1 00:00:00,140 --> 00:00:03,680 S1: Wiz is the number one cloud security platform based on 2 00:00:03,680 --> 00:00:07,760 S1: independent customer reviews that helps security leaders and developers protect 3 00:00:07,760 --> 00:00:11,570 S1: everything they build and run in the cloud. Leading organizations 4 00:00:11,570 --> 00:00:15,110 S1: from the fortune 100 to high growth companies all trust 5 00:00:15,110 --> 00:00:18,770 S1: Wiz to innovate at the pace of cloud while staying secure. 6 00:00:18,800 --> 00:00:21,770 S1: Click the link in the show notes or head to wizaya. 7 00:00:22,220 --> 00:00:26,720 S1: Slash lowercase ul to request a personalized product tour with 8 00:00:26,720 --> 00:00:30,320 S1: a Wiz expert. Unsupervised Learning Podcast listeners also get a 9 00:00:30,320 --> 00:00:35,300 S1: free Cloud Security Health scan by going to wisdom IOL. 10 00:00:35,570 --> 00:00:43,370 S1: That's wisdom. Tolowercase ul. Welcome to Unsupervised Learning, a security, 11 00:00:43,400 --> 00:00:46,100 S1: AI and meaning focused podcast that looks at how best 12 00:00:46,100 --> 00:00:49,490 S1: to thrive as humans in a post AI world. It 13 00:00:49,490 --> 00:00:53,479 S1: combines original ideas, analysis and mental models to bring not 14 00:00:53,479 --> 00:00:56,690 S1: just the news, but why it matters and how to respond. 15 00:01:01,450 --> 00:01:05,290 S1: All right. Welcome to unsupervised learning. This is Daniel Miessler. Okay. 16 00:01:05,290 --> 00:01:08,440 S1: What do we have here? All right. This is, uh, 17 00:01:08,440 --> 00:01:11,080 S1: when sysop makes you release a patch for your end 18 00:01:11,080 --> 00:01:14,619 S1: of life product. Because the vulnerability is so bad, I have. 19 00:01:14,620 --> 00:01:19,959 S2: Abandoned my child. Say it. I abandoned my child. Say 20 00:01:19,959 --> 00:01:23,890 S2: it louder. Say it louder! I've abandoned my child. I've 21 00:01:23,890 --> 00:01:27,580 S2: abandoned my child. I've abandoned my boy. 22 00:01:28,330 --> 00:01:28,990 S3: That's so. 23 00:01:28,990 --> 00:01:33,880 S1: Good. Okay, uh, fabric is now supporting OpenAI's new model 24 00:01:33,910 --> 00:01:37,570 S1: oh one preview. So you basically just pass it the 25 00:01:37,780 --> 00:01:41,440 S1: switch R flag, which it sends the request using user 26 00:01:41,440 --> 00:01:44,800 S1: rather than system, because, oh one preview is not allowing 27 00:01:44,800 --> 00:01:47,590 S1: system access, so you only have to send it in. 28 00:01:47,590 --> 00:01:51,190 S1: User results aren't that much different honestly. And you also 29 00:01:51,190 --> 00:01:54,160 S1: can't send the temperature parameter which we were doing before. 30 00:01:54,160 --> 00:01:58,430 S1: So this raw option allows you to use zero one 31 00:01:58,460 --> 00:02:01,700 S1: preview with fabric and that works great. Got an insane 32 00:02:01,700 --> 00:02:06,260 S1: cookbook use case 401 preview. So this person actually used 33 00:02:06,260 --> 00:02:08,720 S1: it to. I'm going to open this up to actually 34 00:02:08,720 --> 00:02:13,369 S1: create synthetic data and then validate whether the data was 35 00:02:13,400 --> 00:02:17,540 S1: accurate or not. Really really cool. Really really cool stuff. 36 00:02:17,720 --> 00:02:19,220 S1: And just one of the things that you could do 37 00:02:19,250 --> 00:02:21,859 S1: for zero one preview that you can't actually do with 38 00:02:21,889 --> 00:02:26,180 S1: other models. Okay. So let's see here. Back to kickboxing. 39 00:02:26,210 --> 00:02:29,450 S1: I'll be kickboxing today so hopefully it won't be dead 40 00:02:29,480 --> 00:02:33,709 S1: later tonight. I got a connect on LinkedIn button here 41 00:02:33,710 --> 00:02:36,110 S1: in the newsletter. So if you want to go and 42 00:02:36,110 --> 00:02:40,820 S1: connect there if you connect with me I will mutual connect. Okay. 43 00:02:40,970 --> 00:02:45,860 S1: Last week's comments on current AI advances. So last podcast, 44 00:02:45,889 --> 00:02:47,570 S1: if you're listening to this one, you might have already 45 00:02:47,570 --> 00:02:49,490 S1: listened to that one, but I did a real deep 46 00:02:49,490 --> 00:02:51,650 S1: dive on like the current state of like the path 47 00:02:51,650 --> 00:02:55,040 S1: to AGI. I talked about the oh one preview. 48 00:02:55,120 --> 00:02:55,660 S4: Um. 49 00:02:55,690 --> 00:02:58,270 S1: And I just did a whole bunch of analysis, but 50 00:02:58,270 --> 00:03:00,010 S1: you should check that out if you didn't listen to 51 00:03:00,040 --> 00:03:02,770 S1: that one. And I wrote a new essay called the 52 00:03:02,770 --> 00:03:05,920 S1: Art Quality Tier List. Pull this one up. And this 53 00:03:05,919 --> 00:03:09,340 S1: is basically a way of rating art. And keep in 54 00:03:09,340 --> 00:03:11,530 S1: mind this is for if you're a beginner, if you're 55 00:03:11,530 --> 00:03:14,470 S1: already like an advanced art person, you already know art 56 00:03:14,470 --> 00:03:18,430 S1: back and forward, then like you need to teach me stuff. 