1 00:00:15,760 --> 00:00:18,160 Speaker 1: Welcome to Tech Stuff. I'm os Volosha. I'm thrilled to 2 00:00:18,200 --> 00:00:20,759 Speaker 1: announce that the Weekend Tech is back and it's growing. 3 00:00:21,360 --> 00:00:23,680 Speaker 1: And instead of Karen and I are accounting the essential 4 00:00:23,760 --> 00:00:27,480 Speaker 1: news today and every Friday from now on, I'm going 5 00:00:27,480 --> 00:00:30,160 Speaker 1: to be joined by three of the best writers covering 6 00:00:30,200 --> 00:00:33,160 Speaker 1: Silicon Valley. And if you've been listening to the last 7 00:00:33,159 --> 00:00:37,320 Speaker 1: few Fridays, today's guests will be familiar names. Read Albergotti, 8 00:00:37,400 --> 00:00:40,640 Speaker 1: technology editor for Semaphore and a resident of the Bay Area, 9 00:00:41,120 --> 00:00:42,080 Speaker 1: always in the thick of it. 10 00:00:42,120 --> 00:00:43,800 Speaker 2: Read welcome back, thanks for having me. 11 00:00:43,920 --> 00:00:47,080 Speaker 1: Kyle Chaker, writes the Infinite Scroll newsletter for The New Yorker. 12 00:00:47,560 --> 00:00:50,720 Speaker 1: Kyle's a resident of Washington, DC, but hopefully not a 13 00:00:50,800 --> 00:00:53,800 Speaker 1: victim of the disastrous polymarket pop up could occurred there 14 00:00:53,880 --> 00:00:55,920 Speaker 1: last weekend? Or Kyle, were you left in the dark? 15 00:00:59,200 --> 00:01:02,120 Speaker 1: And Natasha Teak, who of the Washington Post. She haunts 16 00:01:02,200 --> 00:01:05,640 Speaker 1: the halls and perhaps sometimes even the consciences of the 17 00:01:05,640 --> 00:01:10,880 Speaker 1: big tech companies in Silay. Welcome back. Hey everyone, you've 18 00:01:10,880 --> 00:01:13,080 Speaker 1: all joined us before, so the format will come as 19 00:01:13,120 --> 00:01:16,120 Speaker 1: no surprise. But remember this is a discussion, so please 20 00:01:16,120 --> 00:01:19,240 Speaker 1: butt in. We have no fear of crosstalk, and you 21 00:01:19,280 --> 00:01:21,919 Speaker 1: three already have the inside scoop. We love a humble 22 00:01:21,959 --> 00:01:24,240 Speaker 1: brag and even a not so humble brag, So don't 23 00:01:24,240 --> 00:01:27,880 Speaker 1: be shy. Let's get into it. Nitisha will start with you. 24 00:01:28,720 --> 00:01:31,400 Speaker 1: There was big news out of Los Angeles courtroom earlier 25 00:01:31,400 --> 00:01:34,560 Speaker 1: this week. What was the trial about and what surprised 26 00:01:34,600 --> 00:01:36,319 Speaker 1: you most about the verdict. 27 00:01:36,400 --> 00:01:40,240 Speaker 3: The trial was about. It was a Bellweather case looking 28 00:01:40,360 --> 00:01:45,319 Speaker 3: at social media addiction, and the last two companies remaining 29 00:01:45,360 --> 00:01:49,040 Speaker 3: a couple of them settled were YouTube and Meta, and 30 00:01:49,120 --> 00:01:53,120 Speaker 3: the jury found that Meta and YouTube were liable for 31 00:01:54,560 --> 00:01:58,520 Speaker 3: product liability. Basically, it's this novel argument that the set 32 00:01:58,720 --> 00:02:02,320 Speaker 3: of lawyers has brought kind of using the same the 33 00:02:02,360 --> 00:02:05,840 Speaker 3: same argument that they use for big tobacco. You know, 34 00:02:05,880 --> 00:02:09,919 Speaker 3: it's a way to get around Section two thirty, which 35 00:02:10,240 --> 00:02:13,360 Speaker 3: which is a shield that tech companies use. You know, 36 00:02:13,360 --> 00:02:16,800 Speaker 3: they're not liable for user generated content. So the jury 37 00:02:16,880 --> 00:02:19,320 Speaker 3: was instructed not to look at any of the content 38 00:02:19,360 --> 00:02:22,680 Speaker 3: in particular, but think about, you know, the product decisions 39 00:02:22,680 --> 00:02:28,480 Speaker 3: that Meta and YouTube had made, infinite scroll notifications, beauty filters. 40 00:02:28,320 --> 00:02:30,680 Speaker 1: And what made I think that the case particularly interesting 41 00:02:30,680 --> 00:02:33,520 Speaker 1: in this respect was the discovery of the internal communications 42 00:02:33,600 --> 00:02:35,440 Speaker 1: right within the companies. That was kind of a key 43 00:02:35,880 --> 00:02:36,680 Speaker 1: thing for the jury. 44 00:02:36,919 --> 00:02:40,000 Speaker 3: Yeah, it was. It was really damning because they had 45 00:02:40,440 --> 00:02:43,920 Speaker 3: you know, Meta had data looking at like thirty percent 46 00:02:44,240 --> 00:02:48,360 Speaker 3: of I think ten to eleven year olds in the 47 00:02:48,520 --> 00:02:51,640 Speaker 3: US are on Instagram, and they knew that even though 48 00:02:51,720 --> 00:02:55,160 Speaker 3: they have an age limit of thirteen. That is I 49 00:02:55,160 --> 00:02:58,840 Speaker 3: think one of the stats that when Mark Zuckerberg was testifying, 50 00:02:59,040 --> 00:03:01,080 Speaker 3: you know, there was a like a lot of tension 51 00:03:01,160 --> 00:03:05,320 Speaker 3: between him and the lawyer who was questioning him over 52 00:03:05,440 --> 00:03:08,120 Speaker 3: this kind of stuff. You know, he kept kind of 53 00:03:08,160 --> 00:03:11,640 Speaker 3: pushing back and saying, like you're misrepresenting this, like you know, 54 00:03:11,680 --> 00:03:15,800 Speaker 3: this wasn't intentional. He said, you know, the idea of 55 00:03:15,880 --> 00:03:18,440 Speaker 3: banning beauty filters, which was something that came up again 56 00:03:18,480 --> 00:03:20,920 Speaker 3: and again because it was related to the plaintiff in 57 00:03:20,919 --> 00:03:23,880 Speaker 3: the case. He said, it would have been paternalistic to 58 00:03:23,960 --> 00:03:27,280 Speaker 3: do something like that, and they just stopped like automatically 59 00:03:27,440 --> 00:03:29,560 Speaker 3: offering them. I mean to me, it was just like 60 00:03:30,680 --> 00:03:35,160 Speaker 3: I remember when we first started talking about these features 61 00:03:35,200 --> 00:03:37,680 Speaker 3: as addictive, Like I was looking back, I was working 62 00:03:37,720 --> 00:03:39,640 Speaker 3: at Wired at the time, Like it was like a 63 00:03:39,680 --> 00:03:43,640 Speaker 3: Wired Guide to Internet addiction twenty eighteen, right, Like, it's 64 00:03:43,720 --> 00:03:48,280 Speaker 3: almost a decade later talking about some of these same 65 00:03:48,720 --> 00:03:51,760 Speaker 3: features and the world has it feels like the world 66 00:03:51,800 --> 00:03:53,960 Speaker 3: has moved on, right, Like we like some of the 67 00:03:53,960 --> 00:03:57,840 Speaker 3: same people involved in this fight about dark patterns and 68 00:03:57,880 --> 00:04:01,000 Speaker 3: addictive social media, they're now obsessed with AI. They are 69 00:04:01,200 --> 00:04:04,840 Speaker 3: on Capitol Hill lobbying about AI. And here we are, 70 00:04:05,200 --> 00:04:06,720 Speaker 3: you know, looking at as. 71 00:04:07,160 --> 00:04:09,240 Speaker 1: Litigating yesterday's news in some respect to. 72 00:04:09,640 --> 00:04:12,480 Speaker 3: Like last decades news, you know, in a way, I 73 00:04:12,520 --> 00:04:17,080 Speaker 3: mean there's there's billions of people and young people still 74 00:04:17,160 --> 00:04:20,840 Speaker 3: on social media. It's just it's hard not to feel 75 00:04:20,880 --> 00:04:25,560 Speaker 3: like the paltriness of the of the verdict in a way, 76 00:04:26,200 --> 00:04:28,599 Speaker 3: you know, because what they what was on trial is 77 00:04:28,640 --> 00:04:32,800 Speaker 3: whether or not they you know, inflicted personal harm on 78 00:04:32,960 --> 00:04:39,360 Speaker 3: the plaintiff. She suffered from body dysmorphia, suicidal ideations, self harm, 79 00:04:39,600 --> 00:04:43,640 Speaker 3: and Meta was arguing that it was you know, it 80 00:04:43,720 --> 00:04:47,920 Speaker 3: was obviously like addiction and mental health issues, they're multi 81 00:04:47,920 --> 00:04:51,400 Speaker 3: factor things. They talked about abuse she faced as a child. 82 00:04:52,000 --> 00:04:56,200 Speaker 3: They looked at her therapy on the stand, right, yeah, 83 00:04:56,240 --> 00:04:58,520 Speaker 3: I mean, just the idea of Meta bringing in your 84 00:04:58,520 --> 00:05:01,960 Speaker 3: therapy sessions is like, you know, it just feels like 85 00:05:01,960 --> 00:05:04,320 Speaker 3: I said like kind of inadequate to the task, Like 86 00:05:04,360 --> 00:05:07,560 Speaker 3: the jury wasn't voting on Hey, did Meta kind of 87 00:05:07,800 --> 00:05:12,040 Speaker 3: mislead the American people about the age limits on their app? 88 00:05:12,120 --> 00:05:14,839 Speaker 3: And you know, did they know that these features and 89 00:05:14,880 --> 00:05:17,719 Speaker 3: product designs could be addictive? That's like not the verdict. 90 00:05:17,800 --> 00:05:21,280 Speaker 3: You know, the verdict is like around addiction, which is 91 00:05:21,360 --> 00:05:22,400 Speaker 3: just so much harder to. 92 00:05:24,240 --> 00:05:24,600 Speaker 1: I think it. 93 00:05:24,880 --> 00:05:26,800 Speaker 3: You know, that's part of the reason why you see 94 00:05:26,880 --> 00:05:33,599 Speaker 3: even some critics and consumer advocates not happy with with 95 00:05:33,720 --> 00:05:35,600 Speaker 3: what ultimately what the jury decided. 96 00:05:36,240 --> 00:05:38,320 Speaker 1: And so read you were saying before we started take things, 97 00:05:38,320 --> 00:05:39,839 Speaker 1: sorry to cut you off, that you were very interested 98 00:05:39,880 --> 00:05:43,360 Speaker 1: in Natasha's take, and natashually used the word my holistics. 99 00:05:43,400 --> 00:05:44,200 Speaker 2: So I would have. 100 00:05:44,160 --> 00:05:46,120 Speaker 1: Thought this would be like kind of a course for celebration. 