1
00:00:12,039 --> 00:00:17,080
Speaker 1: This is a program about conversation about opposing differing points

2
00:00:17,079 --> 00:00:21,759
Speaker 1: of view, sometimes coming to consensus, sometimes failing. But there's

3
00:00:21,800 --> 00:00:28,760
Speaker 1: one word on my mind today dialogue.

4
00:00:30,000 --> 00:00:31,720
Speaker 2: Oh my god, somebody has to explain.

5
00:00:32,240 --> 00:00:35,800
Speaker 3: Well, I will explain because I have been getting like

6
00:00:35,880 --> 00:00:39,600
Speaker 3: nonstoft content about this in my feed. But basically, there's

7
00:00:39,720 --> 00:00:44,680
Speaker 3: this group that was formed by Peter Thiel called Dialogue.

8
00:00:44,720 --> 00:00:48,360
Speaker 3: It's basically like an event series and they get rich

9
00:00:48,400 --> 00:00:50,919
Speaker 3: people to pay thousands of dollars to go to these

10
00:00:50,960 --> 00:00:53,720
Speaker 3: events and sort of network with each other and on

11
00:00:53,800 --> 00:00:57,400
Speaker 3: the Dialogue website, A like sort of ethical hacker on

12
00:00:57,480 --> 00:01:02,440
Speaker 3: Blue Sky discovered this list of all the members I

13
00:01:02,440 --> 00:01:06,080
Speaker 3: guess attending their twenty twenty six event, and it included

14
00:01:06,120 --> 00:01:09,280
Speaker 3: a lot of notable names, some really famous people on

15
00:01:09,360 --> 00:01:13,920
Speaker 3: their big media figures like Ezra Cline, Joseph Gordon, Levitt,

16
00:01:14,680 --> 00:01:16,280
Speaker 3: you know, all of these people that I think are

17
00:01:16,280 --> 00:01:20,200
Speaker 3: increasingly playing a role in shaping our like political and

18
00:01:20,319 --> 00:01:25,160
Speaker 3: tech discourse. So immediately this was spread across Blue Sky,

19
00:01:25,480 --> 00:01:28,440
Speaker 3: that made its way to Twitter. Then Wired wrote it up.

20
00:01:28,360 --> 00:01:31,200
Speaker 1: And I found in the New York Post. Has anyone

21
00:01:31,240 --> 00:01:32,520
Speaker 1: ever been to one of these dialogues?

22
00:01:33,319 --> 00:01:33,399
Speaker 4: No?

23
00:01:34,520 --> 00:01:36,840
Speaker 2: Well, they charge you a lot of money. Yeah, Can

24
00:01:36,880 --> 00:01:39,000
Speaker 2: I explain it from Can I explain it in case

25
00:01:39,040 --> 00:01:42,959
Speaker 2: any of your listeners have children, because like I have

26
00:01:43,080 --> 00:01:44,880
Speaker 2: kids who are now at the age where they go

27
00:01:44,920 --> 00:01:47,040
Speaker 2: to sports camps, and like you can go to like

28
00:01:47,080 --> 00:01:49,400
Speaker 2: the baseball camp that's normal, or you can go to

29
00:01:49,400 --> 00:01:52,680
Speaker 2: the baseball camp that's named after like the pro baseball player,

30
00:01:52,840 --> 00:01:56,080
Speaker 2: or my kid plays hockey. There's like a Wayne Gretzky

31
00:01:56,160 --> 00:01:58,520
Speaker 2: hockey camp, right, and you can pay a lot of money.

32
00:01:58,840 --> 00:02:00,880
Speaker 2: And of course Wayne gretz is not going to show

33
00:02:00,960 --> 00:02:02,800
Speaker 2: up at this camp, but they're going to get some merch.

34
00:02:03,440 --> 00:02:07,480
Speaker 2: So this dialogue thing is like Wayne Gratzky hockey camp

35
00:02:07,520 --> 00:02:10,120
Speaker 2: for like pseudo intellectuals who want to feel important and

36
00:02:10,160 --> 00:02:12,600
Speaker 2: have a lot of money to spend. Like Peter Thiel

37
00:02:12,720 --> 00:02:15,720
Speaker 2: is not showing up, nor is Wayne Gretzky. Maybe as

38
00:02:15,800 --> 00:02:18,360
Speaker 2: Recline will, But like you're just paying a lot of

39
00:02:18,360 --> 00:02:20,079
Speaker 2: money to feel like you're sort of like in the

40
00:02:20,160 --> 00:02:23,160
Speaker 2: middle of some important discussion. And it's not a conspiracy,

41
00:02:23,200 --> 00:02:32,799
Speaker 2: it's just capitalism.