57 00:03:18,430 --> 00:03:21,310 S1: I'm not teaching you anything. This is for someone who 58 00:03:21,310 --> 00:03:24,820 S1: is thoughtful but is not an art expert. Even though 59 00:03:24,820 --> 00:03:28,149 S1: I did take an art history class in college. But anyway, 60 00:03:28,150 --> 00:03:31,809 S1: I don't really remember any of it, and I've basically 61 00:03:31,810 --> 00:03:34,690 S1: just not understood what art was for the longest time. 62 00:03:34,690 --> 00:03:37,690 S1: And I finally came up with a definition which I 63 00:03:37,690 --> 00:03:41,440 S1: think is pretty decent. It is indirect expression of something 64 00:03:41,440 --> 00:03:45,940 S1: that matters to humans. And I basically got a hierarchy 65 00:03:45,970 --> 00:03:48,670 S1: here of like the different ways you could look at it. 66 00:03:48,670 --> 00:03:51,910 S1: So going from not art to like proto art to 67 00:03:51,940 --> 00:03:55,100 S1: like hollow art, which is kind of like a IoT, 68 00:03:55,130 --> 00:03:59,510 S1: is one example where it might actually look pretty cool, 69 00:03:59,510 --> 00:04:02,300 S1: or it might actually make you feel a certain way, 70 00:04:02,300 --> 00:04:05,630 S1: but the artist didn't feel anything. And there was a 71 00:04:05,630 --> 00:04:08,510 S1: little bit of a caveat here. If the artist feels something, 72 00:04:08,510 --> 00:04:11,240 S1: but they're using AI to try to make the art, 73 00:04:11,300 --> 00:04:13,580 S1: bring it out. So there's a little bit of a 74 00:04:13,610 --> 00:04:17,120 S1: caveat there. And then decent art is where the artist 75 00:04:17,120 --> 00:04:21,349 S1: feels it and the audience feels it. Great art is like, yeah, 76 00:04:21,380 --> 00:04:25,100 S1: and people like it. And then S-tier or Brilliant is like, 77 00:04:25,130 --> 00:04:29,839 S1: it's recognized as like a world talent or world class 78 00:04:29,839 --> 00:04:33,650 S1: art and that's, that's basically it. So six different levels. 79 00:04:33,650 --> 00:04:38,330 S1: And I even have a methodology for enjoying art. And 80 00:04:38,330 --> 00:04:44,299 S1: this might be especially useful, especially useful for beginners. So 81 00:04:44,300 --> 00:04:47,750 S1: start by feeling the piece. No thinking no analyzing. Just 82 00:04:47,750 --> 00:04:50,390 S1: take it in and experience how it makes you feel. 83 00:04:50,420 --> 00:04:53,620 S1: Do that for a number of, you know, seconds or minutes, 84 00:04:53,620 --> 00:04:55,419 S1: however long you want to take. Now that you have 85 00:04:55,420 --> 00:04:57,820 S1: a sense of how it affected you, analyze the message 86 00:04:57,820 --> 00:04:59,830 S1: you believe is trying to convey. Then, once you have 87 00:04:59,830 --> 00:05:02,980 S1: the message, think about how it was transmitted to you. Like, 88 00:05:03,010 --> 00:05:05,680 S1: was it a song or was it paint or whatever 89 00:05:05,680 --> 00:05:08,620 S1: it was? And this one's really important. If you're with 90 00:05:08,620 --> 00:05:11,770 S1: a friend, do this in silence. Do one through three 91 00:05:11,770 --> 00:05:14,050 S1: in silence for however long you want to spend on 92 00:05:14,050 --> 00:05:17,710 S1: it for each piece, and then discuss your results with them. 93 00:05:17,710 --> 00:05:20,169 S1: It's a really fun way to do this, you know, 94 00:05:20,200 --> 00:05:22,570 S1: get a glass of wine or whatever it is. The 95 00:05:22,570 --> 00:05:24,580 S1: first time I ended up doing this, I was doing 96 00:05:24,580 --> 00:05:29,169 S1: it with my partner and with my friend Sasha, and 97 00:05:29,170 --> 00:05:31,270 S1: we were doing this, I think it was the the 98 00:05:31,270 --> 00:05:34,330 S1: Big Art Museum in New York, and we had lots 99 00:05:34,330 --> 00:05:37,510 S1: of fun, like we all felt like novices, but we 100 00:05:37,510 --> 00:05:39,880 S1: did something like this methodology and it made it a 101 00:05:39,880 --> 00:05:43,480 S1: lot of fun. Okay, so that is that newsletter just 102 00:05:43,480 --> 00:05:46,479 S1: went out, by the way, and security. So US is 103 00:05:46,480 --> 00:05:52,190 S1: evidently super reliant on Chinese cranes. This is freaking me out. 104 00:05:52,220 --> 00:05:58,640 S1: Particularly from this one company. Zpmc. And this report says 80% 105 00:05:58,640 --> 00:06:02,299 S1: of US ship to shore cranes are owned by this 106 00:06:02,300 --> 00:06:04,220 S1: Chinese company. I thought it was going to be like 107 00:06:04,220 --> 00:06:07,789 S1: 25 or 50%. But yeah, I hope someone is actually 108 00:06:07,790 --> 00:06:10,070 S1: tracking this stuff in