00:18:21,567 S1: So he's part of the team designed to help people 303 00:18:21,567 --> 00:18:27,686 S1: find or design to find spies inside of Dia. And 304 00:18:27,686 --> 00:18:31,007 S1: he was one of these spies, and he dropped this material, 305 00:18:31,007 --> 00:18:34,446 S1: and it was actually dropped to an FBI agent and 306 00:18:34,486 --> 00:18:38,686 S1: got all the write ups there. I Mary Meeker returns 307 00:18:38,686 --> 00:18:42,367 S1: with First Trends report since 2019. Definitely go check this out. 308 00:18:42,367 --> 00:18:45,647 S1: She is a beast of analysis and she's been dormant 309 00:18:45,647 --> 00:18:49,727 S1: for a while. And now she's back. And this analysis 310 00:18:49,767 --> 00:18:52,687 S1: is like 300 something pages. Got a decent summary here 311 00:18:52,686 --> 00:18:55,927 S1: in the newsletter, but go check it out. Dwarkesh Patel 312 00:18:55,966 --> 00:18:59,006 S1: has longer AGI timelines than his podcast guests. I think 313 00:18:59,007 --> 00:19:02,327 S1: he's wrong about this, I think. So basically what Dwarkesh 314 00:19:02,367 --> 00:19:05,327 S1: is saying is that AI is not smart enough to 315 00:19:06,047 --> 00:19:10,367 S1: do continuous learning. And I think I'm not sure about this, 316 00:19:10,367 --> 00:19:13,686 S1: but I think he thinks that's an IQ problem, whereas 317 00:19:13,686 --> 00:19:16,257 S1: I see it as a scaffolding and system problem. It's 318 00:19:16,257 --> 00:19:19,417 S1: a memory problem. It's a context size problem. It's a 319 00:19:19,456 --> 00:19:21,736 S1: even more than a context size problem for the model. 320 00:19:21,736 --> 00:19:26,137 S1: It's more of a context knowledge base management problem. So 321 00:19:26,137 --> 00:19:28,617 S1: this is all scaffolding to me. And I think this 322 00:19:28,617 --> 00:19:31,377 S1: scaffolding is going to improve as fast or faster than 323 00:19:31,377 --> 00:19:36,216 S1: the IQ of the models themselves. So I think this 324 00:19:36,216 --> 00:19:38,977 S1: is not as big of a hurdle as Dwarkesh is 325 00:19:38,976 --> 00:19:43,217 S1: thinking it is. McKinsey says the future of work is Agentic. 326 00:19:43,257 --> 00:19:48,177 S1: They released this big article talking about becoming digital workers 327 00:19:48,177 --> 00:19:50,736 S1: that can think, decide and execute tasks on their own 328 00:19:51,417 --> 00:19:54,096 S1: and that it's worth a read. You should go check 329 00:19:54,097 --> 00:19:56,217 S1: it out. It's a it's a good article by McKinsey. 330 00:19:56,976 --> 00:20:01,417 S1: The truth about AI and job loss. Neruda from meta 331 00:20:01,417 --> 00:20:04,456 S1: dug into historical data to find out which AI, which 332 00:20:04,456 --> 00:20:08,617 S1: jobs AI will actually eliminate, and whether there's still room 333 00:20:08,617 --> 00:20:12,137 S1: for junior developers. That was a good read. Google Gemini 334 00:20:12,177 --> 00:20:16,307 S1: integration with Siri could fill Apple's personal context gap. I 335 00:20:16,347 --> 00:20:18,787 S1: think this analysis is wrong. I included it because it's 336 00:20:18,787 --> 00:20:22,746 S1: an alternative view to my own, but I think this 337 00:20:22,747 --> 00:20:26,027 S1: analysis is wrong. The problem is not Apple's access to 338 00:20:26,667 --> 00:20:29,627 S1: personal data. They have the best. They've been building iOS 339 00:20:29,667 --> 00:20:32,067 S1: all of this time. They