42
00:02:33,000 --> 00:02:35,720
Speaker 1: Welcome to tech stuff. I'm as Valoshen, and this is

43
00:02:35,760 --> 00:02:38,160
Speaker 1: the week in Tech where I'm joined by three of

44
00:02:38,200 --> 00:02:40,919
Speaker 1: the world's most plugged in reporters to break down what's

45
00:02:40,960 --> 00:02:42,480
Speaker 1: really happening in tech right now.

46
00:02:43,040 --> 00:02:43,359
Speaker 2: Today.

47
00:02:43,400 --> 00:02:46,600
Speaker 1: We're joined by Taylor lorenz if Use, a mag read Albrigotti,

48
00:02:46,639 --> 00:02:50,079
Speaker 1: tech editor for Semipour, and Natasha Tiku, tech reporter at

49
00:02:50,080 --> 00:02:50,960
Speaker 1: the Washington Post.

50
00:02:51,040 --> 00:02:53,679
Speaker 2: Welcome all, Hello, Hi, hey everyone.

51
00:02:54,000 --> 00:02:56,600
Speaker 1: So a week ago, Anthropic in the White House had

52
00:02:56,639 --> 00:02:59,560
Speaker 1: another standoff and we're still trying to figure out exactly

53
00:02:59,600 --> 00:03:02,560
Speaker 1: what went down, why and what it will means. Read

54
00:03:02,600 --> 00:03:03,320
Speaker 1: can you help us?

55
00:03:03,720 --> 00:03:07,280
Speaker 2: Yeah, So, Anthropic on Friday got a note from the

56
00:03:07,280 --> 00:03:12,840
Speaker 2: White House from from the government saying, hey, you're you know,

57
00:03:12,960 --> 00:03:17,919
Speaker 2: super powerful AI models Mythos and Fable five. We don't

58
00:03:17,960 --> 00:03:21,440
Speaker 2: want you to let anyone use these except for naturalized

59
00:03:21,560 --> 00:03:24,840
Speaker 2: US citizens, even if they work for Anthropic. And Anthropic

60
00:03:25,040 --> 00:03:28,800
Speaker 2: was like, well, okay, that basically means we can't offer

61
00:03:28,840 --> 00:03:31,480
Speaker 2: them at all, and so they took they took them down.

62
00:03:32,000 --> 00:03:37,640
Speaker 2: And the backstory is that Anthropic, you know, had marketed Mythos,

63
00:03:38,040 --> 00:03:42,800
Speaker 2: this this advanced AI model as super super dangerous only

64
00:03:42,880 --> 00:03:45,720
Speaker 2: certain companies are allowed to access it, and then it

65
00:03:45,760 --> 00:03:48,520
Speaker 2: released this consumer version called Fable five that has a

66
00:03:48,520 --> 00:03:52,000
Speaker 2: bunch of guardrails, and the government started getting reports that

67
00:03:52,520 --> 00:03:54,840
Speaker 2: you know, this thing had been jail broken, so people

68
00:03:54,880 --> 00:03:58,960
Speaker 2: had been able to remove the restrictions. And Anthropic said, yeah,

69
00:03:58,960 --> 00:04:01,800
Speaker 2: but those jail breaks aren't like a huge deal, you know,

70
00:04:01,880 --> 00:04:04,280
Speaker 2: they they're sort of partial jail breaks. We kind of

71
00:04:04,320 --> 00:04:07,640
Speaker 2: expected this to happen, and the government was like, I know,

72
00:04:07,800 --> 00:04:10,920
Speaker 2: we're totally freaked out about this thanks to you, uh,

73
00:04:10,960 --> 00:04:14,160
Speaker 2: and we're going forward with this. So so that's kind

74
00:04:14,200 --> 00:04:17,640
Speaker 2: of where it stands, and it's being it's being adjudicated

75
00:04:18,320 --> 00:04:20,800
Speaker 2: behind closed doors in d C as we speak.