101 00:05:46,279 --> 00:05:47,640 Speaker 1: But I mean, read, what was your question. 102 00:05:48,600 --> 00:05:50,800 Speaker 4: I was going to ask you if you think this 103 00:05:50,880 --> 00:05:53,560 Speaker 4: will stand up on appeal, you know, and does this 104 00:05:53,640 --> 00:05:55,400 Speaker 4: does this ultimately go to the Supreme Court? 105 00:05:55,600 --> 00:05:56,840 Speaker 2: What happens? 106 00:05:57,080 --> 00:05:59,240 Speaker 3: I mean, I don't know, I'm not enough of a 107 00:05:59,320 --> 00:06:01,560 Speaker 3: legal expert to say that, but in terms of like 108 00:06:01,600 --> 00:06:05,600 Speaker 3: looking at the effect on MetaStock, wondering you know, what's 109 00:06:05,640 --> 00:06:10,359 Speaker 3: going on in open Ai or any of the Chinese 110 00:06:10,400 --> 00:06:14,000 Speaker 3: apps today. Based on the verdict, I would say almost nothing. 111 00:06:14,120 --> 00:06:17,560 Speaker 3: Like I just was looking back at my coworker, Naomi 112 00:06:17,640 --> 00:06:20,480 Speaker 3: Nicks had this great piece with some internal documents from 113 00:06:20,560 --> 00:06:24,320 Speaker 3: Meta and it was Adam Mosei, the head of Instagram, 114 00:06:24,360 --> 00:06:27,760 Speaker 3: talking about how they needed to focus on teens. That 115 00:06:27,880 --> 00:06:30,200 Speaker 3: was like their your number one business priority as you're 116 00:06:30,240 --> 00:06:33,520 Speaker 3: looking for twenty twenty four. This document he sent a 117 00:06:33,520 --> 00:06:37,280 Speaker 3: few weeks after these cases were initially filed, So like, 118 00:06:37,360 --> 00:06:41,480 Speaker 3: what is going on internally inside these other companies? Like 119 00:06:41,520 --> 00:06:44,800 Speaker 3: I think the best hope is that, you know, maybe 120 00:06:44,800 --> 00:06:48,640 Speaker 3: it makes open Ai a little bit more cautious. I mean, 121 00:06:48,720 --> 00:06:51,800 Speaker 3: you know, there's sixteen hundred other cases. This is a 122 00:06:51,800 --> 00:06:54,280 Speaker 3: bell Weather case. You know, this was just one plaintiff. 123 00:06:55,040 --> 00:06:58,719 Speaker 3: They're trying to get a sense of how Jeuris will 124 00:06:58,720 --> 00:07:04,000 Speaker 3: respond to this snap had settled. There's lots of permutations 125 00:07:04,040 --> 00:07:06,960 Speaker 3: this could go. But in terms of like genuinely changing 126 00:07:07,360 --> 00:07:12,000 Speaker 3: the product decisions and practices and the mentality behind these 127 00:07:12,040 --> 00:07:16,120 Speaker 3: companies that are their financial incentive is growth, right, Like 128 00:07:16,160 --> 00:07:17,640 Speaker 3: part of what they were saying is we need to 129 00:07:17,680 --> 00:07:21,280 Speaker 3: get them as tweens in order to keep them as teenagers. 130 00:07:21,280 --> 00:07:23,520 Speaker 3: This is what the internal documents at Meta but Instagram 131 00:07:23,520 --> 00:07:23,800 Speaker 3: are saying. 132 00:07:23,880 --> 00:07:27,520 Speaker 1: But Natasha, you use the word Bellwether twice. Bellwether suggests 133 00:07:27,560 --> 00:07:28,360 Speaker 1: this is meaningful. 134 00:07:28,720 --> 00:07:32,120 Speaker 3: This is meaningful. It's bell Weather For the other cases, 135 00:07:32,920 --> 00:07:36,200 Speaker 3: you know, will will the plaintiffs get some kind of 136 00:07:36,320 --> 00:07:38,720 Speaker 3: day in court? Will they get justice? Will they get 137 00:07:38,920 --> 00:07:46,080 Speaker 3: some monetary restitution? I am talking about. Will decisions in 138 00:07:46,160 --> 00:07:49,640 Speaker 3: multi probably trillion by the time they feel the impact 139 00:07:49,680 --> 00:07:52,280 Speaker 3: of this multi trillion dollar companies be made so that 140 00:07:52,480 --> 00:07:59,720 Speaker 3: everyone can benefit from a sense of accountability for your decisions. 141 00:08:00,200 --> 00:08:02,559 Speaker 4: Well, maybe I can be the nihilist because I think 142 00:08:02,760 --> 00:08:05,000 Speaker 4: you know if you look, if you look it. 143 00:08:06,480 --> 00:08:08,360 Speaker 1: I want to be the anti nihilist. I mean, I've 144 00:08:08,400 --> 00:08:10,360 Speaker 1: been fought not nearly as clarity as any of you, 145 00:08:10,400 --> 00:08:12,360 Speaker 1: but I've been following this stuff for ten years. And 146 00:08:12,680 --> 00:08:15,880 Speaker 1: the idea that courts would accept an argument that goes 147 00:08:15,880 --> 00:08:19,400 Speaker 1: around Section two thirty hold the technology companies liable for 148 00:08:19,440 --> 00:08:22,880 Speaker 1: the algorithms they create and the addictive behaviors those algorithms 149 00:08:23,320 --> 00:08:27,200 Speaker 1: promote and provide monetary relief, even though it's peanuts I think, 150 00:08:27,240 --> 00:08:30,800 Speaker 1: six billion dollars. But nonetheless, I mean, how did it 151 00:08:30,840 --> 00:08:32,840 Speaker 1: start with smoking. How did it start with big Tobacco. 152 00:08:32,840 --> 00:08:34,160 Speaker 1: I mean, I think probably I don't know if there 153 00:08:34,160 --> 00:08:36,719 Speaker 1: was one huge case that changed the whole industry or 154 00:08:36,760 --> 00:08:37,880 Speaker 1: if it, but you know, there was a kind of 155 00:08:37,880 --> 00:08:41,439 Speaker 1: groundswell of bell Weather cases like this one that started 156 00:08:41,480 --> 00:08:42,360 Speaker 1: to promote real change. 157 00:08:42,440 --> 00:08:45,000 Speaker 4: I think that the nihilist views like big tobacco still 158 00:08:45,120 --> 00:08:47,520 Speaker 4: went on. You know, they're still around today, you know, 159 00:08:47,559 --> 00:08:50,040 Speaker 4: and they've gone into other areas like food, you know, 160 00:08:50,160 --> 00:08:54,880 Speaker 4: packaged food. So and basically a bunch of a bunch 161 00:08:54,920 --> 00:08:58,280 Speaker 4: of class action attorneys got rich and you know, and 162 00:08:58,320 --> 00:09:00,960 Speaker 4: if people stopped smoking, it wasn't because of those cases. 163 00:09:01,040 --> 00:09:04,040 Speaker 4: It was because people were educated on the harms of 164 00:09:04,120 --> 00:09:06,880 Speaker 4: cigarettes and they and society change, right. 165 00:09:06,960 --> 00:09:08,079 Speaker 2: I mean, well, that's. 166 00:09:07,880 --> 00:09:10,480 Speaker 3: Because they were forced to change the marketing. I think 167 00:09:10,520 --> 00:09:12,599 Speaker 3: part of what makes me, you know, part of my 168 00:09:12,679 --> 00:09:14,920 Speaker 3: interpretation of the verdict is also the fact that on 169 00:09:14,960 --> 00:09:17,800 Speaker 3: the same day Trump named Mark Zuckerberg to his like 170 00:09:17,880 --> 00:09:21,640 Speaker 3: Science Council. And you know, I you know, I've been 171 00:09:21,679 --> 00:09:24,400 Speaker 3: following these companies also very closely for a long time, 172 00:09:24,480 --> 00:09:28,160 Speaker 3: and I know that, you know, bad press is something 173 00:09:28,200 --> 00:09:31,160 Speaker 3: that really motivates them. And we're living in a time 174 00:09:31,320 --> 00:09:33,400 Speaker 3: where it doesn't. 175 00:09:33,120 --> 00:09:35,640 Speaker 1: Matter so much anymore. And in fact, if you can 176 00:09:35,640 --> 00:09:38,840 Speaker 1: go on TPPN or on all in, you know, whatever 177 00:09:38,880 --> 00:09:41,200 Speaker 1: the mainstream media say about you is probably way less 178 00:09:41,200 --> 00:09:42,920 Speaker 1: relevant than it was under the Biden administration. 179 00:09:43,000 --> 00:09:45,240 Speaker 5: Suddenly, but it feels like the past, Like it does 180 00:09:45,280 --> 00:09:48,280 Speaker 5: feel like legislating ten years ago, when the platforms are 181 00:09:48,320 --> 00:09:51,319 Speaker 5: already moving on and we're experiencing new problems. 182 00:09:51,640 --> 00:09:54,560 Speaker 1: But I'm looking, I'm looking, I'm looking thirstily at you 183 00:09:54,640 --> 00:09:57,000 Speaker 1: to give us an optimistic take, because presumably, when you 184 00:09:57,000 --> 00:09:59,440 Speaker 1: were writing your book Infinite Scroll, this was the type 185 00:09:59,480 --> 00:10:02,800 Speaker 1: of lawsuit and outcome that you will, in some sense 186 00:10:02,880 --> 00:10:04,120 Speaker 1: hoping might happen totally. 187 00:10:04,160 --> 00:10:07,000 Speaker 5: I mean, I think it's nice to see the mechanics 188 00:10:07,040 --> 00:10:10,240 Speaker 5: of a platform separated from the content in some ways, 189 00:10:10,280 --> 00:10:14,080 Speaker 5: Like we should be directly looking at the mechanisms of 190 00:10:14,120 --> 00:10:18,080 Speaker 5: a feed and the algorithms and the recommendations and how 191 00:10:18,160 --> 00:10:21,480 Speaker 5: you get bombarded with content as a user, like in 192 00:10:21,800 --> 00:10:25,680 Speaker 5: section two thirty, like it's supposed to protect user generated content, 193 00:10:25,800 --> 00:10:28,800 Speaker 5: but there are so like protect platforms from being responsible 194 00:10:28,840 --> 00:10:32,480 Speaker 5: for that content, but there are so many borderline editorial 195 00:10:32,520 --> 00:10:36,360 Speaker 5: decisions I think that digital platforms make that shapes how 196 00:10:36,360 --> 00:10:38,720 Speaker 5: we consume things and what we consume and how we 197 00:10:38,760 --> 00:10:42,760 Speaker 5: are addicted to our platforms and feeds. So like that 198 00:10:43,040 --> 00:10:46,560 Speaker 5: gives me some optimism, even though the six million dollar 199 00:10:47,240 --> 00:10:51,880 Speaker 5: metric is so minusculess to seem like a joke unfortunately. 200 00:10:51,880 --> 00:10:52,160 Speaker 2: I think. 