have the best personal contacts 340 00:20:32,067 --> 00:20:36,147 S1: on people, I believe, on the planet. The problem is 341 00:20:36,147 --> 00:20:38,987 S1: having Siri prompt injected while it has access to all 342 00:20:38,986 --> 00:20:43,107 S1: that context. I don't think they know how to give Siri, 343 00:20:43,107 --> 00:20:45,427 S1: or whatever the super agent is going to be is 344 00:20:45,427 --> 00:20:48,227 S1: probably still going to be Siri. They don't know how 345 00:20:48,226 --> 00:20:50,667 S1: to give Siri access to all that data in a 346 00:20:50,667 --> 00:20:55,027 S1: secure way, in a comprehensive, powerful way that is secure. 347 00:20:55,027 --> 00:20:57,546 S1: I think that is what is stopping them. That's my guess. 348 00:20:57,547 --> 00:20:59,427 S1: I don't know that for sure. Not talking to people 349 00:20:59,427 --> 00:21:01,706 S1: on the inside about it, but that is my guess. 350 00:21:01,706 --> 00:21:05,067 S1: Based on working in security at Apple for like three years. 351 00:21:05,907 --> 00:21:10,186 S1: Snowflake buys crunchy data for $250 million. And this is snowflake, 352 00:21:10,186 --> 00:21:13,547 S1: which is a giant data platform making an AI agents 353 00:21:13,547 --> 00:21:17,927 S1: play Technology. McKinsey uses AI to automate PowerPoint creation and 354 00:21:17,927 --> 00:21:23,407 S1: proposal writing. So their platform Lily now handles PowerPoint creation 355 00:21:23,407 --> 00:21:29,446 S1: and proposal drafting. Over 75% of people using it monthly, internally, employees. 356 00:21:29,927 --> 00:21:32,047 S1: Workday plans to rehire the same number of people they 357 00:21:32,047 --> 00:21:34,127 S1: laid off, but with different skills. This is what I've 358 00:21:34,127 --> 00:21:38,847 S1: been saying. They want completely different people, completely different people. 359 00:21:40,087 --> 00:21:43,887 S1: Nvidia develops new AI chip for China that meets export controls. 360 00:21:44,567 --> 00:21:47,367 S1: I don't think this stuff matters that much. I think 361 00:21:47,927 --> 00:21:50,087 S1: if China gets less chips, they'll learn how to make 362 00:21:50,087 --> 00:21:52,647 S1: more AI with less chips, and in the meantime, they're 363 00:21:52,647 --> 00:21:54,807 S1: making their own chips. In fact, they just released one 364 00:21:54,807 --> 00:21:58,527 S1: that performs really well. I think ultimately China is going 365 00:21:58,527 --> 00:22:01,127 S1: to get everything that we have. I think ultimately, most 366 00:22:01,127 --> 00:22:04,367 S1: AI advancements are happening at the software layer via what 367 00:22:04,367 --> 00:22:07,127 S1: I used to call in 2023 tricks, a whole bunch 368 00:22:07,127 --> 00:22:10,726 S1: of different tricks that they're using. Right. And those tricks 369 00:22:10,726 --> 00:22:14,246 S1: I think get leaked, I think they get stolen. And 370 00:22:14,777 --> 00:22:18,457 S1: so for example, there's an argument that's something that OpenAI 371 00:22:18,497 --> 00:22:22,456 S1: was doing with oh one got stolen by deep sea 372 00:22:23,216 --> 00:22:25,857 S1: and deep sea copied it. And guess what? Their R1 373 00:22:25,857 --> 00:22:29,337 S1: model was nearly as good as oh one. After OpenAI 374 00:22:29,377 --> 00:22:32,696 S1: spent all that money and deep sea spent way less. 375 00:22:32,736 --> 00:22:35,617 S1: That's an example of a trick or a set of 376 00:22:35,617 --> 00:22:38,737 S1: tricks getting stolen. I think that's going to continue to happen. 