a war room somewhere. And I'm 109 00:06:10,070 --> 00:06:14,390 S1: pretty sure they are doing doing war room exercises. Fortinet 110 00:06:14,390 --> 00:06:17,270 S1: has confirmed a data breach after someone called. I'm not 111 00:06:17,270 --> 00:06:19,550 S1: going to say that word. Something naughty claimed to have 112 00:06:19,550 --> 00:06:23,510 S1: stolen 440 gigs off of their SharePoint server, and they're 113 00:06:23,510 --> 00:06:26,900 S1: talking about third party risk as being the problem. GitLab 114 00:06:26,900 --> 00:06:31,339 S1: release critical updates to fix multiple vulnerabilities. Got 9.9 on 115 00:06:31,339 --> 00:06:36,320 S1: the Richter scale here. Enable remote exploitation with minimal user 116 00:06:36,320 --> 00:06:40,640 S1: interaction and low privileges. Lazarus Group, which is North Korea, 117 00:06:40,670 --> 00:06:44,990 S1: been targeting Python developers with a piece of malware disguised 118 00:06:44,990 --> 00:06:47,570 S1: as coding tests. They've been doing this for about a 119 00:06:47,570 --> 00:06:51,670 S1: year now, thanks to material for sponsoring, Mastercard is buying 120 00:06:51,670 --> 00:06:57,549 S1: Recorded Future from Insight Partners for $2.65 billion, and they 121 00:06:57,550 --> 00:07:02,830 S1: bought it for 780 million. So nice ROI on that one. 122 00:07:02,830 --> 00:07:06,760 S1: And what I noticed here is like this motion from okay, 123 00:07:06,760 --> 00:07:09,880 S1: it's a startup that does this piece of functionality. Now 124 00:07:09,880 --> 00:07:12,250 S1: we're going to bring it into a platform. In this 125 00:07:12,250 --> 00:07:14,620 S1: case the platform is like Mastercard. So you have good 126 00:07:14,620 --> 00:07:18,040 S1: ideas and execution. And their natural home is within some 127 00:07:18,070 --> 00:07:20,260 S1: sort of ecosystem. And I think this is a good 128 00:07:20,260 --> 00:07:23,350 S1: way to think of startups. It's like a petri dish 129 00:07:23,350 --> 00:07:26,620 S1: for features that will live inside of something bigger, a 130 00:07:26,620 --> 00:07:32,350 S1: bigger platform, bigger ecosystem or a framework. Security. Canary Maturity 131 00:07:32,350 --> 00:07:35,110 S1: Model is a framework designed to help organizations assess and 132 00:07:35,110 --> 00:07:38,710 S1: improve their security posture by using Canary tokens. I love 133 00:07:38,710 --> 00:07:41,530 S1: this concept of detection maturity model. I was just talking 134 00:07:41,560 --> 00:07:45,310 S1: to dropzone about this actually. So here's like a percentage 135 00:07:45,310 --> 00:07:48,290 S1: of your most likely minor behaviors that you're likely to 136 00:07:48,320 --> 00:07:50,510 S1: see based on who you are. And you know what 137 00:07:50,510 --> 00:07:53,330 S1: percentage of these can you detect. And thanks to Defend 138 00:07:53,330 --> 00:07:57,140 S1: defy for sponsoring as well. Australia is set to criminalise 139 00:07:57,140 --> 00:08:00,890 S1: doxxing with penalties up to seven years in jail as 140 00:08:00,890 --> 00:08:05,030 S1: part of a new legislation aimed at modernizing the Privacy Act, 141 00:08:05,060 --> 00:08:09,680 S1: criminalizing doxxing. Seven years in jail. I like it. I 142 00:08:09,680 --> 00:08:12,620 S1: like it. I think, yeah, it does really come down 143 00:08:12,620 --> 00:08:16,070 S1: to there are times that this happens accidentally. In fact, 144 00:08:16,070 --> 00:08:18,800 S1: last week I was worried I accidentally did it, which 145 00:08:18,800 --> 00:08:21,380 S1: I could have just rerecorded, but I caught it and 146 00:08:21,380 --> 00:08:23,870 S1: realised it wasn't doxxing so I didn't have to rerecord. 147 00:08:23,870 --> 00:08:26,660 S1: But the point is, hopefully they're doing it for purely 148 00:08:26,660 --> 00:08:29,960 S1: malicious people, which I'm sure they are. This piece discusses 149 00:08:29,960 --> 00:08:34,219 S1: how AI powered autonomous weapons are changing warfare. So a 150 00:08:34,250 --> 00:08:37,699 S1: bunch of Abrams tanks had to pull back after being 151 00:08:37,730 --> 00:08:42,980 S1: targeted by kamikaze drones run by the Russians. Yeah, definitely. 