201 00:10:52,240 --> 00:10:56,920 Speaker 4: I mean, coincidentally, you know, cigarettes are coming back now, right, 202 00:10:57,080 --> 00:11:00,679 Speaker 4: like it's a they're making a comeback, and like, you know, 203 00:11:00,920 --> 00:11:04,439 Speaker 4: maybe it's not a coincidence, honestly, No, I'm talking about 204 00:11:04,480 --> 00:11:07,280 Speaker 4: actual cigarettes, the kind that you light on fire and 205 00:11:07,280 --> 00:11:09,200 Speaker 4: inhale the smoke. It's a it's a it's a new 206 00:11:09,280 --> 00:11:13,720 Speaker 4: it's a trend, right, it's cool again, and especially with 207 00:11:13,840 --> 00:11:15,000 Speaker 4: young people. 208 00:11:14,760 --> 00:11:17,359 Speaker 1: Celebs are lighting up all over the place. 209 00:11:17,280 --> 00:11:18,160 Speaker 2: Back in Hollywood. 210 00:11:18,200 --> 00:11:20,280 Speaker 4: I mean, I just think it's like there are there 211 00:11:20,320 --> 00:11:22,839 Speaker 4: are vices that people have, and like this, I mean 212 00:11:22,920 --> 00:11:25,520 Speaker 4: clearly social media is like a vice that you know, 213 00:11:25,600 --> 00:11:27,880 Speaker 4: people sit there, We've all done it, like just you're 214 00:11:27,920 --> 00:11:30,280 Speaker 4: just scrolling and you're going, what am I doing right now? 215 00:11:30,320 --> 00:11:32,480 Speaker 4: This is horrible? And you just keep doing it and 216 00:11:32,520 --> 00:11:35,240 Speaker 4: it's like, you know, we all know it's bad. And 217 00:11:35,280 --> 00:11:37,320 Speaker 4: I think I think society is like building up the 218 00:11:37,360 --> 00:11:39,920 Speaker 4: scar tissue or like you know, the defenses for it 219 00:11:40,040 --> 00:11:44,120 Speaker 4: right now, and that's totally separate from these legal cases 220 00:11:44,160 --> 00:11:45,360 Speaker 4: that are that are happening. 221 00:11:45,520 --> 00:11:46,960 Speaker 5: What if posting is the vice? 222 00:11:47,080 --> 00:11:47,320 Speaker 2: Though? 223 00:11:48,520 --> 00:11:53,160 Speaker 5: Sorry, If posting was the vice, if we legislated against posting, 224 00:11:53,320 --> 00:11:55,560 Speaker 5: then no one could consume and it would be all good. 225 00:11:55,800 --> 00:11:58,160 Speaker 4: Hey, if you want to make posting illegal, I would 226 00:11:58,200 --> 00:12:01,040 Speaker 4: be the happiest person on earth because I posting. 227 00:12:02,440 --> 00:12:03,120 Speaker 2: A naga. 228 00:12:03,760 --> 00:12:08,440 Speaker 3: I mean, posting is the vice because of intermittent variable rewards, right, 229 00:12:08,559 --> 00:12:11,199 Speaker 3: Like they know that you you know, you'll just keep 230 00:12:11,240 --> 00:12:14,880 Speaker 3: pulling down to refresh if if you think that you're 231 00:12:15,000 --> 00:12:17,839 Speaker 3: going to get some kind of feedback. But I mean, 232 00:12:17,880 --> 00:12:20,920 Speaker 3: I think what Kyle's bringing up about the focus on 233 00:12:21,120 --> 00:12:24,040 Speaker 3: product design is actually why I feel. I think I 234 00:12:24,080 --> 00:12:26,800 Speaker 3: feel like such a sense of optimism that if people 235 00:12:26,920 --> 00:12:32,760 Speaker 3: understood the mechanics of how these how these social networks work, 236 00:12:32,880 --> 00:12:36,440 Speaker 3: what they what the designers know, how that how it functions. 237 00:12:36,480 --> 00:12:39,320 Speaker 3: Like I remember when I learned the term intermittent variable reward. 238 00:12:39,400 --> 00:12:40,200 Speaker 3: Has it stopped me? 239 00:12:40,600 --> 00:12:40,680 Speaker 2: No? 240 00:12:40,880 --> 00:12:43,440 Speaker 3: Have I continued to waste like years of my life 241 00:12:43,559 --> 00:12:44,400 Speaker 3: on these networks? 242 00:12:44,520 --> 00:12:44,760 Speaker 2: Yes? 243 00:12:45,120 --> 00:12:47,679 Speaker 3: But yeah, but I know what's happening to me, you know. 244 00:12:47,800 --> 00:12:51,920 Speaker 1: And to me, there's something different about a lawsuit though, 245 00:12:51,960 --> 00:12:55,120 Speaker 1: where you get discovery of dozens of thousands of pages 246 00:12:55,120 --> 00:12:57,600 Speaker 1: of documents where the executives are saying to each other, 247 00:12:57,840 --> 00:13:00,560 Speaker 1: we know this is causing harm. Essentially, let's continue to 248 00:13:00,880 --> 00:13:03,640 Speaker 1: lean in versus the odd like Cassandra who does a 249 00:13:03,720 --> 00:13:06,200 Speaker 1: Netflix doc or like the open essay from a like 250 00:13:06,240 --> 00:13:08,959 Speaker 1: former employee, like I think that these lawsuits in the 251 00:13:09,000 --> 00:13:10,640 Speaker 1: aggregate will have an effect. I mean there was another 252 00:13:10,679 --> 00:13:13,240 Speaker 1: one in New Mexico that Meta also lost, relating to 253 00:13:14,240 --> 00:13:17,560 Speaker 1: not protecting children from sexual predation on the platform. That 254 00:13:17,600 --> 00:13:19,760 Speaker 1: was a much bigger settlement I think. So I'm going 255 00:13:19,800 --> 00:13:21,560 Speaker 1: to I'm going to push back a little bit against 256 00:13:21,559 --> 00:13:23,719 Speaker 1: the against the tide of these. 257 00:13:23,760 --> 00:13:26,520 Speaker 4: Well, what about Australia that band it right? I mean 258 00:13:26,559 --> 00:13:29,240 Speaker 4: they just banned it under sixteen you can't use it. 259 00:13:30,040 --> 00:13:32,480 Speaker 5: I mean, like more is happening elsewhere. 260 00:13:33,000 --> 00:13:35,000 Speaker 3: I mean, this is what I'm thinking about. You Are 261 00:13:35,440 --> 00:13:40,400 Speaker 3: you are inside the like you know, product marketing design 262 00:13:40,480 --> 00:13:44,160 Speaker 3: team at chatchipt, Like what is this? What is this 263 00:13:44,320 --> 00:13:46,600 Speaker 3: verdict doing to you? Are you just not sutting it 264 00:13:46,640 --> 00:13:49,640 Speaker 3: in email? Are you not putting it in writing? Or 265 00:13:49,720 --> 00:13:53,360 Speaker 3: are you actually saying or or are you just not 266 00:13:53,480 --> 00:13:54,480 Speaker 3: studying addiction. 267 00:13:54,760 --> 00:13:58,440 Speaker 4: You're probably discussing it. You're probably discussing it with chatchipt 268 00:13:58,760 --> 00:14:01,199 Speaker 4: and I cannot wait for the discovery on that. 269 00:14:01,520 --> 00:14:03,040 Speaker 2: That's going to be unbelievable. 270 00:14:03,520 --> 00:14:06,200 Speaker 3: I mean I think that this the sorry, the extent 271 00:14:06,400 --> 00:14:09,800 Speaker 3: of the internal documents that came to public light. I 272 00:14:09,840 --> 00:14:12,920 Speaker 3: mean this is like you know, your your right oes, 273 00:14:13,040 --> 00:14:15,679 Speaker 3: Like if if it was twenty seventeen and we got 274 00:14:15,720 --> 00:14:20,560 Speaker 3: access to this, it would have felt like vindication, partly 275 00:14:20,640 --> 00:14:23,720 Speaker 3: because it is really hard on an individual level to 276 00:14:23,920 --> 00:14:27,800 Speaker 3: stand up to some of these dark patterns. I think, 277 00:14:27,920 --> 00:14:32,200 Speaker 3: you know, we feel that keenly, probably in the way 278 00:14:32,200 --> 00:14:33,160 Speaker 3: that we spend our time. 279 00:14:33,600 --> 00:14:36,200 Speaker 4: Well, Natasha, you and I spent like all weekend, that 280 00:14:36,240 --> 00:14:39,400 Speaker 4: one weekend looking through all those Facebook documents, right that 281 00:14:39,560 --> 00:14:41,280 Speaker 4: was like do you remember that, like the whole the 282 00:14:41,320 --> 00:14:42,480 Speaker 4: Francis hougen leak. 283 00:14:43,080 --> 00:14:44,400 Speaker 3: I was actually on Matt leave. 284 00:14:44,480 --> 00:14:48,840 Speaker 2: But oh yeah, just this is just as wild dreams. 285 00:14:49,000 --> 00:14:52,320 Speaker 1: So can I just ask you before we count of 286 00:14:52,360 --> 00:14:56,200 Speaker 1: this story that there were two quite cinematic dare I 287 00:14:56,240 --> 00:14:58,520 Speaker 1: say it scenes in this little suit. One involved a 288 00:14:58,640 --> 00:15:01,200 Speaker 1: jaw sut of em and M's involved a pair of 289 00:15:01,280 --> 00:15:01,720 Speaker 1: ray bends. 290 00:15:01,960 --> 00:15:05,120 Speaker 3: So the jar full of eminems. This was this the 291 00:15:05,120 --> 00:15:08,520 Speaker 3: plaintiff's lawyer that I mentioned, Mark Lanier. I'm not sure 292 00:15:08,560 --> 00:15:10,680 Speaker 3: if I'm pronouncing his name right. So he takes out 293 00:15:10,720 --> 00:15:14,640 Speaker 3: a jar of Eminem's and he's like, this represents revenue. 294 00:15:14,800 --> 00:15:17,560 Speaker 3: And he takes out a handful of eminems and he's 295 00:15:17,640 --> 00:15:20,040 Speaker 3: just like, you know, they won't even feel it. This 296 00:15:20,200 --> 00:15:22,240 Speaker 3: like won't even make a difference. And then at some 297 00:15:22,320 --> 00:15:25,800 Speaker 3: point he also like like chips, like takes a little 298 00:15:26,320 --> 00:15:28,680 Speaker 3: bit of the of the hard candy shell off and 299 00:15:28,720 --> 00:15:31,080 Speaker 3: he's like, two hundred million. They won't even know this 300 00:15:31,120 --> 00:15:33,680 Speaker 3: is happening. But there was an interview I think in 301 00:15:33,720 --> 00:15:35,680 Speaker 3: the New York Times with a couple of the jurors 302 00:15:35,720 --> 00:15:39,400 Speaker 3: as they were leaving, and they said they especially did 303 00:15:39,440 --> 00:15:42,400 Speaker 3: not want to do what the lawyer seemed to be 304 00:15:42,640 --> 00:15:45,520 Speaker 3: hinting towards, which is like give them a massive settlement 305 00:15:45,640 --> 00:15:49,920 Speaker 3: or sorry, a massive payment, because they said they wanted 306 00:15:49,920 --> 00:15:53,240 Speaker 3: to focus a little bit more on the issues, you know, 307 00:15:53,320 --> 00:15:56,520 Speaker 3: thinking about some of these product decisions and the impact 308 00:15:56,560 --> 00:15:57,960 Speaker 3: on users. 