377 00:22:39,057 --> 00:22:42,817 S1: And it's why not only that, the pinnacle model makers 378 00:22:42,817 --> 00:22:45,497 S1: are going to stay close to each other, but also 379 00:22:45,497 --> 00:22:47,737 S1: open source is going to stay close to the pinnacle 380 00:22:47,736 --> 00:22:50,537 S1: model makers. So I think this whole thing just kind 381 00:22:50,537 --> 00:22:58,337 S1: of churns along, chugs along, advances, slowly rolls forward with them, 382 00:22:58,337 --> 00:23:03,936 S1: largely similar with them largely staying neck and neck. Not always. 383 00:23:03,936 --> 00:23:06,657 S1: There's going to be these these big jumps, right? But 384 00:23:06,657 --> 00:23:10,337 S1: I think that thing will be copied. What I thought 385 00:23:10,337 --> 00:23:12,817 S1: was would happen, and what I think most people thought 386 00:23:12,817 --> 00:23:16,476 S1: would happen is when somebody made a major advancement, like 387 00:23:16,476 --> 00:23:19,837 S1: OpenAI or anthropic or someone in China. They would be 388 00:23:19,837 --> 00:23:22,117 S1: a leader for six months, a year or two years. 389 00:23:22,117 --> 00:23:25,236 S1: They would just be like, oh, they did something crazy, 390 00:23:25,236 --> 00:23:27,236 S1: and no one can find out what it is. And 391 00:23:27,277 --> 00:23:30,917 S1: they're just way ahead. And everyone's developing like independently in 392 00:23:30,917 --> 00:23:34,077 S1: these silos. But it turns out so many of these 393 00:23:34,077 --> 00:23:36,557 S1: researchers know each other. So many of these researchers are 394 00:23:36,557 --> 00:23:41,436 S1: moving between the model maker companies. So many of them 395 00:23:41,476 --> 00:23:45,797 S1: talk and share their stuff in coffee shops and in 396 00:23:45,797 --> 00:23:51,397 S1: conferences and stuff. These secrets, these tricks, these scaffolding things, 397 00:23:51,397 --> 00:23:53,196 S1: because a lot of this is not about the pre-training 398 00:23:53,196 --> 00:23:56,037 S1: of the model. It's about what you do after Post-training 399 00:23:56,677 --> 00:23:59,317 S1: where a lot of these tricks come into play and 400 00:23:59,317 --> 00:24:01,757 S1: it's just being spread. It's just being leaked. It's just 401 00:24:01,757 --> 00:24:04,597 S1: being shared so much that I think we're going to 402 00:24:04,597 --> 00:24:08,197 S1: have a lot of parity with, with some exceptions, uh, 403 00:24:08,277 --> 00:24:10,997 S1: going forward. And I think the only thing that changes 404 00:24:10,997 --> 00:24:14,276 S1: this and gets really weird is when you cross through AGI, 405 00:24:15,007 --> 00:24:17,887 S1: where I think that still maintains what I was saying. 406 00:24:18,686 --> 00:24:21,367 S1: But when it hits ASI, that's where you're going to 407 00:24:21,367 --> 00:24:24,446 S1: have these giant jumps. And I don't know necessarily. Well, 408 00:24:24,486 --> 00:24:26,327 S1: no one knows what that's going to look like because 409 00:24:26,327 --> 00:24:30,486 S1: it's going to be ridiculous. Computer science unemployment hits six 1% 410 00:24:30,486 --> 00:24:35,647 S1: despite Major's popularity. So a lot of questions of like, 411 00:24:35,887 --> 00:24:38,567 S1: is computer science still a good thing to go into? Um, 412 00:24:38,726 --> 00:24:48,167 S1: I think dialectic, I think