152 00:08:42,980 --> 00:08:46,720 S1: Go read kill decision by Daniel Suarez. This is the book. 153 00:08:46,750 --> 00:08:50,650 S1: It is extraordinary. What? Did I read this? I read 154 00:08:50,650 --> 00:08:55,329 S1: it in 2016. Yeah, really, really good. Russia's naval activity 155 00:08:55,330 --> 00:08:59,530 S1: around undersea cables is freaking people out in the US. 156 00:08:59,770 --> 00:09:04,240 S1: They're worried that they're going to sabotage underwater infrastructure through 157 00:09:04,240 --> 00:09:09,550 S1: a unit called Goochie. Goochie goochie something. So this unit 158 00:09:09,580 --> 00:09:14,439 S1: operates submarines, surface vessels and naval drones and has been 159 00:09:14,440 --> 00:09:18,910 S1: spotted near critical deep sea cables. And these carry 95% 160 00:09:18,910 --> 00:09:22,929 S1: of international data. US is drafting a New York joint 161 00:09:22,960 --> 00:09:29,380 S1: statement to bolster security of submarine communication cables. Focus on 162 00:09:29,380 --> 00:09:33,310 S1: excluding Chinese firms from the supply chain. Yeah, I think 163 00:09:33,309 --> 00:09:37,750 S1: we need comprehensive critical infrastructure dependency analysis. And like I 164 00:09:37,750 --> 00:09:40,000 S1: said before, I'm pretty sure this is already happening. I 165 00:09:40,000 --> 00:09:43,020 S1: just hope we're using really good red teamers to emulate 166 00:09:43,020 --> 00:09:47,250 S1: the capabilities of these adversaries. US House has voted to 167 00:09:47,250 --> 00:09:51,120 S1: block the purchase of new drones from DJI. So much 168 00:09:51,120 --> 00:09:54,660 S1: coverage of Counter China stuff lately seems like leadership is 169 00:09:54,660 --> 00:09:57,420 S1: getting the message here, which is great. State Department has 170 00:09:57,420 --> 00:10:01,620 S1: declared that Russia's state owned RT news has become a 171 00:10:01,620 --> 00:10:05,309 S1: key player in the Kremlin's military intelligence operations. I mean, 172 00:10:05,340 --> 00:10:08,040 S1: I remember a friend of mine talking about how awesome 173 00:10:08,040 --> 00:10:10,199 S1: art he was because he was hearing stories he'd never 174 00:10:10,200 --> 00:10:13,200 S1: heard anywhere before. And he's like, yeah, you know, the 175 00:10:13,200 --> 00:10:16,260 S1: US has really messed up. And I'm finding out how 176 00:10:16,290 --> 00:10:19,020 S1: messed up the US is from this channel. And I'm like, okay, 177 00:10:19,020 --> 00:10:22,020 S1: what's the channel? And it's RT and I'm like, dude, 178 00:10:22,020 --> 00:10:25,830 S1: you realize this is like, this is just a Russian 179 00:10:25,830 --> 00:10:29,880 S1: state media, media, right? This is like this is pretty 180 00:10:29,880 --> 00:10:32,640 S1: close to propaganda. And he started watching more and he's like, 181 00:10:32,640 --> 00:10:35,640 S1: oh yeah, it was absolutely. But he didn't know he 182 00:10:35,640 --> 00:10:38,460 S1: was being manipulated until it was called out to him. 183 00:10:38,460 --> 00:10:42,130 S1: And this is like so long ago. This might have 184 00:10:42,130 --> 00:10:45,819 S1: been like 2015 or something. All right, sir, he flash 185 00:10:45,820 --> 00:10:50,350 S1: Biscan Kostov is a civilian radio enthusiast who has become 186 00:10:50,350 --> 00:10:53,949 S1: a key figure in Ukraine's drone defence strategy against Russia, 187 00:10:53,980 --> 00:10:58,540 S1: operating from a mobile intelligence centre in his Volkswagen van. 188 00:10:58,660 --> 00:11:02,470 S1: He monitors Russian radio transmissions and shares his findings with 189 00:11:02,470 --> 00:11:08,230 S1: over 127,000 followers, including soldiers and government officials. I'm not 190 00:11:08,230 --> 00:11:09,939 S1: sure I'd be giving this guy a lot of press. 191 00:11:09,940 --> 00:11:13,450 S1: It is basically putting a bounty on him. I and 192 00:11:13,450 --> 00:11:17,770 S1: tech newspaper had humans, and I create novel research ideas, 193 00:11:17,770 --> 00:11:21,340 S1: and then had human experts rate the ideas to see 194 00:11:21,340 --> 00:11:24,370 S1: which ones they liked, and they actually liked the AI 195 00:11:24,400 --> 00:11:27,250 S1: ideas better. And I think this is the way to 196 00:11:27,280 --> 00:11:31,449 S1: measure the abilities of AI, not with like standalone benchmarks, 197 00:11:31,450 --> 00:11:34,150 S1: but like what you do is you have AI results, 198 00:11:34,150 --> 00:11:37,569 S1: you have human results, and then you have a rating system, 199 00:11:37,570 --> 00:11:40,739 S1: which is humans later. We could use AI for that, 200 00:11:40,740 --> 00:11:43,920 S1: but in the meantime, we have humans doing that. And 201 00:11:43,920 --> 00:11:47,370 S1: then we essentially say, which ones did you like better? 202 00:11:47,370 --> 00:11:49,380 S1: And they're blinded. Of course, they don't know which one 203 00:11:49,380 --> 00:11:52,949 S1: is which. So when they pick the AI ones, that's 204 00:11:52,950 --> 00:11:58,199 S1: a pretty good estimate. Estimation that the AI ones are better. 205 00:11:58,230 --> 00:11:59,940 S1: And I think that's how we should do a whole 206 00:11:59,940 --> 00:12:01,650 S1: bunch of these. Otherwise it's going to be a lot 207 00:12:01,679 --> 00:12:05,310 S1: of bias. OpenAI releases their new zero one preview model 208 00:12:05,309 --> 00:12:09,780 S1: focused on reasoning. We've talked a bunch about this one. Um, yeah. 209 00:12:09,809 --> 00:12:12,240 S1: Got my thoughts on it so far in the podcast. 