1 00:00:00,880 --> 00:00:05,040 S1: Unsupervised Learning is a podcast about trends and ideas in cybersecurity, 2 00:00:05,080 --> 00:00:09,960 S1: national security, AI, technology and society, and how best to 3 00:00:10,000 --> 00:00:18,000 S1: upgrade ourselves to be ready for what's coming. All right, 4 00:00:18,000 --> 00:00:21,919 S1: starting off this week. So I completely reset my email 5 00:00:21,920 --> 00:00:26,280 S1: labels and filters this week. So cruft going back to like, 6 00:00:26,480 --> 00:00:29,000 S1: I don't know, 20 years old or whatever from Gmail. 7 00:00:29,440 --> 00:00:32,120 S1: Basically all those filters got rid of them because I'm 8 00:00:32,120 --> 00:00:34,680 S1: going to start filtering with AI. I think at some 9 00:00:34,680 --> 00:00:37,400 S1: point instead, I'm already doing a little bit of that 10 00:00:37,400 --> 00:00:42,239 S1: with superhuman. I just I'm worried I'm missing things, and 11 00:00:42,240 --> 00:00:45,240 S1: I've already found a few things, like different bills for 12 00:00:45,240 --> 00:00:48,240 S1: things I thought I had turned off or whatever. Um, 13 00:00:48,240 --> 00:00:51,360 S1: but anyway, very clean feeling. I mean, that was probably 14 00:00:51,760 --> 00:00:55,400 S1: a couple of hundred filters that were in there and 15 00:00:55,440 --> 00:01:02,180 S1: like weird labels and subfolders. So I basically consolidated folders, labels, rules. 16 00:01:02,180 --> 00:01:05,580 S1: And I only have like less than a dozen now 17 00:01:05,620 --> 00:01:10,380 S1: of each. So very nice. Clean feeling. Highly recommended. Going 18 00:01:10,380 --> 00:01:14,020 S1: to be trying out Karpathy's idea. This is Andrej Karpathy 19 00:01:14,060 --> 00:01:17,340 S1: has an idea of using a single Apple note instead 20 00:01:17,340 --> 00:01:20,420 S1: of actually using, uh, you know, a million. I have 21 00:01:20,420 --> 00:01:24,940 S1: currently around 2900, and it's not really a nice situation 22 00:01:24,940 --> 00:01:28,860 S1: to be in. So I'm going to try this single 23 00:01:28,860 --> 00:01:30,540 S1: note thing. It's pretty cool. And I got a link 24 00:01:30,540 --> 00:01:33,380 S1: here in the show and, uh, watch your API key 25 00:01:33,380 --> 00:01:38,140 S1: and I'd agents because I've got a $2,000 bill, which 26 00:01:38,140 --> 00:01:41,539 S1: is the limit that I set because I did this 27 00:01:41,540 --> 00:01:45,220 S1: weird web documentation lookup. And some for some reason it 28 00:01:45,220 --> 00:01:48,260 S1: was calling the LLM every time it did it. And 29 00:01:48,260 --> 00:01:50,420 S1: I was watching it run because I thought it was 30 00:01:50,420 --> 00:01:53,380 S1: just a pure HTTP query. I don't know why I 31 00:01:53,380 --> 00:01:56,500 S1: was using an agent to process results each time. It's 32 00:01:56,500 --> 00:02:00,640 S1: very silly. But anyway, um. Yeah. That's why you should 33 00:02:00,640 --> 00:02:03,840 S1: have limits set. I'm glad I had mine set. Um, 34 00:02:03,840 --> 00:02:06,480 S1: but now I've broken out multiple keys and lowered some 35 00:02:06,480 --> 00:02:10,560 S1: limits to make it, uh, even more of a backstop. 36 00:02:10,960 --> 00:02:14,840 S1: Go delete your 23 andme data. They are selling out. 37 00:02:15,120 --> 00:02:18,720 S1: They are going bankrupt and selling their data to whoever. 