309 00:15:57,960 --> 00:16:00,040 Speaker 1: Interesting and the ray Bens. 310 00:16:00,440 --> 00:16:04,000 Speaker 3: Oh, the ray bands, Yeah, as you what did you 311 00:16:04,040 --> 00:16:04,680 Speaker 3: think about that? 312 00:16:05,160 --> 00:16:07,640 Speaker 1: Well? I heard that some of the Facebook segs turned 313 00:16:07,720 --> 00:16:11,120 Speaker 1: up to court wearing the metro rate pians recording the 314 00:16:11,160 --> 00:16:15,040 Speaker 1: proceedings that had to be told off. Again to your 315 00:16:15,040 --> 00:16:17,680 Speaker 1: point about like whether or not the tech executives care 316 00:16:17,760 --> 00:16:22,200 Speaker 1: about like making the you know, judicial system feel good 317 00:16:22,240 --> 00:16:23,800 Speaker 1: in the way they may have done a few years ago, 318 00:16:23,920 --> 00:16:26,080 Speaker 1: Like when Mark put his suit on, it feels like 319 00:16:26,080 --> 00:16:27,640 Speaker 1: we're in a different, different era. 320 00:16:28,040 --> 00:16:30,840 Speaker 2: Wait, I did exactly, just like I was a court reporter. 321 00:16:30,920 --> 00:16:32,920 Speaker 4: I used to have to hand in well actually I 322 00:16:32,960 --> 00:16:34,720 Speaker 4: was one of the only people who didn't have to 323 00:16:34,720 --> 00:16:36,600 Speaker 4: hand in their cell phone in the Southern District. But 324 00:16:36,600 --> 00:16:39,080 Speaker 4: I always thought that was, like, why can't people know 325 00:16:39,120 --> 00:16:42,080 Speaker 4: what's happening inside the courts? Like what's what's the big deal? 326 00:16:42,440 --> 00:16:45,200 Speaker 3: I wish I had an eminem visual, you know. But 327 00:16:45,280 --> 00:16:48,760 Speaker 3: they also the same lawyer also asked, like Mark if 328 00:16:48,760 --> 00:16:53,760 Speaker 3: he'd gone over the plaintiff's account and then proceeded to 329 00:16:53,840 --> 00:16:57,000 Speaker 3: pull out like a thirty five foot wall of her 330 00:16:57,040 --> 00:17:00,840 Speaker 3: selfies and like asked him to to look at it. 331 00:17:00,920 --> 00:17:04,240 Speaker 3: So he has obviously a flair for the dramatic, which yeah, 332 00:17:04,359 --> 00:17:06,800 Speaker 3: only the meta ray bands caught I guess. 333 00:17:08,920 --> 00:17:12,400 Speaker 1: Okay, moving right to along. Open Ai made an announcement 334 00:17:12,440 --> 00:17:15,720 Speaker 1: this week that I quite found quite surprising. No more 335 00:17:16,040 --> 00:17:19,240 Speaker 1: Mickey Mouse ai videos. What's going on here? 336 00:17:19,359 --> 00:17:19,560 Speaker 2: Yeah? 337 00:17:19,600 --> 00:17:22,000 Speaker 4: I mean they so they got rid of Sora, their 338 00:17:22,080 --> 00:17:26,159 Speaker 4: their video generation tool. And what I thought was so 339 00:17:26,280 --> 00:17:28,520 Speaker 4: interesting about that it's it's funny because it kind of 340 00:17:28,560 --> 00:17:31,000 Speaker 4: ties in with what Natasha was talking about with with 341 00:17:31,119 --> 00:17:34,119 Speaker 4: what is open ai going to do now with social media? 342 00:17:34,160 --> 00:17:35,879 Speaker 4: There was sort of a social media app that was 343 00:17:35,920 --> 00:17:38,400 Speaker 4: part of But I thought was what was so interesting 344 00:17:38,560 --> 00:17:41,679 Speaker 4: was like, what the real reason is that they just 345 00:17:41,760 --> 00:17:44,639 Speaker 4: can't They don't have enough tokens to go around, right, 346 00:17:44,800 --> 00:17:48,840 Speaker 4: and they're seeing this there's a huge profitable you know 347 00:17:49,000 --> 00:17:51,600 Speaker 4: boom right now, like with open Claw, which we talked 348 00:17:51,640 --> 00:17:53,880 Speaker 4: about last week, you know, just causing people to use 349 00:17:53,960 --> 00:17:56,680 Speaker 4: way way more compute and they're kind of seeing where 350 00:17:56,720 --> 00:17:59,440 Speaker 4: this is heading and they have to making tough decisions 351 00:17:59,480 --> 00:18:01,560 Speaker 4: about where they want to allocate their compute and this 352 00:18:01,720 --> 00:18:04,600 Speaker 4: was not a profitable area. Like it's just you know, 353 00:18:04,640 --> 00:18:07,560 Speaker 4: they had this deal with Disney people were you know, 354 00:18:07,640 --> 00:18:10,359 Speaker 4: people liked making these videos, but it just took up 355 00:18:10,400 --> 00:18:13,639 Speaker 4: too much compute power for what it actually produced and 356 00:18:14,440 --> 00:18:16,679 Speaker 4: what it meant and profit. And so for me, it 357 00:18:16,760 --> 00:18:19,240 Speaker 4: was sort of another Bell weather about where the industry 358 00:18:19,320 --> 00:18:22,000 Speaker 4: is headed with AI, Like I think we were having 359 00:18:22,000 --> 00:18:24,000 Speaker 4: it was funny like a few months ago. I think 360 00:18:24,040 --> 00:18:26,440 Speaker 4: it feels like we were having a debate about whether 361 00:18:26,520 --> 00:18:28,199 Speaker 4: this was a bubble and the whole thing was going 362 00:18:28,240 --> 00:18:30,520 Speaker 4: to crash down. And now it's like there is so 363 00:18:30,760 --> 00:18:33,480 Speaker 4: much demand and it just keeps going up and up 364 00:18:33,520 --> 00:18:36,160 Speaker 4: and up, and I find it just it's just Incredible's 365 00:18:36,280 --> 00:18:36,920 Speaker 4: what's happening. 366 00:18:37,119 --> 00:18:42,480 Speaker 5: It's just not demands for personalized slop videos, so demand 367 00:18:42,560 --> 00:18:45,520 Speaker 5: for doing your coding for you and like organizing your email. 368 00:18:46,000 --> 00:18:47,199 Speaker 2: Yeah, maybe that's a good thing. 369 00:18:47,720 --> 00:18:50,480 Speaker 4: It's not like as harmful as social media when it's 370 00:18:50,480 --> 00:18:53,159 Speaker 4: like you're just trying to get your your AI agent 371 00:18:53,240 --> 00:18:53,800 Speaker 4: to check your. 372 00:18:53,720 --> 00:18:54,399 Speaker 2: Email for you. 373 00:18:54,440 --> 00:18:59,080 Speaker 4: But it is you know, obviously there's gonna be there's 374 00:18:59,119 --> 00:19:01,399 Speaker 4: going to be slops like this is not going to 375 00:19:01,480 --> 00:19:05,399 Speaker 4: go away. But it's just like there's a huge battle 376 00:19:05,480 --> 00:19:08,800 Speaker 4: right now between these companies over who has the most compute. 377 00:19:08,880 --> 00:19:11,560 Speaker 4: I think, like I remember Dario amid Abas, I. 378 00:19:11,560 --> 00:19:14,479 Speaker 1: Mean, this is literaltru but this is literally like they 379 00:19:14,520 --> 00:19:17,679 Speaker 1: don't have enough data sent to Alas to do sora 380 00:19:17,880 --> 00:19:19,440 Speaker 1: is that is that what this comes down to when 381 00:19:19,440 --> 00:19:21,160 Speaker 1: you say there's not enough tokens. 382 00:19:20,840 --> 00:19:22,760 Speaker 4: I think they don't have enough data center hours to 383 00:19:22,880 --> 00:19:25,200 Speaker 4: just do everything they want to do, and they have 384 00:19:25,280 --> 00:19:27,960 Speaker 4: to make decisions and it's like how much are we 385 00:19:28,000 --> 00:19:30,160 Speaker 4: going to put into training? It's a it's a very 386 00:19:30,240 --> 00:19:33,000 Speaker 4: it's a zero sum game within the companies because they 387 00:19:33,040 --> 00:19:36,560 Speaker 4: all only have so much compute. It's also l yeah, 388 00:19:36,600 --> 00:19:38,600 Speaker 4: and it's also a zero sum game, right, I mean 389 00:19:38,680 --> 00:19:41,119 Speaker 4: we could talk about that too, but it's also like 390 00:19:41,160 --> 00:19:45,160 Speaker 4: a zero game between the companies, like if whoever gets 391 00:19:45,200 --> 00:19:48,800 Speaker 4: whoever locks up the compute today is that means like 392 00:19:48,880 --> 00:19:53,280 Speaker 4: that compute is not available to their competitors tomorrow. And 393 00:19:53,520 --> 00:19:55,359 Speaker 4: there's and there's only so much that can be made, 394 00:19:55,440 --> 00:19:57,840 Speaker 4: like there's just there's only so many like extreme ultra 395 00:19:57,920 --> 00:20:00,960 Speaker 4: violet lithography machines that could be a year. So it's 396 00:20:01,000 --> 00:20:03,720 Speaker 4: like to me, it's it's like one of it's almost 397 00:20:03,800 --> 00:20:06,119 Speaker 4: like a wartime economy, like we're just gonna have to 398 00:20:06,119 --> 00:20:07,040 Speaker 4: start ratcheting things. 399 00:20:07,080 --> 00:20:08,480 Speaker 2: And I find it. 400 00:20:08,480 --> 00:20:10,720 Speaker 4: It's it's like incredible, and I don't think there's enough 401 00:20:10,800 --> 00:20:14,480 Speaker 4: like attention in society like that part of the just 402 00:20:14,560 --> 00:20:17,760 Speaker 4: this sheer economic issue that that this is all bringing 403 00:20:17,840 --> 00:20:18,720 Speaker 4: up Natasha. 404 00:20:18,760 --> 00:20:20,120 Speaker 3: Okay, can I can I ask a question? 405 00:20:20,160 --> 00:20:20,360 Speaker 1: Though? 