210 00:12:12,240 --> 00:12:14,489 S1: And the big difference I just want to point this 211 00:12:14,490 --> 00:12:16,349 S1: one out. That's why I kept the story in here. 212 00:12:16,350 --> 00:12:20,040 S1: Big difference is that it uses chain of thought reasoning 213 00:12:20,040 --> 00:12:23,520 S1: and the fact that it actually spends time and actually tokens, 214 00:12:23,520 --> 00:12:28,380 S1: which means money thinking before responding, because before it would 215 00:12:28,380 --> 00:12:30,330 S1: be tokens in and tokens out. And that was pretty 216 00:12:30,330 --> 00:12:34,559 S1: much it. Klarna's CEO says that I could replace enterprise. 217 00:12:34,590 --> 00:12:39,520 S1: Enterprise software giants like Salesforce and Workday claims that conversational AI, 218 00:12:39,550 --> 00:12:44,410 S1: like OpenAI's strawberry, can handle natural language commands to build 219 00:12:44,410 --> 00:12:48,040 S1: custom apps. 100%, this is what I've been saying Sbca 220 00:12:48,160 --> 00:12:51,790 S1: that's going to happen. And yeah, this is my Fpca piece, 221 00:12:51,790 --> 00:12:55,660 S1: and this is what I think future software looks like. 222 00:12:55,660 --> 00:12:58,240 S1: You've got, uh, see, I can find the thing. There 223 00:12:58,240 --> 00:13:01,930 S1: we go. State policy questions, action. I think that is 224 00:13:01,929 --> 00:13:05,380 S1: the vibe. AI powered SAR satellites are now capable of 225 00:13:05,380 --> 00:13:09,069 S1: detecting aircraft from space due to new radar tech, which 226 00:13:09,100 --> 00:13:12,189 S1: enables real time monitoring of air traffic, good for both 227 00:13:12,190 --> 00:13:17,380 S1: civilian and military uses. Cardio tech? That's a good name. 228 00:13:17,559 --> 00:13:22,720 S1: Leveraging AI to tackle cardiovascular disease. Leading cause of death. 229 00:13:22,720 --> 00:13:27,550 S1: Partnering with 65 hospitals to build a massive human heart tissue. 230 00:13:27,580 --> 00:13:34,300 S1: Multi-omics dataset to identify new drug candidates. Super exciting because 231 00:13:34,300 --> 00:13:37,620 S1: the AI I needs data to form its model of 232 00:13:37,620 --> 00:13:40,050 S1: the world, and this is a whole bunch of that 233 00:13:40,050 --> 00:13:42,990 S1: data to model it, because then you could figure out 234 00:13:42,990 --> 00:13:45,750 S1: that's kind of the punchline here. You can figure out 235 00:13:45,750 --> 00:13:49,140 S1: if drugs might work or that's the plan anyway. If 236 00:13:49,140 --> 00:13:52,110 S1: you have a really good model of how the heart works, 237 00:13:52,110 --> 00:13:57,239 S1: you can have drugs, potentially synthetic testing of drugs to 238 00:13:57,270 --> 00:13:59,969 S1: see if they will work without doing trials or without 239 00:13:59,970 --> 00:14:03,780 S1: doing as many trials, or for as long or as expensive. 240 00:14:03,780 --> 00:14:07,110 S1: Salesforce just launched Agent Force, a suite of AI powered 241 00:14:07,110 --> 00:14:12,510 S1: agents designed to enhance human workers across various business functions. Yeah, yeah, 242 00:14:12,540 --> 00:14:16,559 S1: definitely designed to enhance human workers. Not designed at all 243 00:14:16,590 --> 00:14:20,580 S1: to replace them. Yep, yep. Waymo's latest data shows that 244 00:14:20,610 --> 00:14:24,840 S1: human drivers are responsible for most serious collisions involving its 245 00:14:24,840 --> 00:14:30,030 S1: driverless cars. 16 out of 23 severe crashes being rear 246 00:14:30,060 --> 00:14:34,720 S1: ended by human driven vehicles. Tesla's Cybertruck is spiking in 247 00:14:34,720 --> 00:14:39,850 S1: the truck segment. Well, electric truck segment with 61% of 248 00:14:39,850 --> 00:14:45,790 S1: sales surge in July, a surge of 61% in July. 249 00:14:45,820 --> 00:14:50,020 S1: Did way better than the R1t's and the Ford F-150 lightning. 250 00:14:50,020 --> 00:14:52,210 S1: I think we didn't. We just see a story saying 251 00:14:52,210 --> 00:14:54,790 S1: the lightning was crushing it. I don't know why there's 252 00:14:54,790 --> 00:14:56,440 S1: a big jump right now, but I have seen a 253 00:14:56,440 --> 00:15:01,420 S1: whole bunch more very recently. Anecdotally, USPS has rolled out 254 00:15:01,420 --> 00:15:04,150 S1: a new delivery vehicle. Let's see if we can get 255 00:15:04,150 --> 00:15:06,400 S1: a picture of this thing, which I saw. Yeah, look 256 00:15:06,400 --> 00:15:09,430 S1: at that. Looks like a duck. Looks like a duck, Bill, 257 00:15:09,460 --> 00:15:12,670 S1: doesn't it? Yeah. Evidently the drivers love them. They're very, 258 00:15:12,670 --> 00:15:17,050 S1: very practical, but they look very strange. Dmitri Greenberg has 259 00:15:17,050 --> 00:15:20,050 S1: managed to run Linux and Ultrix on a business card, 260 00:15:20,080 --> 00:15:23,170 S1: turning it into a tiny computer, eight kilobytes of Ram 261 00:15:23,170 --> 00:15:28,360 S1: and 32kB of flash storage. New studies showing that debunk bot, 262 00:15:28,360 --> 00:15:33,990 S1: an AI chatbot can effectively persuade users to abandon conspiracy theories. 