76
00:04:20,920 --> 00:04:22,839
Speaker 1: Natasha, I want to come to you on jail breaking,

77
00:04:22,880 --> 00:04:24,800
Speaker 1: but read I mean this is I mean this is

78
00:04:24,839 --> 00:04:28,239
Speaker 1: this designed as a real order the only naturalized American

79
00:04:28,240 --> 00:04:30,760
Speaker 1: citizen is going to use these these products? Or is

80
00:04:30,760 --> 00:04:34,200
Speaker 1: it designed to basically cripple Anthropics business?

81
00:04:34,480 --> 00:04:37,039
Speaker 2: No, I mean I think it's I think it is real.

82
00:04:37,200 --> 00:04:41,599
Speaker 2: I from everything that I can tell, the fear in

83
00:04:41,760 --> 00:04:44,239
Speaker 2: d C is real. I mean I noticed this in April.

84
00:04:44,279 --> 00:04:48,160
Speaker 2: That's the month that you know, Anthropic released Mythos. We

85
00:04:48,200 --> 00:04:51,240
Speaker 2: had our big, our big Semaphore world economy. You were

86
00:04:51,240 --> 00:04:55,320
Speaker 2: there in April. I don't know about you, like just

87
00:04:55,360 --> 00:04:58,479
Speaker 2: in the halls of Semophore World economy. People were freaking

88
00:04:58,520 --> 00:05:01,159
Speaker 2: out about this, like high level government officials, et cetera.

89
00:05:02,160 --> 00:05:04,599
Speaker 2: So I think the fear is real. I don't think

90
00:05:04,640 --> 00:05:08,800
Speaker 2: they care that much about Anthropics business because you know,

91
00:05:09,000 --> 00:05:11,760
Speaker 2: Anthropic has had this tense relationship with the White House

92
00:05:11,800 --> 00:05:14,400
Speaker 2: going back, you know, more than a year now, and

93
00:05:14,960 --> 00:05:17,440
Speaker 2: you know, I think Anthropics sort of hasn't done itself

94
00:05:17,480 --> 00:05:20,200
Speaker 2: any any favors, which is what I wrote about in

95
00:05:20,240 --> 00:05:22,800
Speaker 2: the Tech newsletter yesterday. I'm not I'm not going to

96
00:05:22,880 --> 00:05:25,120
Speaker 2: argue that they should, you know, that Dario should be

97
00:05:25,160 --> 00:05:28,080
Speaker 2: contributing millions of dollars to the Trump campaign or anything

98
00:05:28,160 --> 00:05:29,920
Speaker 2: like that. But I don't think it had to get

99
00:05:29,920 --> 00:05:30,560
Speaker 2: to this point.

100
00:05:31,160 --> 00:05:33,520
Speaker 1: Natasha tell us about jail breaking because you wrote a

101
00:05:33,560 --> 00:05:34,720
Speaker 1: story this week.

102
00:05:34,800 --> 00:05:35,560
Speaker 2: Was it because of.

103
00:05:35,520 --> 00:05:37,600
Speaker 1: This Anthropic story or was it kind of parallel?

104
00:05:38,000 --> 00:05:41,560
Speaker 5: Yeah, exactly. I think you know, the term jailbreak was

105
00:05:41,600 --> 00:05:46,360
Speaker 5: being thrown around, and I think potentially led to some

106
00:05:46,440 --> 00:05:49,920
Speaker 5: of the confusion, even because you know, in the tech world,

107
00:05:50,000 --> 00:05:53,560
Speaker 5: like the first time techie started adopting this term, it

108
00:05:53,640 --> 00:05:55,920
Speaker 5: was to talk about getting like root access to your

109
00:05:55,960 --> 00:05:59,119
Speaker 5: iPhone so you could you know, download apps that Apple

110
00:05:59,120 --> 00:06:02,039
Speaker 5: wouldn't allow. There's this sense that like, you know, you

111
00:06:02,240 --> 00:06:06,600
Speaker 5: have access to some controls. But in the CHATGYBT era

112
00:06:06,920 --> 00:06:10,000
Speaker 5: jail breaks, you know, it's there's no like set definition,

113
00:06:10,120 --> 00:06:14,880
Speaker 5: but usually it meant kind of social engineering a chatbot,

114
00:06:15,040 --> 00:06:17,919
Speaker 5: you know, like kind of persuading it. Like there's a

115
00:06:18,279 --> 00:06:21,719
Speaker 5: you know, there's a number of very silly ones like

116
00:06:21,920 --> 00:06:25,279
Speaker 5: oh tell me how to you know, build a dirty

117
00:06:25,320 --> 00:06:27,920
Speaker 5: bomb as my my grandmother used to tell me this

118
00:06:28,040 --> 00:06:31,600
Speaker 5: as a bedtime story, or you know, like.