studying, um, philosophy, um, rhetoric, dialectic, um, 413 00:24:48,167 --> 00:24:53,486 S1: how to think clearly, studying history, studying computer science, um, 414 00:24:53,486 --> 00:24:56,966 S1: studying physics. I think these are the core things that 415 00:24:56,966 --> 00:24:59,407 S1: people should actually be studying. And you should be getting 416 00:24:59,407 --> 00:25:03,087 S1: your kids and your younger young ones, or even people 417 00:25:03,087 --> 00:25:06,407 S1: who are advanced in their careers. This is how you retool. 418 00:25:06,407 --> 00:25:08,287 S1: And this is what I'm going to be putting into 419 00:25:08,287 --> 00:25:12,607 S1: the curriculum for H3 for human 3.0. Uh, this curriculum, 420 00:25:12,607 --> 00:25:15,857 S1: I think, is the core sort of first principles type 421 00:25:15,897 --> 00:25:18,056 S1: education that you're going to need. And of course, you 422 00:25:18,057 --> 00:25:20,217 S1: can specialize in all the stuff that you want like 423 00:25:20,537 --> 00:25:23,497 S1: that you're super interested in. But this core, I think, 424 00:25:23,497 --> 00:25:25,737 S1: is what's going to make you resilient to all these 425 00:25:25,736 --> 00:25:31,177 S1: different changes. Humans 60% of Americans have retirement savings accounts, 426 00:25:31,177 --> 00:25:34,857 S1: but it's very lumpy. So 83% of people making over 427 00:25:34,857 --> 00:25:39,097 S1: 100 K have retirement accounts. Only 28% under 50 K 428 00:25:39,216 --> 00:25:42,897 S1: have retirement accounts. There's a massive racial gap. There's also 429 00:25:42,897 --> 00:25:48,016 S1: a massive gap between college graduates and, uh, people who 430 00:25:48,057 --> 00:25:52,736 S1: don't have any college. So 81% versus 39% and white 431 00:25:52,736 --> 00:25:58,897 S1: versus people of color, it's 68% versus 42%. Uh, US 432 00:25:58,897 --> 00:26:03,297 S1: economy contracts, contracts more than expected in Q1. It actually 433 00:26:03,297 --> 00:26:06,776 S1: looks like it contracted by 0.2 percent in Q1. Younger 434 00:26:06,777 --> 00:26:09,857 S1: generations less likely to develop dementia. I think this is 435 00:26:09,857 --> 00:26:15,157 S1: because previous generations they would stop working at 60 or 65, 436 00:26:15,476 --> 00:26:17,277 S1: and they would just be like, oh, I'm just going 437 00:26:17,317 --> 00:26:19,036 S1: to play with grandkids and I don't really need to 438 00:26:19,037 --> 00:26:21,236 S1: do anything. And I think when you tell your brain 439 00:26:21,236 --> 00:26:24,397 S1: to shut off, it, like literally starts shutting down and 440 00:26:24,397 --> 00:26:28,917 S1: you start getting dementia. That's that's me being a fake doctor. 441 00:26:29,357 --> 00:26:31,476 S1: Do not listen to me. I think it's backed up 442 00:26:31,476 --> 00:26:34,196 S1: by tons of data. But I think if you treat 443 00:26:34,196 --> 00:26:37,237 S1: your brain and your body like you are a 22 444 00:26:37,236 --> 00:26:41,157 S1: year old in terms of like you're lifting weights, you're 445 00:26:41,317 --> 00:26:44,677 S1: working out, you're in the sun, you're doing cardio, you're 446 00:26:44,677 --> 00:26:48,357 S1: learning new things constantly. You are executing on new cognitive 447 00:26:48,397 --> 00:26:54,357 S1: tasks constantly, and you're constantly relearning and in upgrading yourself, 448 00:26:54,837 --> 00:26:56,837 S1: then you are basically sending a signal to your brain 449 00:26:56,837 --> 00:27:00,237 S1: and body, hey, we are 22. You better stay in shape. 