263 00:15:33,990 --> 00:15:38,640 S1: The bot made significant progress in changing people's beliefs, challenging 264 00:15:38,640 --> 00:15:43,170 S1: the notion that facts and logic cannot combat conspiracies. I mean, 265 00:15:43,200 --> 00:15:45,150 S1: this makes sense to me. So this is what I 266 00:15:45,150 --> 00:15:49,109 S1: wrote you. You can convince. If you can convince somebody 267 00:15:49,110 --> 00:15:51,540 S1: of something, you can convince them as well. This is 268 00:15:51,540 --> 00:15:55,530 S1: why I'm optimistic about having AI on us all the time, 100%. 269 00:15:55,530 --> 00:15:58,050 S1: It can be Orwellian, or it could be a defender 270 00:15:58,050 --> 00:16:01,800 S1: or protector or tutor or coach, etc. that's always with us. 271 00:16:01,830 --> 00:16:05,220 S1: Community college had to cancel its CS career fair because 272 00:16:05,220 --> 00:16:08,610 S1: no companies reached out. Super sad, but super expected. If 273 00:16:08,610 --> 00:16:11,250 S1: you have people coming out of college with four year 274 00:16:11,250 --> 00:16:14,520 S1: degrees and master's in CS and they can't find jobs, 275 00:16:14,520 --> 00:16:18,990 S1: what do you think? Do you think companies are rejecting right? 276 00:16:19,020 --> 00:16:21,360 S1: The companies who would come to the job fair, they're 277 00:16:21,510 --> 00:16:24,479 S1: rejecting the master's degree people, but they're going to go 278 00:16:24,480 --> 00:16:26,880 S1: to community college? I don't think so. This is why 279 00:16:26,880 --> 00:16:30,920 S1: we need human 3.0. Future is connecting directly to individuals 280 00:16:30,920 --> 00:16:35,000 S1: not going through these credential forms. Google has officially killed 281 00:16:35,000 --> 00:16:37,940 S1: off cache links, which is sad because that's how I 282 00:16:37,940 --> 00:16:40,370 S1: used to see old versions of the site. And you 283 00:16:40,370 --> 00:16:43,310 S1: can see the site even if it was down, like, whatever. 284 00:16:43,340 --> 00:16:47,840 S1: United Airlines is partnering with SpaceX to bring free Starlink 285 00:16:47,840 --> 00:16:50,960 S1: Wi-Fi to all of its planes. I'm a united person. 286 00:16:50,960 --> 00:16:53,960 S1: I'm so happy that I'm a united person right now, man, 287 00:16:53,990 --> 00:16:57,410 S1: because I fly JSX as well, and the Starlink is 288 00:16:57,410 --> 00:16:59,780 S1: so good. It's better than a lot of my friends. 289 00:16:59,780 --> 00:17:03,020 S1: Wi-Fi at their house. Ukraine just launched its biggest drone 290 00:17:03,020 --> 00:17:07,460 S1: attack on Moscow yet, hitting the region with 144 drones. 291 00:17:07,460 --> 00:17:10,669 S1: I still don't understand how Ukraine could possibly be winning 292 00:17:10,670 --> 00:17:13,760 S1: this war. Completely insane to me, but in a good way. 293 00:17:13,790 --> 00:17:17,570 S1: Sweden is increasing how much it's paying migrants to go home. 294 00:17:17,600 --> 00:17:22,310 S1: It's now up to $34,000, roughly paying them to just leave. 295 00:17:22,340 --> 00:17:27,800 S1: NASA's advanced composite solar sail successfully been deployed its ultra 296 00:17:27,800 --> 00:17:31,380 S1: thin solar sail is now in low Earth orbit means 297 00:17:31,380 --> 00:17:34,140 S1: it's visible in the night sky from various locations. It 298 00:17:34,140 --> 00:17:37,980 S1: could be as bright as serious. Sirius is the brightest 299 00:17:37,980 --> 00:17:41,070 S1: star out there. The dog star. It is very bright, 300 00:17:41,190 --> 00:17:43,740 S1: so I can't wait to go find this thing and 301 00:17:43,740 --> 00:17:45,780 S1: I wonder. I mean, it would be crazy to see 302 00:17:45,780 --> 00:17:48,720 S1: something as bright as Sirius actually moving, so I can't 303 00:17:48,720 --> 00:17:53,160 S1: wait to see this thing. Got a comet? They're calling 304 00:17:53,160 --> 00:17:56,910 S1: it the comet of the century. I feel like these astronomer, uh, 305 00:17:56,910 --> 00:18:00,480 S1: journalists are always saying it's the brightest. You will not 306 00:18:00,480 --> 00:18:04,860 S1: see another solar eclipse for 900 years or whatever. And 307 00:18:04,859 --> 00:18:08,880 S1: then it turns out, well, not one that comes directly here. 308 00:18:08,880 --> 00:18:11,760 S1: Not one that falls on a Friday. Not one that 309 00:18:11,760 --> 00:18:14,640 S1: you could see while standing upside down or whatever. And 310 00:18:14,640 --> 00:18:16,470 S1: then you look and you're like, okay, when is the 311 00:18:16,470 --> 00:18:20,520 