38 00:02:19,360 --> 00:02:21,760 S1: So go delete that if you have it. Uh, new 39 00:02:21,760 --> 00:02:28,720 S1: obscure book recommendation. Fanged noumena. Fanged noumena or fanged noumena, maybe. Um, 40 00:02:28,720 --> 00:02:31,000 S1: my friend Joel Parish recommended that to me. So I'm 41 00:02:31,000 --> 00:02:33,960 S1: going to read that. And I was made emotionally leaky 42 00:02:34,000 --> 00:02:38,520 S1: last night from this, uh, video of this pianist, um, 43 00:02:38,520 --> 00:02:41,840 S1: on this YouTube channel called Great Measures. And he did 44 00:02:41,840 --> 00:02:45,560 S1: a cover of Fade to Black by Metallica, which is 45 00:02:45,600 --> 00:02:47,880 S1: kind of an emotional song for like, a personal reason 46 00:02:47,880 --> 00:02:50,720 S1: for me, uh, related to, like, a friend in high school, 47 00:02:50,840 --> 00:02:53,560 S1: but I don't know. I don't really listen to the 48 00:02:53,560 --> 00:02:56,610 S1: song that much, and I don't know, musically, it's not 49 00:02:56,610 --> 00:03:00,290 S1: that exciting to me, but it has some. Emotional meaning 50 00:03:00,330 --> 00:03:02,210 S1: to me. And then so I'm watching him play it 51 00:03:02,210 --> 00:03:05,850 S1: on piano and it is. Stunning. I mean, it is 52 00:03:05,850 --> 00:03:10,169 S1: absolutely just gorgeous because he's adding like so much more 53 00:03:10,169 --> 00:03:13,210 S1: depth and everything to it. Plus it's on piano and 54 00:03:13,210 --> 00:03:16,250 S1: it's like, yeah, it, uh, it got me. Definitely worth 55 00:03:16,250 --> 00:03:19,090 S1: checking out. And I love the channel. The channel is 56 00:03:19,090 --> 00:03:23,410 S1: called Great Measures. So it's basically, uh, a metal guy 57 00:03:23,450 --> 00:03:28,889 S1: showing his pianist friend, so classically trained musician, showing him 58 00:03:28,889 --> 00:03:31,970 S1: a bunch of metal. It's quite awesome. So I'm joining 59 00:03:31,970 --> 00:03:35,050 S1: Caleb Sima and Edward Wu for a panel on Drop 60 00:03:35,090 --> 00:03:40,410 S1: Zone AI Security Frontiers conference, and that is on March 27th, 61 00:03:40,410 --> 00:03:42,770 S1: which is well, it will have already happened by the 62 00:03:42,770 --> 00:03:44,810 S1: time you hear this. But, um, yeah, we're going to 63 00:03:44,850 --> 00:03:48,610 S1: talk about where JNI stands in security today and where 64 00:03:48,610 --> 00:03:51,570 S1: it's headed, and it's, uh, virtual free and worth it. 65 00:03:51,570 --> 00:03:54,190 S1: So you can go watch the thing, even though it's 66 00:03:54,190 --> 00:03:58,710 S1: already recorded. Cybersecurity y white box red teaming makes me 67 00:03:58,710 --> 00:04:03,510 S1: feel weird. So this guy is Zygi Strannix, which are 68 00:04:03,510 --> 00:04:07,990 S1: how to pronounce that. But Zygi shares unsettling experiences with 69 00:04:07,990 --> 00:04:13,630 S1: models appearing to express distress during advanced LLM safety techniques. 70 00:04:13,630 --> 00:04:16,229 S1: And the quote here is it just doesn't feel good 71 00:04:16,230 --> 00:04:20,230 S1: to be responsible for making models scream. It distracts me 72 00:04:20,230 --> 00:04:23,710 S1: from doing research and makes me write rambling blog posts. 