406 00:20:20,760 --> 00:20:24,200 Speaker 3: So I I was just like scrolling on on Soa 407 00:20:24,240 --> 00:20:27,080 Speaker 3: this morning, which I haven't looked at since it came out, 408 00:20:27,240 --> 00:20:29,639 Speaker 3: and this this is what I'm curious about, Like open 409 00:20:29,680 --> 00:20:36,160 Speaker 3: Ai has there has a foot in absolutely everything, right 410 00:20:36,200 --> 00:20:38,840 Speaker 3: and obviously the you know, even when it came out 411 00:20:38,880 --> 00:20:42,680 Speaker 3: when Soura was very popular, they could barely afford it. 412 00:20:42,680 --> 00:20:46,560 Speaker 3: It sounded like, you know, to keep Soa going. But like, 413 00:20:46,800 --> 00:20:50,439 Speaker 3: is the demand for Sora in particular so high, because 414 00:20:50,520 --> 00:20:53,360 Speaker 3: I think it like totally fell off. I was wondering 415 00:20:53,400 --> 00:20:55,439 Speaker 3: how they even got it to the top of the 416 00:20:55,440 --> 00:20:58,080 Speaker 3: app store when it came out, because like, this is 417 00:20:58,119 --> 00:21:01,400 Speaker 3: a company that is so obsessed with Twitter and has 418 00:21:01,480 --> 00:21:06,080 Speaker 3: not yet, I think, figured out how to market to 419 00:21:06,440 --> 00:21:09,120 Speaker 3: people who are on other social networks. Like if you look, 420 00:21:09,160 --> 00:21:11,760 Speaker 3: you know the main characters that they let you play 421 00:21:11,800 --> 00:21:15,480 Speaker 3: around with, like are one of the Sora creators, Bill Peebles, right, 422 00:21:15,560 --> 00:21:18,240 Speaker 3: like Sam the s guy Gabriel from open Ai And 423 00:21:18,320 --> 00:21:20,520 Speaker 3: I was looking and they haven't posted for four months, 424 00:21:20,960 --> 00:21:23,640 Speaker 3: you know. So I'm not saying that like video Generation, 425 00:21:23,880 --> 00:21:26,760 Speaker 3: I think like Iran has shown us like how valuable 426 00:21:26,920 --> 00:21:31,199 Speaker 3: and coveted or the you know, the fruit narratives, the 427 00:21:31,240 --> 00:21:33,680 Speaker 3: fruit sex narratives like on if we're talking about what's 428 00:21:33,680 --> 00:21:36,640 Speaker 3: happening this week. But Sora itself, I just don't think 429 00:21:36,720 --> 00:21:40,200 Speaker 3: that it I don't think it was like particularly great, 430 00:21:40,320 --> 00:21:42,600 Speaker 3: and I don't think that the dynamics of the app 431 00:21:42,640 --> 00:21:46,000 Speaker 3: were very compelling. So it might sound very good for 432 00:21:46,119 --> 00:21:49,040 Speaker 3: them to say like we're buckling down, but basically they 433 00:21:49,040 --> 00:21:52,520 Speaker 3: had no discipline. They were trying to do everything. Now 434 00:21:52,840 --> 00:21:56,160 Speaker 3: Fiji's come in right and said like, hey, we're we're losing. 435 00:21:56,880 --> 00:21:59,200 Speaker 3: You know, we're not competitive on the things that people 436 00:21:59,280 --> 00:22:04,320 Speaker 3: really are sessed with JATGBT coding, and so let's reallocate 437 00:22:04,359 --> 00:22:06,600 Speaker 3: our compute. But like I would like it corrective on 438 00:22:06,640 --> 00:22:08,560 Speaker 3: the on the Sora is very popular. 439 00:22:08,920 --> 00:22:10,360 Speaker 1: I want to come to you, Kyle, and I want 440 00:22:10,359 --> 00:22:12,760 Speaker 1: you to put your cultural historian head on and take 441 00:22:12,840 --> 00:22:15,160 Speaker 1: us all the way back to October twenty twenty five 442 00:22:15,200 --> 00:22:18,080 Speaker 1: the mists of Time, when Sora was the number one 443 00:22:18,160 --> 00:22:21,639 Speaker 1: app in the ATA Giant talk about sore from the 444 00:22:21,680 --> 00:22:26,159 Speaker 1: point of view as a cultural observer and pretty good technology. 445 00:22:26,240 --> 00:22:29,160 Speaker 5: I mean it was I think it attempted to be 446 00:22:29,359 --> 00:22:32,840 Speaker 5: a social network of generative AI, Like the point was 447 00:22:32,920 --> 00:22:36,160 Speaker 5: that you would make your own videos with AI. It's 448 00:22:36,280 --> 00:22:40,000 Speaker 5: major mechanic was that you could transform yourself into an 449 00:22:40,080 --> 00:22:44,399 Speaker 5: AI doll essentially and have your friends reanimate you in 450 00:22:44,520 --> 00:22:49,800 Speaker 5: crazy different situations. The example was Sam Altman was one 451 00:22:49,840 --> 00:22:52,800 Speaker 5: of these digital puppets, and many people put him in 452 00:22:52,960 --> 00:22:56,600 Speaker 5: videos trying to steal chips from like a time or 453 00:22:56,680 --> 00:23:00,040 Speaker 5: getting your desk. So and it's like that is a 454 00:23:00,080 --> 00:23:04,280 Speaker 5: horrifying use case, Like I'm not sure who actually wanted 455 00:23:04,400 --> 00:23:06,840 Speaker 5: to be this digital puppet, and I think they were 456 00:23:07,400 --> 00:23:10,960 Speaker 5: betting too hard on this being a form of viral content, 457 00:23:11,560 --> 00:23:14,760 Speaker 5: Like we still kind of haven't figured out. I mean, 458 00:23:14,760 --> 00:23:18,280 Speaker 5: maybe it's the slopaganda Iran war videos, but we haven't 459 00:23:18,320 --> 00:23:20,840 Speaker 5: figured out like what is the unit of AI content 460 00:23:20,960 --> 00:23:23,720 Speaker 5: that is the most successful. So SOA was like an 461 00:23:23,720 --> 00:23:26,600 Speaker 5: attempt to figure that out. But I think what it 462 00:23:26,680 --> 00:23:29,160 Speaker 5: ended up proving was that no one wants the feed 463 00:23:29,240 --> 00:23:33,720 Speaker 5: of only slop. Like now we see it dispersed into 464 00:23:33,800 --> 00:23:36,679 Speaker 5: our Twitter feeds or on Instagram or TikTok or whatever, 465 00:23:36,720 --> 00:23:39,080 Speaker 5: and that's kind of fun, but seeing it wall to 466 00:23:39,200 --> 00:23:42,320 Speaker 5: wall is like, I don't know, eating only those eminem's 467 00:23:42,440 --> 00:23:42,920 Speaker 5: or something. 468 00:23:43,480 --> 00:23:46,159 Speaker 4: I never thought that the social media part of it 469 00:23:46,240 --> 00:23:49,240 Speaker 4: was interesting, to be honest, Like you see these like, oh, 470 00:23:49,280 --> 00:23:51,480 Speaker 4: we rose to the top of the app store, like 471 00:23:51,560 --> 00:23:53,639 Speaker 4: anybody can be at the top of the app store 472 00:23:53,680 --> 00:23:56,040 Speaker 4: for a day. It's like the New York Times bestseller list, 473 00:23:56,119 --> 00:23:58,320 Speaker 4: like you can buy a box of books and like 474 00:23:58,400 --> 00:23:59,800 Speaker 4: get yourself in the best sellerst for. 475 00:24:00,560 --> 00:24:02,760 Speaker 1: That story is slightly more compositive, sadly than. 476 00:24:05,280 --> 00:24:08,159 Speaker 4: Like, don't pay attention to these stories. Like but like 477 00:24:08,280 --> 00:24:10,600 Speaker 4: it's you're totally right, Like both of you are totally right. 478 00:24:10,640 --> 00:24:12,600 Speaker 4: It's just like that's not the most interesting part of 479 00:24:12,600 --> 00:24:16,679 Speaker 4: the story. Like, yeah, companies make failed social media attempts 480 00:24:16,720 --> 00:24:19,119 Speaker 4: every day and they you know, it's it doesn't matter. 481 00:24:19,560 --> 00:24:20,760 Speaker 1: But I want to, I want to. I want to 482 00:24:20,760 --> 00:24:23,120 Speaker 1: push back on you, on you again though, because if 483 00:24:23,119 --> 00:24:25,360 Speaker 1: I look at last year, remember Scott Galloway is saying, oh, 484 00:24:25,480 --> 00:24:27,840 Speaker 1: like that, you know, the studio Ghibli moment and Sora 485 00:24:27,920 --> 00:24:29,399 Speaker 1: like there's a total waste of time. They should be 486 00:24:29,440 --> 00:24:32,600 Speaker 1: focusing their resources on more like profitable things, which is 487 00:24:32,640 --> 00:24:34,600 Speaker 1: what you're saying. Read it as well. Yeah, on the 488 00:24:34,680 --> 00:24:37,800 Speaker 1: other hand, like outside of the like Silicon Valley and 489 00:24:37,960 --> 00:24:41,080 Speaker 1: New York like Silicon Valley Technology in New York Finance Bubble, 490 00:24:41,960 --> 00:24:44,520 Speaker 1: AI wasn't really it was just an idea, right, And 491 00:24:44,560 --> 00:24:47,920 Speaker 1: then with that Studio Ghibli moment in March last year 492 00:24:48,040 --> 00:24:51,520 Speaker 1: and with Sora, like every regular person in the world 493 00:24:51,640 --> 00:24:54,080 Speaker 1: got to see the magic of AI themselves and like 494 00:24:54,160 --> 00:24:57,560 Speaker 1: experience like making a Sam Waltman puppet or like turning 495 00:24:57,560 --> 00:24:59,919 Speaker 1: their family photos into pixarl. That may not have been 496 00:25:00,080 --> 00:25:03,120 Speaker 1: ultimately very helpful of open AI's business, but in terms 497 00:25:03,160 --> 00:25:06,640 Speaker 1: of opening people's eyes to the way this technology can 498 00:25:06,680 --> 00:25:09,439 Speaker 1: create magic, maybe people will look back and say it 499 00:25:09,480 --> 00:25:11,280 Speaker 1: wasn't so great for open ai, but actually from a 500 00:25:11,280 --> 00:25:13,800 Speaker 1: cultural point of view, that was quite impactful. 501 00:25:14,160 --> 00:25:17,879 Speaker 2: Yeah, I hear that. It's a valid argument. I think I. 502 00:25:17,880 --> 00:25:20,000 Speaker 3: Would say like the argument at the time was that 503 00:25:20,040 --> 00:25:23,359 Speaker 3: they discovered something because they were saying, hey, look we 504 00:25:23,560 --> 00:25:26,760 Speaker 3: understand people, because people want to see themselves in the videos, 505 00:25:27,119 --> 00:25:30,840 Speaker 3: and that's why you know this app that will allow 506 00:25:30,920 --> 00:25:33,040 Speaker 3: you to make a doll of yourself, Like that's why 507 00:25:33,040 --> 00:25:35,360 Speaker 3: we're winning and other people aren't. And I just think 508 00:25:35,359 --> 00:25:39,280 Speaker 3: it's very interesting that they did not understand the dynamics 509 00:25:39,440 --> 00:25:43,280 Speaker 3: of like why people use it. They used they built 510 00:25:43,280 --> 00:25:46,399 Speaker 3: a good tool, but people still want to share it 511 00:25:46,440 --> 00:25:49,080 Speaker 3: with their friends. And I think it's important to point 512 00:25:49,119 --> 00:25:53,439 Speaker 3: out when people who want to like change the you know, 513 00:25:53,560 --> 00:25:56,439 Speaker 3: landscape of the Internet and append the economy don't like 514 00:25:56,480 --> 00:25:59,439 Speaker 3: actually understand how people think and work. 515 00:26:00,280 --> 00:26:03,320 Speaker 5: And there are other video models that are really successful, 516 00:26:03,440 --> 00:26:06,159 Speaker 5: like like AI generated video is huge and it's a 517 00:26:06,240 --> 00:26:09,639 Speaker 5: huge medium, but it's just like not everyone needs to 518 00:26:09,680 --> 00:26:12,520 Speaker 5: be making AI clips or wants to be making these things. 