S1: actual next eclipse? Oh, nine months from now or whatever. Anyway, 312 00:18:20,520 --> 00:18:21,810 S1: this is going to be a cool one. It's going 313 00:18:21,840 --> 00:18:25,050 S1: to be bright. So I'm going to try to remember 314 00:18:25,050 --> 00:18:27,050 S1: these times. I'm sure they'll say more about it at 315 00:18:27,050 --> 00:18:29,359 S1: the time so I can go take a look. US 316 00:18:29,390 --> 00:18:33,230 S1: is closing a trade loophole that e-commerce people like Temu 317 00:18:33,230 --> 00:18:38,510 S1: and Shine. Shine, shine have been exploiting. Loophole allows them 318 00:18:38,510 --> 00:18:42,530 S1: to ship goods directly to customers without paying tariffs. So 319 00:18:42,530 --> 00:18:44,630 S1: the US government is going after that. There's a leaked 320 00:18:44,660 --> 00:18:48,619 S1: PDF that details Mr. Beast's unique company culture, and it 321 00:18:48,619 --> 00:18:51,290 S1: got a one page summary of it here. Pretty, pretty 322 00:18:51,320 --> 00:18:54,830 S1: smart business stuff. Although he is, I think being investigated 323 00:18:54,830 --> 00:18:58,850 S1: for some some stuff. See here. Uh, person says sunlight 324 00:18:58,880 --> 00:19:01,850 S1: cured their migraines. It's not a study. It's really just 325 00:19:01,850 --> 00:19:04,880 S1: somebody talking about it. But I figure most people who 326 00:19:04,880 --> 00:19:07,850 S1: have migraines have tried everything else, so why not try this? 327 00:19:07,880 --> 00:19:10,669 S1: It's basically sunlight getting lots of sunlight, I think in 328 00:19:10,670 --> 00:19:14,900 S1: the morning. Love this concept from Laura Hogan. Be a thermostat, 329 00:19:14,930 --> 00:19:18,410 S1: not a thermometer. That's that's cool. I mean, it's cool 330 00:19:18,410 --> 00:19:21,050 S1: if you don't even read the article. Be a thermostat, 331 00:19:21,050 --> 00:19:24,750 S1: not a thermometer. Content driven development is a strategy for 332 00:19:24,780 --> 00:19:28,500 S1: making progress on side projects by focusing on small, shareable 333 00:19:28,500 --> 00:19:30,869 S1: pieces of work. So you finish a thing you share, 334 00:19:30,869 --> 00:19:33,870 S1: you finish a thing you share. In 1913, Vienna was 335 00:19:33,869 --> 00:19:38,490 S1: quite a place to hang out. Hitler, Leon, Trotsky, Tito, 336 00:19:38,490 --> 00:19:43,290 S1: Sigmund Freud, Joseph Stalin all hanging out in the same city. 337 00:19:43,290 --> 00:19:46,560 S1: And by the way, this is before Hitler. And I 338 00:19:46,590 --> 00:19:49,740 S1: guess Stalin turned evil. And whoever else in that list 339 00:19:49,740 --> 00:19:53,220 S1: was evil. Not sure how evil Trotsky was, because that's 340 00:19:53,220 --> 00:19:57,149 S1: how poorly read I am about Russian history. Discovery merkel 341 00:19:57,150 --> 00:20:00,180 S1: map CLI command line tool lets you search and enumerate 342 00:20:00,180 --> 00:20:07,740 S1: subdomains using the merkel Map API. Okay, 71TB, ZFS, NAS 343 00:20:07,770 --> 00:20:12,570 S1: 24 four terabyte drives. It has lasted without a single 344 00:20:12,570 --> 00:20:16,350 S1: drive failure for over a decade, thanks to a strategy 345 00:20:16,350 --> 00:20:19,230 S1: of keeping the server off when not in use. Cheating? 346 00:20:19,230 --> 00:20:22,969 S1: I feel like it's cheating. Doctor Mordechai Guri She has 347 00:20:22,970 --> 00:20:26,990 S1: unveiled a new side channel attack called Rambo, which uses 348 00:20:26,990 --> 00:20:31,850 S1: radio signals from a device's Ram to exfiltrate data over 349 00:20:31,880 --> 00:20:35,720 S1: Airgapped networks. I was like, I know this is Israeli. 350 00:20:35,720 --> 00:20:38,840 S1: I know for sure, I guess Tel Aviv, but it 351 00:20:38,840 --> 00:20:43,610 S1: was actually a different Israeli university. But yes, it was Israeli. 352 00:20:43,640 --> 00:20:47,750 S1: Israelis are the side channel goats. Six techniques I use 353 00:20:47,750 --> 00:20:50,690 S1: to create a great user experience for shell scripts. This 354 00:20:50,690 --> 00:20:53,690 S1: was a great read. User friendly shell scripts with techniques 355 00:20:53,690 --> 00:20:59,030 S1: like comprehensive error handling, colorful output, and progress reporting. These 356 00:20:59,030 --> 00:21:01,130 S1: things really do make a difference. They make it feel 357 00:21:01,130 --> 00:21:03,320 S1: like an actual app. Two people made a tool that 358 00:21:03,320 --> 00:21:07,280 S1: lets you transfer playlists between Apple Music and Spotify and 359 00:21:07,280 --> 00:21:10,100 S1: YouTube music. Those are kind of the three big ones. 360 00:21:10,100 --> 00:21:13,070 S1: Nothing is a timer that celebrates the art of doing 361 00:21:13,100 --> 00:21:17,450 S1: absolutely nothing. And how to create a pipeline with pinecone 362 00:21:17,450 --> 00:21:22,199 