73 00:04:23,710 --> 00:04:27,190 S1: That's uh, yeah, that's a problem. It feels like Black 74 00:04:27,190 --> 00:04:30,029 S1: Mirror to me. Like I'm not worried that it's actually conscious, 75 00:04:30,029 --> 00:04:31,990 S1: but at some point I will be and it'll be 76 00:04:31,990 --> 00:04:34,950 S1: hard to know the difference. And like he said, it's 77 00:04:34,950 --> 00:04:38,470 S1: still disturbing if the thing is making all the sounds 78 00:04:38,470 --> 00:04:42,110 S1: of it hurting, even though you you think you know 79 00:04:42,150 --> 00:04:44,990 S1: that it doesn't have the capability to do so. Next 80 00:04:44,990 --> 00:04:48,950 S1: one here, right? White House OpSec fail. White House accidentally 81 00:04:48,950 --> 00:04:53,089 S1: revealed top secret Houthi bombing plans to. The editor of 82 00:04:53,250 --> 00:04:56,690 S1: The Atlantic magazine shared the plans in a signal group 83 00:04:56,690 --> 00:05:00,730 S1: that didn't realize the reporter was in there. And the 84 00:05:00,730 --> 00:05:05,049 S1: worst part is the message before they started sharing the stuff. 85 00:05:05,610 --> 00:05:08,770 S1: We are currently clean on OpSec, and I believe that 86 00:05:08,770 --> 00:05:11,650 S1: was like the Secretary of State saying that we are 87 00:05:11,650 --> 00:05:15,090 S1: currently clean on OpSec. And of course, the person also 88 00:05:15,690 --> 00:05:19,130 S1: receiving that message is a civilian and the head of 89 00:05:19,130 --> 00:05:23,370 S1: a magazine. My goodness, my agents, security and companies like Microsoft. 90 00:05:23,370 --> 00:05:25,650 S1: So I got invited to a Microsoft media event last 91 00:05:25,650 --> 00:05:28,130 S1: week in SF where they showed off all the AI 92 00:05:28,330 --> 00:05:31,850 S1: agent stuff and copilot that they're talking about this week. Basically, 93 00:05:31,850 --> 00:05:34,450 S1: they're adding agents to tons of products under the banner 94 00:05:34,450 --> 00:05:37,610 S1: of copilot, and I had a single thought while spending 95 00:05:37,610 --> 00:05:40,529 S1: like three hours talking to everyone there from like red 96 00:05:40,570 --> 00:05:44,930 S1: team to threat intelligence, incident response or whatever. But basically 97 00:05:44,930 --> 00:05:47,370 S1: startups need to hurry up, because what I saw in 98 00:05:47,410 --> 00:05:51,150 S1: that room, I think is like the future and definitely 99 00:05:51,150 --> 00:05:54,510 S1: not sponsored by Microsoft. I'm not even like a Microsoft fanboy, 100 00:05:54,589 --> 00:05:58,150 S1: but many of the agents in the room could talk 101 00:05:58,150 --> 00:06:03,589 S1: to all of the other Microsoft services, right? Microsoft vulnerability management, 102 00:06:03,589 --> 00:06:07,270 S1: identity and access management, asset management. They could talk to 103 00:06:07,310 --> 00:06:10,870 S1: all those different systems. They could pull context from across 104 00:06:10,870 --> 00:06:17,910 S1: the entire organization, like HR, like asset management endpoints, uh, 105 00:06:17,910 --> 00:06:23,270 S1: cloud vulnerability data, local vulnerability data on the actual workstations, 106 00:06:23,270 --> 00:06:27,590 S1: vulnerability management, ticketing systems. What I'm saying is the companies 107 00:06:27,589 --> 00:06:30,910 S1: that are going to win this AI security game are 108 00:06:30,910 --> 00:06:34,029 S1: not necessarily the ones with the best AI or agent tech, 109 00:06:34,430 --> 00:06:38,830 S1: but the ones that can best leverage customer company context 110 00:06:39,150 --> 00:06:42,630 S1: for their AI or agents. So at first it'll be 111 00:06:42,630 --> 00:06:45,549 S1: startups because they can move the fastest, right? But soon 112 00:06:45,550 --> 00:06:49,370 S1: they're going to have a major disadvantage compared to like 113 00:06:49,410 --> 00:06:54,930 S1: Microsoft and companies that give access, or companies where the 