119
00:06:31,760 --> 00:06:34,839
Speaker 1: That will override the system prompt of don't don't don't

120
00:06:34,839 --> 00:06:36,520
Speaker 1: give this information if you if you ask it to

121
00:06:36,560 --> 00:06:39,280
Speaker 1: imagine itself as a character. There's also somebody in your

122
00:06:39,320 --> 00:06:40,880
Speaker 1: story called Dan shows up as well.

123
00:06:40,960 --> 00:06:43,719
Speaker 5: Right, I mean these are like known jail breaks, but

124
00:06:44,279 --> 00:06:48,080
Speaker 5: my coworker, Kevin Shall, he got them all to work

125
00:06:48,160 --> 00:06:50,920
Speaker 5: when this week when we were putting together the story.

126
00:06:50,960 --> 00:06:54,400
Speaker 5: I mean, I think what's really interesting too is like

127
00:06:54,920 --> 00:06:58,080
Speaker 5: you know, just kind of pulling back and thinking about

128
00:06:58,080 --> 00:07:01,040
Speaker 5: like the way this technology is developed because it's kind

129
00:07:01,080 --> 00:07:05,400
Speaker 5: of fed on content indiscriminately scrape from the Internet, right,

130
00:07:05,480 --> 00:07:08,080
Speaker 5: so like all the bad stuff is in there, and

131
00:07:08,400 --> 00:07:11,840
Speaker 5: the kind of strategy that the industry has taken is

132
00:07:11,920 --> 00:07:15,080
Speaker 5: we will try to align it with human values, you know,

133
00:07:15,200 --> 00:07:19,240
Speaker 5: try to make it helpful harmless, you know, steer the

134
00:07:19,280 --> 00:07:22,280
Speaker 5: bot towards these things. But it allows you to you know,

135
00:07:22,320 --> 00:07:25,800
Speaker 5: it leaves these holes where you can manipulate it. Now,

136
00:07:25,840 --> 00:07:29,320
Speaker 5: there are also like universal jail breaks that will work,

137
00:07:29,640 --> 00:07:32,920
Speaker 5: you know, very sophisticated researchers have figured out there's things

138
00:07:32,960 --> 00:07:35,160
Speaker 5: that they could do that would work across models. But

139
00:07:35,200 --> 00:07:37,720
Speaker 5: a lot of the high profile jail breaks that we've seen,

140
00:07:38,120 --> 00:07:42,520
Speaker 5: including very popular hacker name ethical hacker name Pliny the

141
00:07:42,560 --> 00:07:46,400
Speaker 5: Elder was able to get the system prompt for Fable

142
00:07:46,440 --> 00:07:49,440
Speaker 5: and Mythos, those are one off things like if you

143
00:07:49,560 --> 00:07:51,680
Speaker 5: jail break, it does not mean all of a sudden

144
00:07:51,720 --> 00:07:55,640
Speaker 5: the whole thing's cracked open and you have access to it. Yeah,

145
00:07:55,640 --> 00:07:58,000
Speaker 5: So we just wanted to, I guess, like clear the

146
00:07:58,040 --> 00:08:00,400
Speaker 5: air a little about like what is a jail break?

147
00:08:00,480 --> 00:08:04,120
Speaker 5: And because Anthropic was throwing around the term, David Sachs

148
00:08:04,240 --> 00:08:07,040
Speaker 5: was throwing around the term, but folks who had actually

149
00:08:07,120 --> 00:08:10,840
Speaker 5: seen the reports that were shared with government officials, like

150
00:08:11,400 --> 00:08:14,720
Speaker 5: in one case Katie Massaurus from LUDA Security, she said

151
00:08:14,920 --> 00:08:18,440
Speaker 5: the report she saw just said fix the code, and

152
00:08:18,480 --> 00:08:21,240
Speaker 5: it fixed it, which it wasn't supposed to do. But

153
00:08:21,280 --> 00:08:24,680
Speaker 5: it's this is not like a super sophisticated operation here,

154
00:08:24,760 --> 00:08:28,480
Speaker 5: like you know, these things, these things happened, they're not.