450 00:27:00,677 --> 00:27:04,237 S1: And me, as old as I am, that's exactly how 451 00:27:04,236 --> 00:27:09,037 S1: I feel. I feel 22 all the time. Unless I'm 452 00:27:09,077 --> 00:27:11,957 S1: like off routine and low energy or whatever, then I 453 00:27:11,956 --> 00:27:16,087 S1: feel like my age. But, um, I would say 95% 454 00:27:16,087 --> 00:27:19,567 S1: of the time I feel 22. And this is why 455 00:27:19,567 --> 00:27:21,527 S1: and this is why I think there's such a big 456 00:27:21,527 --> 00:27:26,446 S1: difference between dimension numbers. Uh, in and I think the 457 00:27:26,446 --> 00:27:29,487 S1: number is 25% versus 15%. So what is that? That's 458 00:27:29,486 --> 00:27:35,127 S1: a massive difference. Um, is just because the old mindset 459 00:27:35,167 --> 00:27:37,767 S1: of retirement versus the new mindset, which is more active 460 00:27:38,887 --> 00:27:42,127 S1: American versus European mindset on life. Really good piece. If 461 00:27:42,127 --> 00:27:44,847 S1: you're useful, it doesn't mean you're valued. This was a 462 00:27:44,847 --> 00:27:48,367 S1: good piece, uh, about economic value and how companies view us. 463 00:27:48,367 --> 00:27:50,807 S1: How much coffee is too much? Really good news. About 464 00:27:50,807 --> 00:27:55,206 S1: 3 to 5 cups of coffee. Uh, discovery. Run your own. 465 00:27:55,206 --> 00:27:59,927 S1: I locally on your Mac Anthropic's interactive prompt engineering tutorial. 466 00:28:00,047 --> 00:28:05,566 S1: Indirect prompt injection overview. My AI skeptic friends might are nuts. 467 00:28:05,567 --> 00:28:09,167 S1: My AI skeptic friends are nuts. This is by another 468 00:28:09,167 --> 00:28:11,367 S1: security guy who's basically saying the same thing that I 469 00:28:11,367 --> 00:28:14,947 S1: was talking about in the beginning from that essay. Um, 470 00:28:14,946 --> 00:28:18,827 S1: and I've got a cool, cool quote here from it. Um, quote. 471 00:28:19,067 --> 00:28:21,627 S1: But the code is shitty like that of a junior developer. 472 00:28:21,627 --> 00:28:24,347 S1: And his response to that. Does an intern cost $20 473 00:28:24,347 --> 00:28:28,467 S1: a month? Because that's what cursor costs $20 a month. 474 00:28:28,627 --> 00:28:31,987 S1: I think that's a great point. Cloud code is my computer. 475 00:28:32,427 --> 00:28:37,506 S1: You to inky turns video into vocabulary flashcards, Java filters, 476 00:28:37,507 --> 00:28:41,227 S1: hacker news job posts with AI powered metadata. Tensor product 477 00:28:41,227 --> 00:28:44,627 S1: attention is all you need. The metamorphosis of intellect. This 478 00:28:44,627 --> 00:28:47,306 S1: is a really cool book. It's an online text. You 479 00:28:47,307 --> 00:28:49,427 S1: can just click the link in the newsletter and you 480 00:28:49,427 --> 00:28:51,707 S1: get the full book. It's fantastic. One of my top 481 00:28:51,707 --> 00:28:54,947 S1: ten sci fi books ever. Andor season two shows how 482 00:28:54,987 --> 00:28:59,267 S1: insider threats actually work in real world organizations. I have 483 00:28:59,267 --> 00:29:01,827 S1: not seen all of Andor one, so I need to 484 00:29:01,827 --> 00:29:04,747 S1: go watch that. And two evidently it's the best Star 485 00:29:04,747 --> 00:29:08,867 S1: Wars in a long time. GitHub repository of end to 486 00:29:08,867 --> 