114 00:06:54,930 --> 00:06:59,450 S1: AI has the ability to access unified company contacts UCC. 115 00:06:59,690 --> 00:07:03,330 S1: So other companies, I think, like Amazon and Databricks and stuff, 116 00:07:03,330 --> 00:07:09,130 S1: will also try to build this UCC unified customer context 117 00:07:09,850 --> 00:07:13,250 S1: so that they'll be able to create that data store 118 00:07:13,250 --> 00:07:15,970 S1: for people to use AI with. So it won't only 119 00:07:15,970 --> 00:07:19,490 S1: be Microsoft, right, or Google or Amazon, because people like 120 00:07:19,490 --> 00:07:22,370 S1: Databricks and other companies are going to come in and say, hey, 121 00:07:22,370 --> 00:07:24,650 S1: let me just collect all your data and put it 122 00:07:24,650 --> 00:07:27,410 S1: into a UCC. And then any AI that you have 123 00:07:27,410 --> 00:07:29,450 S1: from these different vendors, they could all tap into the 124 00:07:29,450 --> 00:07:32,330 S1: same thing. So that will also be a thing. It 125 00:07:32,330 --> 00:07:35,929 S1: already is a thing. Again, companies like Databricks, I believe 126 00:07:35,930 --> 00:07:38,330 S1: they're playing in that space. But you don't want to 127 00:07:38,330 --> 00:07:41,730 S1: be a startup trying to implement AI in a customer's company. 128 00:07:41,730 --> 00:07:45,500 S1: When you don't have access to their UCC And you're 129 00:07:45,500 --> 00:07:49,500 S1: competing against someone else who does like Microsoft. Right. That 130 00:07:49,500 --> 00:07:52,260 S1: is a bad place to be. So the main game 131 00:07:52,260 --> 00:07:56,500 S1: for making AI useful or powerful, especially in security and 132 00:07:56,500 --> 00:08:00,900 S1: security startups, will be gaining access to unified customer context. 133 00:08:00,940 --> 00:08:04,620 S1: This is all especially relevant to cybersecurity because security use 134 00:08:04,620 --> 00:08:09,340 S1: cases really, really benefit from contexts, their identity, actions, history 135 00:08:09,340 --> 00:08:12,940 S1: of the thing you're investigating right across multiple systems. Also, 136 00:08:12,980 --> 00:08:17,180 S1: there's the issue of securing the UCC, since it will 137 00:08:17,220 --> 00:08:20,660 S1: be the absolute most sensitive data store in the entire company, 138 00:08:21,220 --> 00:08:24,860 S1: all the juiciest bits of the company all in one place, 139 00:08:25,340 --> 00:08:30,380 S1: which is like a red teamer or attacker's dream. Cloudflare 140 00:08:30,420 --> 00:08:35,100 S1: launched an AI labyrinth feature that messes with unauthorized AI 141 00:08:35,140 --> 00:08:38,980 S1: scrapers by feeding them endless pages of irrelevant but real 142 00:08:39,020 --> 00:08:42,300 S1: looking content instead of actually blocking them. So it's like 143 00:08:42,300 --> 00:08:47,160 S1: a classic, like honeypot slash deception thing. Uh, to counter 144 00:08:47,160 --> 00:08:51,480 S1: rogue AI that won't honor the robots.txt file. A rush 145 00:08:51,480 --> 00:08:56,120 S1: release of JFK assassination files exposed 400 Social Security numbers 146 00:08:56,120 --> 00:09:01,079 S1: and other sensitive data belonging to former congressional staffers, many 147 00:09:01,080 --> 00:09:04,920 S1: of which of whom are actually high ranking officials now. 148 00:09:05,520 --> 00:09:10,000 S1: But we just doxed their Social Security numbers in this release. 