S1: and semantic image search CLI CLE ideas. I love it 363 00:21:22,230 --> 00:21:25,859 S1: when experts completely disagree about a really important thing. Like 364 00:21:25,859 --> 00:21:29,790 S1: for example, I have no idea if China is thriving 365 00:21:29,790 --> 00:21:31,560 S1: right now and is about to take over the world, 366 00:21:31,560 --> 00:21:34,980 S1: or if they are about to have this demographic collapse 367 00:21:34,980 --> 00:21:37,590 S1: and their economy is tanking, and they've got the housing 368 00:21:37,590 --> 00:21:41,669 S1: crisis and they're just in major upheaval, and she is 369 00:21:41,670 --> 00:21:45,150 S1: facing pressure and time is running out. And they have 370 00:21:45,180 --> 00:21:48,090 S1: brain drain going to the west. And they're super screwed 371 00:21:48,090 --> 00:21:50,160 S1: because they're slowing down AI and they're going to get 372 00:21:50,160 --> 00:21:52,619 S1: passed up. I don't know if that's true or if 373 00:21:52,619 --> 00:21:55,080 S1: they're about to invent their own chips. They're about to 374 00:21:55,109 --> 00:21:57,990 S1: not need Nvidia. They're about to take over everything because 375 00:21:57,990 --> 00:22:00,930 S1: the West is stupid. And I could find the world's 376 00:22:00,930 --> 00:22:04,050 S1: best experts who tell me both of those things and 377 00:22:04,050 --> 00:22:06,720 S1: tell me that the other set of experts is completely wrong. 378 00:22:06,720 --> 00:22:09,750 S1: I find that fascinating, and I've read a whole bunch 379 00:22:09,750 --> 00:22:14,790 S1: of books. I've read probably 12 deep, like large books 380 00:22:14,790 --> 00:22:18,929 S1: about China or dedicated just to China and its development, 381 00:22:18,930 --> 00:22:21,679 S1: probably the most important one being like Chip Wars, although 382 00:22:21,680 --> 00:22:23,450 S1: that was more about chips than it was about China. 383 00:22:23,450 --> 00:22:25,850 S1: But it definitely talked about China a lot. But I 384 00:22:25,850 --> 00:22:28,220 S1: read a whole bunch that were dedicated just to China. 385 00:22:28,220 --> 00:22:30,920 S1: And I mean, it didn't tell me the answer. I mean, 386 00:22:30,950 --> 00:22:33,050 S1: I have all these opinions and I listen to these 387 00:22:33,050 --> 00:22:35,419 S1: opinions of like, the smartest people, but the fact that 388 00:22:35,420 --> 00:22:38,870 S1: I still don't know, after reading all those books and 389 00:22:38,869 --> 00:22:41,180 S1: seeing all the data and listening to all these smart 390 00:22:41,180 --> 00:22:44,449 S1: experts and I still don't know, that is exhilarating to me. 391 00:22:44,450 --> 00:22:48,020 S1: I'm like, wow. I mean, I love the idea of 392 00:22:48,050 --> 00:22:51,710 S1: like and I'm happy with myself for not like picking 393 00:22:51,710 --> 00:22:53,750 S1: one and just saying, oh, this is what I think 394 00:22:53,780 --> 00:22:56,780 S1: is going to happen. I tend to it's like strong opinions, 395 00:22:56,780 --> 00:22:59,120 S1: loosely held, you know, and the recommendation of the week 396 00:22:59,150 --> 00:23:03,920 S1: actively guard against age related lock in, which starts around 30. 397 00:23:03,950 --> 00:23:06,320 S1: I'm guessing I made that up, but roughly around there, 398 00:23:06,320 --> 00:23:08,840 S1: maybe even late 20s. And for some people it doesn't 399 00:23:08,840 --> 00:23:12,320 S1: kick in. I've been very resistant to it, although I 400 00:23:12,350 --> 00:23:16,400 S1: am writing this also to myself because you actively have 401 00:23:16,400 --> 00:23:19,650 S1: to fight it. Listen to new music, read new books 402 00:23:19,650 --> 00:23:22,410 S1: with new ideas in them, talk to new people. Go 403 00:23:22,410 --> 00:23:26,100 S1: to strange restaurants. Try new foods. Don't let your experiences 404 00:23:26,100 --> 00:23:29,640 S1: reduce into a tighter and tighter death spiral. Variation keeps 405 00:23:29,640 --> 00:23:32,370 S1: your mind young and the aphorism of the week. Choosing 406 00:23:32,369 --> 00:23:35,100 S1: not to read great books has the same effect as 407 00:23:35,100 --> 00:23:38,550 S1: not being allowed to. Choosing not to read great books 408 00:23:38,550 --> 00:23:43,889 S1: has the same effect as not being allowed to. Unsupervised 409 00:23:43,890 --> 00:23:46,379 S1: learning is produced and edited by Daniel Miessler on a 410 00:23:46,410 --> 00:23:51,270 S1: Neumann U87 AI microphone using Hindenburg. Intro and outro music 411 00:23:51,270 --> 00:23:54,330 S1: is by zombie with a Y. And to get the 412 00:23:54,330 --> 00:23:56,429 S1: text and links from this episode, sign up for the 413 00:23:56,430 --> 00:24:02,070 S1: newsletter version of the show at Daniel miessler.com/newsletter. We'll see 414 00:24:02,070 --> 00:24:02,850 S1: you next time.