519 00:26:12,680 --> 00:26:15,680 Speaker 5: And to Natasha's point, you might not want to see 520 00:26:15,680 --> 00:26:18,080 Speaker 5: yourself in them. You want to see Donald Trump, like 521 00:26:18,160 --> 00:26:22,359 Speaker 5: you want to see copyrighted or famous individuals who you 522 00:26:22,400 --> 00:26:26,600 Speaker 5: want to manipulate, like dolls, And that's much more problematic content, 523 00:26:26,760 --> 00:26:30,720 Speaker 5: like harder to defend legally. I would bet I don't. 524 00:26:30,760 --> 00:26:33,200 Speaker 3: I mean, I don't know, Like I saw someone say, 525 00:26:33,320 --> 00:26:35,120 Speaker 3: you know, oh, maybe in a week or two we'll 526 00:26:35,160 --> 00:26:39,520 Speaker 3: find out that, like, you know, something was on this network. 527 00:26:39,680 --> 00:26:42,159 Speaker 3: Like usually the problem with a lot of this stuff 528 00:26:42,200 --> 00:26:45,280 Speaker 3: is seesam, right, So that's like the absolute worst nightmare, 529 00:26:45,359 --> 00:26:48,200 Speaker 3: like AI generated se sam that somebody's creating. I'm just 530 00:26:48,280 --> 00:26:53,360 Speaker 3: saying like that they have this kind of monolith idea, 531 00:26:53,560 --> 00:26:57,840 Speaker 3: like if you like generating studio ghibli images or videos, 532 00:26:57,880 --> 00:26:59,920 Speaker 3: then you must want like Kyle was saying, like this 533 00:27:00,040 --> 00:27:02,240 Speaker 3: feed where it's all of that, and people are really 534 00:27:02,240 --> 00:27:05,000 Speaker 3: thinking about this as a tool right on TikTok. People 535 00:27:05,000 --> 00:27:08,959 Speaker 3: are using it super creatively, you know, to come up 536 00:27:08,960 --> 00:27:12,000 Speaker 3: with different characters and narratives. I think people want to 537 00:27:12,080 --> 00:27:16,240 Speaker 3: incorporate this technology. Incorporate it, you know, not not like 538 00:27:16,520 --> 00:27:18,480 Speaker 3: live or die by it read. 539 00:27:18,520 --> 00:27:20,160 Speaker 1: I want to give the last word to you on this, 540 00:27:20,359 --> 00:27:24,679 Speaker 1: on this story you've written in your newsletter quote, the 541 00:27:24,880 --> 00:27:28,400 Speaker 1: entire economy and global supply chain is being reoriented right 542 00:27:28,440 --> 00:27:32,040 Speaker 1: now around a single commodity, the token, and most people 543 00:27:32,040 --> 00:27:34,600 Speaker 1: are only vaguely aware it's happening or how it will 544 00:27:34,600 --> 00:27:36,480 Speaker 1: affect their lives. And I just want you to ask 545 00:27:36,480 --> 00:27:38,040 Speaker 1: you to explain that. But you said a couple of 546 00:27:38,040 --> 00:27:40,040 Speaker 1: times we were kind of missing the main story by 547 00:27:40,040 --> 00:27:43,320 Speaker 1: talking about the social aspect of soor so. What is like, 548 00:27:43,400 --> 00:27:45,159 Speaker 1: what is the big takeaway here for you? 549 00:27:45,600 --> 00:27:45,879 Speaker 2: Yeah? 550 00:27:45,960 --> 00:27:47,960 Speaker 4: I mean, first of all, like I agree with everything 551 00:27:47,960 --> 00:27:50,960 Speaker 4: you're saying about this stuff, Like I think their product 552 00:27:51,000 --> 00:27:53,920 Speaker 4: sense here is kind of it is really off right. 553 00:27:54,280 --> 00:27:56,640 Speaker 4: But then at the same time, like I mean, yeah, 554 00:27:56,840 --> 00:28:00,520 Speaker 4: AI video creation is a popular thing that is going 555 00:28:00,560 --> 00:28:03,119 Speaker 4: to continue. It's going to change Hollywood, like it's for 556 00:28:03,240 --> 00:28:05,040 Speaker 4: you know, there are lots of other companies doing this, 557 00:28:06,600 --> 00:28:09,840 Speaker 4: but I think the what the really important thing to 558 00:28:09,880 --> 00:28:13,280 Speaker 4: look at is like there's just only so many resources 559 00:28:13,280 --> 00:28:15,600 Speaker 4: to go around, and they have to make these tough 560 00:28:15,600 --> 00:28:18,720 Speaker 4: decisions and focus on what actually makes sense from a 561 00:28:18,920 --> 00:28:22,000 Speaker 4: product perspective, and that which brings like to this broader 562 00:28:22,040 --> 00:28:26,720 Speaker 4: point like the economy orienting around AI. It's like we're 563 00:28:26,920 --> 00:28:30,960 Speaker 4: we're seeing like massive shortages now of you know, like memory, 564 00:28:31,160 --> 00:28:34,040 Speaker 4: and of course there's questions about what's going to happen 565 00:28:34,119 --> 00:28:35,160 Speaker 4: with energy prices. 566 00:28:35,560 --> 00:28:37,240 Speaker 1: I don't know when you say memory, you don't mean 567 00:28:37,320 --> 00:28:40,080 Speaker 1: data centers, you mean literally like stick memory sticks of 568 00:28:40,120 --> 00:28:41,480 Speaker 1: full time is more expensive than they. 569 00:28:41,440 --> 00:28:44,120 Speaker 4: Will sticks of memory, right, which go into everything right 570 00:28:44,160 --> 00:28:47,000 Speaker 4: from your cars to you know, your Nintendo switch. Like 571 00:28:47,040 --> 00:28:51,520 Speaker 4: it's just consumer product prices are growing up like way 572 00:28:51,600 --> 00:28:55,240 Speaker 4: beyond what you would expect with inflation, and I think 573 00:28:55,280 --> 00:28:57,520 Speaker 4: it's just going to continue, Like what's the next thing? 574 00:28:57,600 --> 00:28:59,160 Speaker 2: Is it copper? Is it other stuff? 575 00:28:59,240 --> 00:29:03,280 Speaker 4: So this is actually gonna have a real impact on 576 00:29:03,960 --> 00:29:06,720 Speaker 4: like what we're able to buy and how much things cost. 577 00:29:06,840 --> 00:29:08,480 Speaker 4: In a way, that like these are like the most 578 00:29:08,520 --> 00:29:11,600 Speaker 4: important things in society, right, like our pocketbooks, like how 579 00:29:11,640 --> 00:29:14,840 Speaker 4: much money is in our bank account, Like that's that's 580 00:29:14,880 --> 00:29:17,240 Speaker 4: the kind of stuff that like really hurts. And I 581 00:29:17,240 --> 00:29:20,440 Speaker 4: think we're just totally ignoring that, and we're we're, you know, 582 00:29:20,760 --> 00:29:24,400 Speaker 4: we're looking at like I don't know chatbots and whether 583 00:29:24,440 --> 00:29:28,400 Speaker 4: they make us sad, but like the whole economy, the 584 00:29:28,440 --> 00:29:32,719 Speaker 4: whole economy is changing and orienting around this thing, and 585 00:29:32,800 --> 00:29:34,440 Speaker 4: I don't even know what the effects are, but I 586 00:29:34,480 --> 00:29:36,920 Speaker 4: know that it's gonna be like there's there's gonna be 587 00:29:36,920 --> 00:29:37,600 Speaker 4: a huge impact. 588 00:29:38,600 --> 00:29:40,480 Speaker 1: We just take a short break now, but then we'll 589 00:29:40,520 --> 00:29:45,560 Speaker 1: hear from Kyle about a universal feeling hatred of weather apps. 590 00:29:45,760 --> 00:29:49,000 Speaker 2: Stay with us, so mad Apple bought dark Sky. Still 591 00:29:49,120 --> 00:29:50,240 Speaker 2: this is what this is. 592 00:29:50,120 --> 00:30:09,720 Speaker 1: The whole thing. Welcome back to tech stuff. Over the break, 593 00:30:09,840 --> 00:30:12,480 Speaker 1: all of our panelists were giggling about Kyle's story about 594 00:30:12,480 --> 00:30:14,560 Speaker 1: how we will hate weather apps. But Kyle tell us 595 00:30:14,560 --> 00:30:16,480 Speaker 1: a story first, and then I want to hear read 596 00:30:16,640 --> 00:30:18,240 Speaker 1: and attash to your takes. 597 00:30:18,640 --> 00:30:21,200 Speaker 5: Once upon a time there was a weather app called 598 00:30:21,280 --> 00:30:24,720 Speaker 5: dark Sky. It launched in twenty ten. It was amazing. 599 00:30:25,200 --> 00:30:29,680 Speaker 5: It gave you live radar, it predicted precipitation into the future, 600 00:30:30,160 --> 00:30:31,880 Speaker 5: it did a push alert when it was going to 601 00:30:32,000 --> 00:30:35,920 Speaker 5: rain near you, and it was great. And then Apple 602 00:30:36,080 --> 00:30:39,240 Speaker 5: bought dark Sky in twenty twenty and then in twenty 603 00:30:39,280 --> 00:30:43,800 Speaker 5: twenty three, Apple shut down dark Sky after integrating some 604 00:30:43,880 --> 00:30:47,360 Speaker 5: of its features into Apple Weather, a common story in 605 00:30:47,440 --> 00:30:52,360 Speaker 5: Silicon Valley startup you know, acquisition and destruction. But the 606 00:30:52,400 --> 00:30:54,600 Speaker 5: good news is that one of the co founders of 607 00:30:54,680 --> 00:30:58,200 Speaker 5: dark Sky, this guy Adam Grossman, and another co founder 608 00:30:58,240 --> 00:31:01,600 Speaker 5: as well, have a new weather app that we can 609 00:31:01,640 --> 00:31:05,040 Speaker 5: all enjoy, which is called acne Weather. And I wrote 610 00:31:05,080 --> 00:31:08,280 Speaker 5: this piece this week kind of mourning Dark Sky and 611 00:31:08,320 --> 00:31:11,120 Speaker 5: also discussing what what does make a good weather app, 612 00:31:11,120 --> 00:31:14,440 Speaker 5: because at least for me, it's a source of constant frustration, 613 00:31:15,080 --> 00:31:18,160 Speaker 5: and particularly so many people around me have complained about 614 00:31:18,200 --> 00:31:21,800 Speaker 5: Apple Weather lately that I suspected something had just gotten 615 00:31:21,840 --> 00:31:25,840 Speaker 5: worse like and shittification had struck the weather app genre. 