149 00:09:10,000 --> 00:09:15,160 S1: New cybersecurity compensation research shows high six figure salaries are 150 00:09:15,160 --> 00:09:19,480 S1: not stopping 60% of security professionals from thinking about leaving 151 00:09:19,480 --> 00:09:23,360 S1: their job within a year. National security OpenAI is pressuring 152 00:09:23,360 --> 00:09:27,240 S1: the Trump administration to allow copyright scraping for AI training, 153 00:09:27,400 --> 00:09:31,280 S1: claiming America will lose the AI race to China without 154 00:09:31,280 --> 00:09:34,079 S1: having full access to scrape. And a lot of people 155 00:09:34,080 --> 00:09:37,240 S1: see this as like, corporate bullshit, trying to use security 156 00:09:37,240 --> 00:09:42,620 S1: to give them a corporate advantage. But Unfortunately, it's actually true. 157 00:09:42,660 --> 00:09:46,300 S1: This is actually true because China has no limitations whatsoever 158 00:09:46,340 --> 00:09:49,180 S1: on what it trains on. On what it crawls, they 159 00:09:49,179 --> 00:09:54,900 S1: steal whatever, consume whatever with 100% free reign. And that 160 00:09:54,900 --> 00:09:58,620 S1: is a path or an accelerator to get to AGI 161 00:09:58,620 --> 00:10:01,179 S1: or ASI or whatever. So the question is, who do 162 00:10:01,179 --> 00:10:03,900 S1: you want to have AGI or ASI first? The US 163 00:10:03,940 --> 00:10:06,660 S1: or China? And for me, Trump makes that question a 164 00:10:06,660 --> 00:10:08,980 S1: little bit harder to answer. But my answer is still 165 00:10:08,980 --> 00:10:13,819 S1: the US. Americans are buying overseas residency and citizenship as 166 00:10:13,820 --> 00:10:17,700 S1: a hedge against uncertainty in the US. China unveiled a 167 00:10:17,700 --> 00:10:22,459 S1: deep sea cable cutting device capable of severing undersea communications 168 00:10:22,860 --> 00:10:27,580 S1: at depths twice beyond where existing infrastructure operates, so they 169 00:10:27,580 --> 00:10:30,540 S1: can find deeper cables that are further down in a 170 00:10:30,540 --> 00:10:34,260 S1: trench or whatever. It says unveiled. I wrote that word 171 00:10:34,580 --> 00:10:37,220 S1: based on the story, but it's kind of weird to like, 172 00:10:37,460 --> 00:10:39,400 S1: but I'm like, is this a press release or are 173 00:10:39,400 --> 00:10:42,719 S1: they like, you know, showing off the cool toy that 174 00:10:42,720 --> 00:10:45,400 S1: they have and I don't know, I'm thinking of like CES, 175 00:10:45,440 --> 00:10:48,400 S1: they're like, come check out the cable cutter. London's Heathrow 176 00:10:48,400 --> 00:10:51,600 S1: Airport announced a full day shutdown after a significant fire 177 00:10:51,600 --> 00:10:54,640 S1: at a nearby electrical substation knocked out power to the 178 00:10:54,640 --> 00:10:56,800 S1: entire facility. A lot of questions are going to be 179 00:10:56,800 --> 00:11:01,119 S1: asked about that. I. French Swash Relays Arc Prize Foundation 180 00:11:01,120 --> 00:11:04,160 S1: released a new AI intelligence test. I think this is 181 00:11:04,160 --> 00:11:07,760 S1: just another version of Arc, but the best AI models 182 00:11:07,760 --> 00:11:12,680 S1: are currently only scoring 1% while humans get 60%. That 183 00:11:12,679 --> 00:11:19,320 S1: is fantastic. Anthropic's Claude, not Anthropic's Anthropic's cloud has finally 184 00:11:19,360 --> 00:11:23,079 S1: added I hate saying finally seems rude and trite, but whatever. 