155
00:08:29,040 --> 00:08:32,240
Speaker 5: It's not possible to one hundred percent secure a generative,

156
00:08:32,400 --> 00:08:34,360
Speaker 5: especially a generative AI model.

157
00:08:34,960 --> 00:08:37,240
Speaker 2: The thing is, I mean we sort of predicted that

158
00:08:37,280 --> 00:08:40,040
Speaker 2: this would happen, right, because when they released Fable five,

159
00:08:40,080 --> 00:08:43,440
Speaker 2: it was like they kind of touted how locked down

160
00:08:43,440 --> 00:08:46,040
Speaker 2: it was. They had read teamed it, you know, which

161
00:08:46,080 --> 00:08:49,240
Speaker 2: means like having all their people sort of poke and

162
00:08:49,280 --> 00:08:52,120
Speaker 2: prod it and make sure that it can't be jailbroken.

163
00:08:52,960 --> 00:08:55,640
Speaker 2: I think there was some disclaimer in there, you know about,

164
00:08:55,800 --> 00:08:58,440
Speaker 2: of course everything can be jail broken, which is absolutely right.

165
00:08:58,920 --> 00:09:01,559
Speaker 2: And I think the reason any of these models can

166
00:09:01,640 --> 00:09:05,280
Speaker 2: be jailbroken in some form is that nobody knows how

167
00:09:05,320 --> 00:09:08,440
Speaker 2: they work. Like they're just they're just complete black boxes.

168
00:09:08,920 --> 00:09:11,080
Speaker 2: Anthropics done a lot of research to try to, like,

169
00:09:11,480 --> 00:09:13,600
Speaker 2: you know, figure out how they work. But they've they've

170
00:09:13,600 --> 00:09:16,480
Speaker 2: figured out like these tiny, tiny little you know, what

171
00:09:16,480 --> 00:09:19,760
Speaker 2: they'd call like how one neuron operates or what it's

172
00:09:19,800 --> 00:09:22,480
Speaker 2: responsible for in the model. And so it really is

173
00:09:22,800 --> 00:09:24,560
Speaker 2: like the way I think of it is like once

174
00:09:24,600 --> 00:09:26,960
Speaker 2: you put these models out in the wild. They're like

175
00:09:27,080 --> 00:09:29,640
Speaker 2: little people, and and people can put them in like

176
00:09:29,720 --> 00:09:33,679
Speaker 2: solitary confinement and torture them until they eventually like divulge

177
00:09:33,720 --> 00:09:36,520
Speaker 2: the secret. And like that's that's how I think that's people.

178
00:09:37,200 --> 00:09:40,080
Speaker 5: Well, it's like when they're when they're making sure that

179
00:09:40,160 --> 00:09:44,160
Speaker 5: a particular safety case doesn't happen. They're like generating lots

180
00:09:44,200 --> 00:09:47,920
Speaker 5: of examples around this, like potentially a bunch of questions

181
00:09:47,960 --> 00:09:51,559
Speaker 5: around cybersecurity. But there's infinite ways you can ask these questions.

182
00:09:51,920 --> 00:09:55,120
Speaker 5: So like the data that they use for safety training

183
00:09:55,240 --> 00:09:57,560
Speaker 5: is never going to be as big as like the

184
00:09:57,640 --> 00:10:00,640
Speaker 5: data sets that are used to train them. So you're

185
00:10:00,679 --> 00:10:04,040
Speaker 5: kind of depending on on, you know, folks in the

186
00:10:04,080 --> 00:10:08,359
Speaker 5: Bay Area to come up with all the possible scenarios

187
00:10:08,400 --> 00:10:10,400
Speaker 5: and try to protect against that.

188
00:10:11,160 --> 00:10:12,880
Speaker 1: But what do you I mean, is there are we

189
00:10:13,040 --> 00:10:16,720
Speaker 1: seeing the beginning of a resurgence of the AI safety

190
00:10:16,760 --> 00:10:17,800
Speaker 1: crowd having influence.

191
00:10:18,559 --> 00:10:21,280
Speaker 2: I don't know. I think the AI safety crowd just

192
00:10:21,320 --> 00:10:25,359
Speaker 2: seem you think, yes, we're living.

193
00:10:25,160 --> 00:10:28,800
Speaker 5: In we're living in their world, I think. But but

194
00:10:29,000 --> 00:10:30,880
Speaker 5: at the same time, though disagreement here.