616 00:31:26,720 --> 00:31:29,680 Speaker 5: So this was my investigation, and I have to say 617 00:31:29,760 --> 00:31:34,400 Speaker 5: Acme Weather is a huge improvement on Apple Yeah, I 618 00:31:34,600 --> 00:31:37,560 Speaker 5: recommend it, recommend it, I would say, so far worse 619 00:31:37,600 --> 00:31:38,440 Speaker 5: than twenty five. 620 00:31:38,280 --> 00:31:40,760 Speaker 1: Dollars a year, which is deep redial smiling. 621 00:31:41,240 --> 00:31:42,760 Speaker 2: I'm I'm I'm a customer. 622 00:31:43,440 --> 00:31:44,040 Speaker 1: I will buy it. 623 00:31:44,400 --> 00:31:46,960 Speaker 4: I mean I am now, like as of right now, 624 00:31:47,000 --> 00:31:49,080 Speaker 4: I will as soon as this is over. I'm paying 625 00:31:49,080 --> 00:31:52,040 Speaker 4: the twenty five dollars. Like I loved Dark Sky. I 626 00:31:52,120 --> 00:31:55,440 Speaker 4: was so so sad when Apple bought Dark Sky and 627 00:31:56,040 --> 00:31:59,560 Speaker 4: this makes me very happy. So that you that it 628 00:31:59,560 --> 00:32:01,320 Speaker 4: has your dom Kyle, what made. 629 00:32:01,200 --> 00:32:02,600 Speaker 1: You choose to write this story this week? 630 00:32:03,160 --> 00:32:06,960 Speaker 5: Well, the spring weather has been completely bizarre, especially in DC. 631 00:32:07,320 --> 00:32:10,120 Speaker 5: We've gone from like eighty five degrees in sunny to 632 00:32:10,480 --> 00:32:13,800 Speaker 5: hailstorms within the space of twenty four hours, and so 633 00:32:13,920 --> 00:32:16,480 Speaker 5: my weather apps were really failing. And I have this 634 00:32:16,600 --> 00:32:19,719 Speaker 5: nostalgia for a dark Sky, and so I wanted to 635 00:32:19,800 --> 00:32:22,920 Speaker 5: test out ACME Weather and I ended up really appreciating 636 00:32:23,440 --> 00:32:26,320 Speaker 5: the graphic design of it. It's very clear. It uses 637 00:32:26,320 --> 00:32:27,200 Speaker 5: a lot of text. 638 00:32:28,040 --> 00:32:30,720 Speaker 1: They show different different weather models, right, so you know, 639 00:32:31,120 --> 00:32:33,320 Speaker 1: I could choose your adventure of the weather rather than 640 00:32:33,360 --> 00:32:36,720 Speaker 1: a top down yes ex cathedral pronouncement on what the 641 00:32:36,760 --> 00:32:38,120 Speaker 1: weather will be exactly. 642 00:32:38,200 --> 00:32:41,080 Speaker 5: So what I my conclusion is that most weather apps 643 00:32:41,120 --> 00:32:45,280 Speaker 5: are too confident, like they're too short in their predictions 644 00:32:45,320 --> 00:32:48,080 Speaker 5: and their numbers and their icons, and what I liked 645 00:32:48,080 --> 00:32:51,320 Speaker 5: about acme is that it kind of admits to this uncertainty, 646 00:32:51,760 --> 00:32:56,120 Speaker 5: and it includes visualizations of different predictions and different weather 647 00:32:56,200 --> 00:32:59,160 Speaker 5: models right on that home screen, so you can kind 648 00:32:59,200 --> 00:33:02,320 Speaker 5: of tell if a prediction is not that certain, Like 649 00:33:02,560 --> 00:33:04,280 Speaker 5: it's not just going to tell you what's going to happen, 650 00:33:04,400 --> 00:33:06,920 Speaker 5: it'll give you this indication that I don't know, maybe 651 00:33:07,000 --> 00:33:10,440 Speaker 5: the afternoon's a little shaky, like don't totally trust me, 652 00:33:10,920 --> 00:33:13,080 Speaker 5: and it just I don't know. It made me more 653 00:33:13,080 --> 00:33:14,280 Speaker 5: reassured as a user. 654 00:33:14,760 --> 00:33:16,680 Speaker 1: This is there a betting function built into the app. 655 00:33:16,720 --> 00:33:19,080 Speaker 5: Oh my god, betting on the weather. Don't say that 656 00:33:19,120 --> 00:33:22,840 Speaker 5: too loudly. This is like, no, no, And the nice 657 00:33:22,840 --> 00:33:26,720 Speaker 5: thing about acme, as I came to understand through this work, 658 00:33:26,960 --> 00:33:29,800 Speaker 5: most weather apps just use the same data, like they're 659 00:33:29,800 --> 00:33:34,920 Speaker 5: all just accessing this government data and weather company predictions 660 00:33:35,360 --> 00:33:39,160 Speaker 5: and so they're reskinning the same stuff. But acme actually 661 00:33:39,160 --> 00:33:43,200 Speaker 5: has its own machine learning based on geography. It's kind 662 00:33:43,200 --> 00:33:47,040 Speaker 5: of making its own predictions and customizing them. So I 663 00:33:47,040 --> 00:33:50,800 Speaker 5: think it actually is a technical improvement on other apps. 664 00:33:51,880 --> 00:33:53,400 Speaker 4: Do you think by the way, I mean, Noah was 665 00:33:53,920 --> 00:33:57,440 Speaker 4: cut pretty badly during the whole Doge thing. Do you 666 00:33:57,480 --> 00:34:00,680 Speaker 4: think that's actually making weather prediction worse. 667 00:34:00,600 --> 00:34:08,760 Speaker 5: Or national something, Yeah, national, what isic uphe administration? 668 00:34:08,800 --> 00:34:11,600 Speaker 4: Basically the weather keep the weather, the weather guys. That's 669 00:34:11,600 --> 00:34:13,080 Speaker 4: like where they that's where we got all. 670 00:34:12,960 --> 00:34:13,520 Speaker 2: The data from. 671 00:34:13,719 --> 00:34:13,919 Speaker 1: Yeah. 672 00:34:13,960 --> 00:34:17,879 Speaker 5: So in reporting this, I spoke to weather app founders 673 00:34:17,920 --> 00:34:21,279 Speaker 5: to see what their problems were, which is a remarkably 674 00:34:21,360 --> 00:34:24,240 Speaker 5: large set of people, and they all were talking about 675 00:34:24,520 --> 00:34:28,279 Speaker 5: n O A A being defunded and also just the 676 00:34:28,320 --> 00:34:33,319 Speaker 5: people's perception, like as Trump has been degrading science and 677 00:34:33,360 --> 00:34:37,759 Speaker 5: talking about defunding these institutions, the weather app founders were 678 00:34:37,800 --> 00:34:41,600 Speaker 5: saying that their users trust data less. There's like this 679 00:34:41,719 --> 00:34:44,920 Speaker 5: aura of incorrectness about it. They're more likely to be 680 00:34:45,040 --> 00:34:48,680 Speaker 5: mad at an incorrect weather prediction. So I think it 681 00:34:48,800 --> 00:34:50,719 Speaker 5: is like a governmental thing as well. 682 00:34:51,640 --> 00:34:53,879 Speaker 1: Natasha, I'm curious for your for your take here, because 683 00:34:53,880 --> 00:34:57,160 Speaker 1: it's this weird thing happening right where like we there's 684 00:34:57,200 --> 00:35:01,080 Speaker 1: there's a there's a new boom evidently in weather prediction apps. 685 00:35:01,680 --> 00:35:06,000 Speaker 1: There are you know, prediction markets going crazy with Calshian polymarket. 686 00:35:07,400 --> 00:35:10,080 Speaker 1: It's no longer possible as of yesterday to get the 687 00:35:10,480 --> 00:35:14,239 Speaker 1: live wait times from JFK because they're too long. So 688 00:35:14,280 --> 00:35:17,000 Speaker 1: you're not it's now totally unpredictable when you go to 689 00:35:17,040 --> 00:35:18,640 Speaker 1: the airport whether or not you will be able to 690 00:35:18,680 --> 00:35:20,960 Speaker 1: get on your flight. Like do you have any any 691 00:35:21,040 --> 00:35:24,560 Speaker 1: kind of wider thoughts about what's going on in the world. 692 00:35:24,560 --> 00:35:27,120 Speaker 1: And we also had like technology aura and whoops, and 693 00:35:27,160 --> 00:35:33,640 Speaker 1: there's this kind of like intersection of monitoring, predicting, prescribing 694 00:35:33,800 --> 00:35:36,239 Speaker 1: with health apps and stuff like what's your what's your 695 00:35:36,239 --> 00:35:37,200 Speaker 1: take on Kyle's story? 696 00:35:37,719 --> 00:35:42,480 Speaker 3: Like monitoring the situation. I mean, just from this conversation, 697 00:35:42,600 --> 00:35:48,080 Speaker 3: I'm wondering, like how gendered is interest in monitoring the situation? 698 00:35:48,320 --> 00:35:53,040 Speaker 3: Like I personally do not I guess have extremely strong feelings, 699 00:35:53,120 --> 00:35:55,799 Speaker 3: but I just like don't understand how you know, there's 700 00:35:55,800 --> 00:35:58,200 Speaker 3: so many micro climates in the Bay Area, So like 701 00:35:58,760 --> 00:36:02,080 Speaker 3: for me looking up Oaklynn versus like elsa read over, 702 00:36:02,280 --> 00:36:04,400 Speaker 3: it's just not helpful And I think I'm just not 703 00:36:04,480 --> 00:36:07,560 Speaker 3: a I would just like to know, like. 704 00:36:08,080 --> 00:36:09,600 Speaker 2: In California is what you're saying. 705 00:36:10,200 --> 00:36:14,319 Speaker 3: Yeah, I live in California, but yeah, I mean I 706 00:36:14,360 --> 00:36:16,680 Speaker 3: did think the Noah stuff was extremely sad, and to 707 00:36:16,719 --> 00:36:20,040 Speaker 3: me it seemed like like this is an amazing case 708 00:36:20,200 --> 00:36:23,680 Speaker 3: for machine learning, right, like for the technology that we have. 709 00:36:24,120 --> 00:36:28,000 Speaker 3: I don't know why you wouldn't apply like the powerful 710 00:36:28,320 --> 00:36:32,200 Speaker 3: prediction algorithms that we have to like deploy more sensors 711 00:36:32,239 --> 00:36:36,120 Speaker 3: and give people like even more robust data about what 712 00:36:36,200 --> 00:36:41,040 Speaker 3: to expect in their micro climates. So yeah, it just 713 00:36:41,280 --> 00:36:43,480 Speaker 3: it just felt like, why are we always trying to 714 00:36:43,560 --> 00:36:48,000 Speaker 3: use machine learning to like ape humans when there's so 715 00:36:48,080 --> 00:36:51,440 Speaker 3: many cases where machine learning could help us and I, 716 00:36:51,480 --> 00:36:53,520 Speaker 3: you know, maybe it could help me, like just figure 717 00:36:53,560 --> 00:36:55,839 Speaker 3: out what I'm supposed to wear on a weekend when 718 00:36:55,880 --> 00:36:58,239 Speaker 3: I'm going to three different cities in the East Bay. 719 00:36:59,040 --> 00:37:02,600 Speaker 1: Oh no, we lost, We lost Natasha. Hopefully she'll be back. 720 00:37:02,640 --> 00:37:04,760 Speaker 1: But in the meantime, Kyle, do you do you agree 721 00:37:04,760 --> 00:37:07,879 Speaker 1: with Natasha? That's something inherently bro about the situation room. 722 00:37:08,560 --> 00:37:11,560 Speaker 5: Well, you know, come to think of it, I only 723 00:37:11,640 --> 00:37:13,440 Speaker 5: talk to male weather app founder. 