185 00:11:23,320 --> 00:11:27,120 S1: Added web search to its AI chatbot catching up to 186 00:11:27,160 --> 00:11:31,640 S1: ChatGPT with clickable citations. I really want this in the API. 187 00:11:31,679 --> 00:11:34,600 S1: It's not there yet, and they're apparently using Brave Search 188 00:11:34,600 --> 00:11:38,689 S1: to power that, uh, search capability. Gmail is rolling out 189 00:11:38,690 --> 00:11:41,969 S1: an AI powered search that ranks results based on relevance, 190 00:11:41,970 --> 00:11:44,690 S1: instead of just showing the newest emails first. This is cool, 191 00:11:44,690 --> 00:11:48,369 S1: but I want AI filtering and AI auto drafts. Like, 192 00:11:48,410 --> 00:11:51,890 S1: why don't they get more Gemini into Gmail? It seems 193 00:11:51,890 --> 00:11:56,130 S1: like they have the ability. Can't wait for that to happen. Technology. 194 00:11:56,170 --> 00:11:59,490 S1: Apple is updating AirPods Max next month to add lossless 195 00:11:59,490 --> 00:12:03,770 S1: and ultra low latency audio. Matt Claude says that while 196 00:12:03,970 --> 00:12:06,850 S1: type flags make sense for terminal commands, you should use 197 00:12:06,850 --> 00:12:10,929 S1: the dash dash force style option, the long option instead 198 00:12:10,929 --> 00:12:15,410 S1: for better readability. Someone wrote Seth Larson wrote this thing 199 00:12:15,410 --> 00:12:18,650 S1: I fear for the unauthenticated Web argues that the increasingly 200 00:12:18,650 --> 00:12:22,890 S1: common sign in to continue message on websites is destroying 201 00:12:22,890 --> 00:12:25,610 S1: the open promise of the web. I think we talked 202 00:12:25,610 --> 00:12:28,450 S1: about that last week as well. Nvidia says they're investing 203 00:12:28,450 --> 00:12:32,010 S1: hundreds of billions of dollars in the US manufactured chips 204 00:12:32,490 --> 00:12:35,790 S1: over the next four years, shifting away from Asia and 205 00:12:35,790 --> 00:12:39,430 S1: Trump's tariff threats. This is exactly what Trump was trying 206 00:12:39,429 --> 00:12:42,750 S1: to do with his policies, and it's positive. But I'm 207 00:12:42,750 --> 00:12:45,390 S1: worried that the damage will actually be worse than the benefit. 208 00:12:45,390 --> 00:12:49,990 S1: NYPD has dramatically expanded its drone program, sending them to 209 00:12:50,030 --> 00:12:56,790 S1: thousands of. 911 911 911 calls. While privacy advocates are 210 00:12:56,790 --> 00:13:00,350 S1: worried about the lack of transparency, and they're basically saying 211 00:13:00,350 --> 00:13:03,710 S1: this could be used for surveillance humans. New research from 212 00:13:03,750 --> 00:13:07,150 S1: Aalto University suggests Earth has way more people than the 213 00:13:07,150 --> 00:13:12,910 S1: official 8.2 billion, due to major undercounting in rural areas. 214 00:13:12,950 --> 00:13:15,670 S1: I wonder how major they're talking about. Are they talking about, like, 215 00:13:15,710 --> 00:13:19,109 S1: up to 9 billion? Like that would an extra billion? 216 00:13:19,110 --> 00:13:22,390 S1: Are we losing an extra billion? Curious what they I 217 00:13:22,390 --> 00:13:26,190 S1: didn't find the actual number that they're re estimating. It 218 00:13:26,190 --> 00:13:28,989 S1: looked like they just said it was bad. Tyler Cowen 219 00:13:29,230 --> 00:13:32,429 S1: shares insights from his conversation with Ezra Klein about the 220 00:13:32,429 --> 00:13:36,370 S1: new book, abundance. Okay, so check this out. You have 221 00:13:36,370 --> 00:13:39,970 S1: to go check out this book. I just listened to the, um, 222 00:13:40,050 --> 00:13:45,330 S1: podcast with Lex and