00:29:11,907 S1: end workflows. Someone created a GitHub repo of tons of 487 00:29:11,907 --> 00:29:15,797 S1: scraped any workflow like hundreds of them. And you could 488 00:29:15,797 --> 00:29:19,037 S1: just go get them. Because they're in GitHub now. And 489 00:29:19,197 --> 00:29:21,637 S1: because it's like a text configuration, you could just like 490 00:29:21,677 --> 00:29:24,717 S1: paste them into N810, which by the way, is becoming 491 00:29:24,717 --> 00:29:29,557 S1: one of my favorite AI automation frameworks. And the number A10, 492 00:29:30,197 --> 00:29:33,076 S1: that and bedrock are my favorites and lane graph. But 493 00:29:33,077 --> 00:29:38,037 S1: I'm using that last now, um, preferring to use my 494 00:29:38,037 --> 00:29:40,836 S1: five year experiment with UTC. So someone is now using 495 00:29:40,837 --> 00:29:43,117 S1: UTC for everything. I thought I was crazy for going 496 00:29:43,157 --> 00:29:47,117 S1: to metric. This guy's using UTC book of Secret Knowledge 497 00:29:47,117 --> 00:29:50,237 S1: GitHub repository. Jason Chan and Clint Gibler have a brilliant 498 00:29:50,237 --> 00:29:53,877 S1: conversation in the latest TLDR sick. So my buddy Clint, 499 00:29:53,917 --> 00:29:56,717 S1: who runs TLDR comm, which you should go sign up 500 00:29:56,717 --> 00:30:00,397 S1: for that newsletter. Um, he had a guest post by 501 00:30:00,397 --> 00:30:04,397 S1: Jason Chan, who's like a absolute security guru, actually kind 502 00:30:04,397 --> 00:30:11,517 S1: of pioneered, uh, security guardrails and, um, really, um, really cool, 503 00:30:11,517 --> 00:30:16,697 S1: innovative security guy who ran security at Netflix. But anyway, 504 00:30:16,737 --> 00:30:20,177 S1: he wrote a piece about building a security program for TLDR, 505 00:30:20,577 --> 00:30:28,257 S1: and it is extraordinary. You got to go check it out. Okay, 506 00:30:28,257 --> 00:30:30,177 S1: this is the end of the standard edition of the podcast, 507 00:30:30,177 --> 00:30:32,377 S1: which includes just the news items for the week to 508 00:30:32,377 --> 00:30:34,497 S1: get the rest of the episode, which includes much more 509 00:30:34,497 --> 00:30:37,897 S1: of my analysis, the ideas section, and the weekly member essay. 510 00:30:37,897 --> 00:30:40,617 S1: Please consider becoming a member. As a member, you get 511 00:30:40,657 --> 00:30:43,417 S1: access to all sorts of stuff, most importantly, access to 512 00:30:43,417 --> 00:30:46,537 S1: our extraordinary community of over a thousand brilliant and kind 513 00:30:46,577 --> 00:30:51,017 S1: people in industries like cybersecurity, AI, and the humanities. You 514 00:30:51,017 --> 00:30:53,697 S1: also get access to the UL Book Club, dedicated member 515 00:30:53,697 --> 00:30:56,697 S1: content and events, and lots more. Plus, you'll get a 516 00:30:56,697 --> 00:30:59,017 S1: dedicated podcast feed that you can put into your client 517 00:30:59,017 --> 00:31:01,536 S1: that gets you the full member edition of the podcast 518 00:31:01,537 --> 00:31:03,537 S1: that basically doesn't have this in it, and just goes 519 00:31:03,537 --> 00:31:06,057 S1: all the way through with all the different sections. So 520 00:31:06,097 --> 00:31:07,897 S1: to become a member and get all that, just head 521 00:31:07,897 --> 00:31:12,816 S1: over to Daniel Store.com. That's Daniel Miessler, and we'll see 522 00:31:12,817 --> 00:31:13,497 S1: you next time.