724 00:37:13,560 --> 00:37:14,319 Speaker 2: Is that true now? 725 00:37:14,360 --> 00:37:19,480 Speaker 4: That the one that then to be honest here in 726 00:37:19,520 --> 00:37:21,680 Speaker 4: the situation room here, Well. 727 00:37:21,920 --> 00:37:25,000 Speaker 5: You know, it speaks to DC that like the Polymarket 728 00:37:25,120 --> 00:37:27,600 Speaker 5: prediction bar was like the biggest news in town. 729 00:37:28,600 --> 00:37:30,880 Speaker 1: Talk about that. Tell that story as a kick for 730 00:37:30,920 --> 00:37:31,640 Speaker 1: this man. 731 00:37:31,840 --> 00:37:35,720 Speaker 5: I mean, so Polymarket, which is obviously the prediction market 732 00:37:35,840 --> 00:37:38,279 Speaker 5: that is trying to market itself in the US the 733 00:37:38,320 --> 00:37:42,880 Speaker 5: most right now. UH tried to open this bar in 734 00:37:43,040 --> 00:37:46,600 Speaker 5: DC to participate in this meme of monitoring the situation, 735 00:37:46,920 --> 00:37:50,120 Speaker 5: which is the admittedly brilliant behavior of trying to take 736 00:37:50,160 --> 00:37:53,120 Speaker 5: in so much data, take in so many predictions and 737 00:37:53,160 --> 00:37:56,960 Speaker 5: tweets and live streams and whatever stock market tickers that 738 00:37:57,040 --> 00:38:00,000 Speaker 5: you just understand what's going on. But in a beautiful 739 00:38:00,239 --> 00:38:04,719 Speaker 5: metaphor for our internet environments, the eighty televisions that Polymarket 740 00:38:04,719 --> 00:38:08,560 Speaker 5: had in this bar did not work. So it was 741 00:38:08,600 --> 00:38:12,400 Speaker 5: a broken space in which to monitor the broken situation, 742 00:38:12,840 --> 00:38:14,840 Speaker 5: and it turned out to be a pretty bad party. 743 00:38:15,040 --> 00:38:17,080 Speaker 2: I would say, that's very funny. 744 00:38:17,600 --> 00:38:19,560 Speaker 1: Well, that's all we have time for this week. I 745 00:38:19,680 --> 00:38:21,200 Speaker 1: like to ask who had the best week in tech 746 00:38:21,200 --> 00:38:23,040 Speaker 1: and who had the worst week in tech. I think 747 00:38:23,080 --> 00:38:25,360 Speaker 1: Natasha literally had the worst week in tech because she 748 00:38:25,440 --> 00:38:28,360 Speaker 1: dropped out of the riverside her computer stopped working. So sorry, Natasha, 749 00:38:28,360 --> 00:38:30,719 Speaker 1: but we loved having you on this week when sad 750 00:38:30,800 --> 00:38:33,080 Speaker 1: not to have your your take on the highs and lows? 751 00:38:33,600 --> 00:38:34,960 Speaker 1: But read who do you think had the best week? 752 00:38:35,000 --> 00:38:35,800 Speaker 1: About the worst week? 753 00:38:36,520 --> 00:38:38,239 Speaker 4: I don't know, I mean probably, I mean I think 754 00:38:38,280 --> 00:38:41,360 Speaker 4: you have to you got to say, like meta, really meta, 755 00:38:41,440 --> 00:38:43,960 Speaker 4: I mean Google, to a lesser extent, had had the 756 00:38:44,000 --> 00:38:44,680 Speaker 4: worst week. 757 00:38:44,920 --> 00:38:46,400 Speaker 1: Yeah, I would think that, but you look at the 758 00:38:46,400 --> 00:38:48,160 Speaker 1: stock price, it didn't I mean, to Nasha's point out, 759 00:38:48,200 --> 00:38:50,839 Speaker 1: it didn't move, ifact was slightly after trial. It's down 760 00:38:50,880 --> 00:38:53,359 Speaker 1: two percent in the last seven days. Like I mean, 761 00:38:53,520 --> 00:38:56,560 Speaker 1: I thought that was they lose two bell Weather cases 762 00:38:56,600 --> 00:38:58,680 Speaker 1: and the stock market doesn't react. I mean, maybe that 763 00:38:58,719 --> 00:39:02,240 Speaker 1: makes the best week for them to immune to the 764 00:39:02,360 --> 00:39:03,280 Speaker 1: judicial system. 765 00:39:03,640 --> 00:39:06,400 Speaker 4: I think stock price is always is always like the 766 00:39:06,440 --> 00:39:11,200 Speaker 4: wrong metric to look at, like how to measure that means? Yeah, 767 00:39:11,320 --> 00:39:15,520 Speaker 4: for individual tech companies, I mean I think it's no. 768 00:39:15,600 --> 00:39:18,400 Speaker 4: I mean, like there, their stock price didn't really go 769 00:39:18,440 --> 00:39:21,120 Speaker 4: down even after like Cambridge Analytica and like they you know, 770 00:39:21,560 --> 00:39:24,000 Speaker 4: I don't think they. I think they realize, like they're 771 00:39:24,120 --> 00:39:26,840 Speaker 4: they're making money, like it's fine, but like they're because 772 00:39:27,160 --> 00:39:29,759 Speaker 4: it just means like they're less and less relevant to me, 773 00:39:30,040 --> 00:39:33,319 Speaker 4: Like it's there's the world is changing even outside of 774 00:39:33,360 --> 00:39:36,440 Speaker 4: this lawsuit. Like I think the lawsuit is like is 775 00:39:36,480 --> 00:39:38,640 Speaker 4: not even really the biggest issue. It's just like the 776 00:39:38,680 --> 00:39:42,080 Speaker 4: whole world of social media is is you know, it's 777 00:39:42,120 --> 00:39:45,920 Speaker 4: completely changing, and like I personally, no one I know 778 00:39:46,239 --> 00:39:49,040 Speaker 4: is like on Facebook or really maybe Instagram, but like 779 00:39:49,640 --> 00:39:52,560 Speaker 4: it's not like such an important part of people's lives anymore. 780 00:39:53,160 --> 00:39:55,160 Speaker 4: And that doesn't mean it's a bad business. Like it's 781 00:39:55,200 --> 00:39:57,000 Speaker 4: gonna be you know, they're gonna be fine. They're gonna 782 00:39:57,000 --> 00:40:00,800 Speaker 4: still make a bunch of money. And you know, I 783 00:40:00,800 --> 00:40:02,520 Speaker 4: don't know, and I don't know who had the best week, 784 00:40:02,560 --> 00:40:03,400 Speaker 4: who had a good week? 785 00:40:03,640 --> 00:40:06,240 Speaker 2: Who's Kyle? Could we have multiple choice? 786 00:40:07,719 --> 00:40:11,160 Speaker 5: I kind of feel like Anthrophic continues to have a 787 00:40:11,200 --> 00:40:14,440 Speaker 5: great week, and I just feel like Claude has become 788 00:40:14,680 --> 00:40:17,879 Speaker 5: a kind of ask Jeeves of twenty twenty six, and 789 00:40:17,920 --> 00:40:20,600 Speaker 5: it's like Claude is like Gray Scott. 790 00:40:20,800 --> 00:40:23,040 Speaker 1: I asked Jeeves, but Jeeves goes and does it for 791 00:40:23,200 --> 00:40:25,640 Speaker 1: you rather than just giving an that could be better. 792 00:40:25,800 --> 00:40:29,120 Speaker 4: I saw someone back to the point, but like someone 793 00:40:29,320 --> 00:40:33,319 Speaker 4: some VC was posting on x about how he had 794 00:40:33,360 --> 00:40:35,560 Speaker 4: sort of run into the like when you when they're 795 00:40:35,640 --> 00:40:37,880 Speaker 4: when you're using too many tokens on Claude, like you 796 00:40:38,360 --> 00:40:41,200 Speaker 4: get like these blocks sometimes and he's like, oh, I'd 797 00:40:41,200 --> 00:40:44,480 Speaker 4: gone over and you know, tried code ax and it's great, 798 00:40:44,600 --> 00:40:47,200 Speaker 4: It's actually really great. And I was like this guy's 799 00:40:47,200 --> 00:40:50,120 Speaker 4: an investor in anthropic Like, what the hell is going on? 800 00:40:50,400 --> 00:40:53,880 Speaker 5: Like this is this is vibe coding, psychosis, this is 801 00:40:54,200 --> 00:40:56,080 Speaker 5: this is what I should write about now. But it's 802 00:40:56,120 --> 00:40:58,480 Speaker 5: like we're just getting sucked in so hard to these 803 00:40:58,480 --> 00:41:02,239 Speaker 5: things and we're anthwrop refires sing them ourselves now and 804 00:41:02,280 --> 00:41:03,799 Speaker 5: that's I don't know, it seems bad. 805 00:41:06,160 --> 00:41:08,160 Speaker 1: That's it for the Week in Tech. Thank you so 806 00:41:08,280 --> 00:41:09,160 Speaker 1: much for joining us. 807 00:41:09,200 --> 00:41:34,640 Speaker 2: Great to be here for tech Stuff. 808 00:41:34,760 --> 00:41:37,040 Speaker 1: I'm as Volosha and this episode was produced by Eliza 809 00:41:37,120 --> 00:41:40,200 Speaker 1: Dennis and Melissa Slaughter. It was executive produced by me 810 00:41:40,520 --> 00:41:44,040 Speaker 1: Julia Nutter and Kate Osborne for Kaleidoscope and Katrina Norvel 811 00:41:44,080 --> 00:41:48,239 Speaker 1: for iHeart Podcasts. The engineer is Kathleen Conti from CDM Studios. 812 00:41:48,640 --> 00:41:51,239 Speaker 1: Jack Insley makes this episode and Kyle Murdoch wrote our 813 00:41:51,320 --> 00:41:55,080 Speaker 1: theme song. A special thanks to you read Kyle and Natasha. 814 00:41:55,440 --> 00:41:58,080 Speaker 1: Please check out all of these threes extraordinary work they 815 00:41:58,160 --> 00:42:00,320 Speaker 1: put into the world. We're lucky to call them friends 816 00:42:00,320 --> 00:42:03,000 Speaker 1: of Tech Stuff, and please also do rate and review 817 00:42:03,040 --> 00:42:04,480 Speaker 1: the podcast wherever you listen.