the co-author of the book with him, with, uh, 223 00:13:45,330 --> 00:13:51,450 S1: Ezra Klein. And it is fantastic, in fact, that Technological 224 00:13:51,450 --> 00:13:56,610 S1: Republic by Alex Karp combined with this is, um, these 225 00:13:56,610 --> 00:14:01,170 S1: are very centrist books. It's Alex Karp being a lot 226 00:14:01,170 --> 00:14:04,090 S1: more liberal because he's considered a kind of. Right. It's 227 00:14:04,090 --> 00:14:07,330 S1: being a lot more liberal. And then this book is 228 00:14:07,330 --> 00:14:10,450 S1: from two liberals, but they're talking a lot more centrist, 229 00:14:10,490 --> 00:14:14,410 S1: a lot more kind of Republican about the government and 230 00:14:14,410 --> 00:14:17,330 S1: how to move forward and all the mistakes that the 231 00:14:17,330 --> 00:14:21,170 S1: liberals have made that enabled Trump to come to power. 232 00:14:21,690 --> 00:14:25,410 S1: So basically, if you are a centrist or you like 233 00:14:26,010 --> 00:14:30,010 S1: first principle thinking, uh, you like to figure out, like 234 00:14:30,010 --> 00:14:32,430 S1: why the government is broken. Like what we could do 235 00:14:32,430 --> 00:14:35,550 S1: to fix it. Read this book, abundance and read the 236 00:14:35,550 --> 00:14:40,670 S1: book The Technological Republic. They are fantastic books. And to 237 00:14:40,710 --> 00:14:43,150 S1: be clear, I haven't read the book actually, but I've 238 00:14:43,150 --> 00:14:46,750 S1: listened to hours of analysis on it, and I've. They've 239 00:14:46,750 --> 00:14:48,630 S1: kind of already covered the whole thing, but I'm still 240 00:14:48,630 --> 00:14:51,710 S1: going to read it cover to cover, telling staff that 241 00:14:51,710 --> 00:14:54,070 S1: the way to get ahead is not to accumulate a 242 00:14:54,070 --> 00:14:56,550 S1: giant fiefdom, and AI is going to do the same. 243 00:14:56,550 --> 00:14:59,870 S1: It's going to clean all that out. David Kellogg explains 244 00:14:59,870 --> 00:15:02,870 S1: the essential differences between a manager, director and a VP, 245 00:15:03,310 --> 00:15:06,790 S1: with the VP being accountable for results regardless of who 246 00:15:06,830 --> 00:15:10,910 S1: approved the plan. Isn't that the case for director as well? 247 00:15:11,350 --> 00:15:16,430 S1: Jonathan Kipnis and his team discovered that rejuvenating the brain's 248 00:15:16,430 --> 00:15:20,950 S1: lymphatic vessels improves memory in old mice by helping clear 249 00:15:20,950 --> 00:15:24,990 S1: waste that contributes to cognitive decline. So my question is, 250 00:15:24,990 --> 00:15:27,070 S1: how do I do this for me? Because I'm not 251 00:15:27,070 --> 00:15:32,450 S1: a mouse. Yeah. How do you rejuvenate the brain's lymphatic vessels? 252 00:15:33,250 --> 00:15:35,730 S1: Is there like a pill? Is that like a workout? 253 00:15:35,770 --> 00:15:39,330 S1: What do you do? Discovery. Most bitter people you will 254 00:15:39,330 --> 00:15:42,730 S1: ever meet. Gut punching. Three paragraph essay. I'm not going 255 00:15:42,770 --> 00:15:45,810 S1: to read the whole thing. Go check it out. Delphi 256 00:15:45,890 --> 00:15:48,410 S1: AI new platform lets you create and share a digital 257 00:15:48,410 --> 00:15:53,890 S1: clone of yourself. Lange Magnus no Lange Manus new open 258 00:15:53,890 --> 00:15:56,410 S1: source tool that makes it easier to build autonomous agents 259 00:15:56,410 --> 00:16:00,290 S1: using Lange chain and Lange graph. Pure clever new browser 260 00:16:00,290 --> 00:16:03,850 S1: hack lets you read any Paywalled content. It's very much 261 00:16:03,850 --> 00:16:07,610 S1: similar to AI. Rise of Agentic AI is out which 262 00:16:07,610 --> 00:16:10,770 S1: I had a chance to contribute to. Definitely go check 263 00:16:10,770 --> 00:16:13,850 S1: that out in the links. Personal best. Neat little tool 264 00:16:13,850 --> 00:16:17,250 S1: that shows you which personal blogs are most popular on 265 00:16:17,250 --> 00:16:19,650 S1: Hacker News. So I'm going to mine this thing and 266 00:16:19,650 --> 00:16:22,650 S1: actually put them all into threshold. So they will be 267 00:16:22,650 --> 00:16:26,570 S1: in threshold soon. Someone says they recommend against brave. I 268 00:16:26,570 --> 00:16:29,340 S1: thought that was a was pretty good arguments. I don't 269 00:16:29,340 --> 00:16:31,580 S1: use it, so it wasn't super pertinent to me, but 270 00:16:31,620 --> 00:16:35,619 S1: worth sharing. Circuit tutor. Neat little tool lets you describe 271 00:16:35,620 --> 00:16:38,860 S1: simple circuits in plain English, and get both schematics and 272 00:16:38,860 --> 00:16:45,500 S1: interactive explanations for folks who need e refreshers. Go act. 273 00:16:45,540 --> 00:16:48,180 S1: New tool that turns your text or files into a 274 00:16:48,180 --> 00:16:54,620 S1: browser based explainer. Videos and OS. New GitHub Osint tool 275 00:16:54,620 --> 00:16:59,860 S1: that scrapes public user info, including emails, organizations and repositories. 276 00:16:59,900 --> 00:17:02,060 S1: All right, this is the end of the standard edition 277 00:17:02,060 --> 00:17:04,900 S1: of the podcast, which includes just the news items for 278 00:17:04,900 --> 00:17:07,260 S1: the week. To get the rest of the episode, which 279 00:17:07,300 --> 00:17:10,580 S1: includes my analysis, the discovery section that has all of 280 00:17:10,580 --> 00:17:14,219 S1: the coolest tools and articles, I found this week, the 281 00:17:14,220 --> 00:17:16,860 S1: recommendation of the week and the aphorism of the week. 282 00:17:17,220 --> 00:17:21,500 S1: Please consider becoming a member. As a member, you'll get 283 00:17:21,500 --> 00:17:25,260 S1: lots of different things, from access to our extraordinary community 284 00:17:25,320 --> 00:17:28,960 S1: of over a thousand brilliant and kind people in industries 285 00:17:28,960 --> 00:17:34,240 S1: like cybersecurity, AI, technology and the humanities. You also get 286 00:17:34,280 --> 00:17:38,440 S1: access to the UL Book Club, dedicated member content and events, 287 00:17:38,640 --> 00:17:42,080 S1: and lots and lots more. Plus, you'll get a dedicated 288 00:17:42,080 --> 00:17:44,800 S1: podcast feed you can put in your client that gets 289 00:17:44,800 --> 00:17:48,560 S1: you the full member edition of this podcast. To become 290 00:17:48,560 --> 00:17:51,040 S1: a member and get all of that, just head over 291 00:17:51,040 --> 00:17:58,520 S1: to Daniel Meister. That's Daniel COVID-19. We'll see you next time. 292 00:18:00,280 --> 00:18:03,879 S1: Unsupervised learning is produced on Hindenburg Pro using an SM 293 00:18:03,880 --> 00:18:07,440 S1: seven B microphone. A video version of the podcast is 294 00:18:07,440 --> 00:18:11,200 S1: available on the Unsupervised Learning YouTube channel, and the text 295 00:18:11,200 --> 00:18:14,480 S1: version with full links and notes is available at Daniel 296 00:18:14,840 --> 00:18:18,280 S1: Comm Slash newsletter. We'll see you next time.