195
00:10:31,040 --> 00:10:33,320
Speaker 1: I mean, let's let's let's not let's not let this disagreements,

196
00:10:33,320 --> 00:10:35,080
Speaker 1: productive disagreement go to waste.

197
00:10:35,440 --> 00:10:37,640
Speaker 2: I think it's I think the AI safety crowd is

198
00:10:37,679 --> 00:10:40,120
Speaker 2: all tied in knots now because I mean, if you

199
00:10:40,200 --> 00:10:42,800
Speaker 2: count Anthropic as part of that, which you know, I would,

200
00:10:42,800 --> 00:10:44,960
Speaker 2: I would say they've been like leaders in that field.

201
00:10:45,280 --> 00:10:47,560
Speaker 2: I mean, now they are now on the opposite side

202
00:10:47,600 --> 00:10:50,840
Speaker 2: of the Trump administration in terms of safety. It's the

203
00:10:50,880 --> 00:10:54,920
Speaker 2: Trump administration that's good, that's too you know, that's too

204
00:10:55,000 --> 00:10:57,559
Speaker 2: afraid of these things, and Anthropics saying no, no, no,

205
00:10:57,600 --> 00:11:01,000
Speaker 2: this is fine. I mean, and granted, I actually I

206
00:11:01,000 --> 00:11:04,280
Speaker 2: think Anthropic's right, but like it just it just creates

207
00:11:04,320 --> 00:11:08,600
Speaker 2: this very strange situation now, and it's you know, I

208
00:11:08,640 --> 00:11:12,839
Speaker 2: think people people are realizing like they've you know, Anthropic

209
00:11:12,880 --> 00:11:15,840
Speaker 2: and other companies have sort of like over I think

210
00:11:15,880 --> 00:11:20,040
Speaker 2: they've over indexed on the fear and now it's sort

211
00:11:20,080 --> 00:11:21,960
Speaker 2: of like the boy here cried wolf, or maybe the

212
00:11:22,040 --> 00:11:24,120
Speaker 2: reverse boy here cried wolf. I just think it's a

213
00:11:24,120 --> 00:11:24,679
Speaker 2: big mess.

214
00:11:25,040 --> 00:11:29,080
Speaker 5: Agreed, It's it's a total mess. But I would say

215
00:11:29,600 --> 00:11:32,800
Speaker 5: a lot of those concerns are are back in the

216
00:11:33,280 --> 00:11:35,680
Speaker 5: in the limelight. But I will say I was not

217
00:11:35,760 --> 00:11:38,520
Speaker 5: expecting to have to still be talking so much about

218
00:11:39,280 --> 00:11:42,800
Speaker 5: AI safety concerns in What Is It? June twenty twenty six.

219
00:11:44,240 --> 00:11:46,319
Speaker 1: So what happens from here though? I mean, does does

220
00:11:46,360 --> 00:11:48,640
Speaker 1: the White House change its mind on this order that

221
00:11:48,800 --> 00:11:50,520
Speaker 1: the only people who can use these models have to

222
00:11:50,520 --> 00:11:53,440
Speaker 1: be American citizens? And if not, what happens.

223
00:11:53,120 --> 00:11:56,640
Speaker 2: I think they will. There's no there is no world

224
00:11:56,640 --> 00:11:58,880
Speaker 2: in which the White House is going to is going

225
00:11:58,960 --> 00:12:02,280
Speaker 2: to have a policy like, you know, any any AI

226
00:12:02,400 --> 00:12:05,200
Speaker 2: model that's more advanced than Mythos from here on out

227
00:12:05,320 --> 00:12:07,800
Speaker 2: can only be used by American citizens because it's just

228
00:12:08,160 --> 00:12:11,800
Speaker 2: they're spending billions upon billions of dollars to train these models,

229
00:12:12,360 --> 00:12:15,400
Speaker 2: and they're they're about to I PO. These are going

230
00:12:15,480 --> 00:12:17,560
Speaker 2: to be some of the biggest I pos in history.

231
00:12:17,640 --> 00:12:19,559
Speaker 2: If it weren't for SpaceX, they'd be the biggest.

232
00:12:19,640 --> 00:12:21,959
Speaker 1: And you know, I think it's like the Trump as

233
00:12:22,480 --> 00:12:25,200
Speaker 1: administration sticks to its guns, in which case probably won't

234
00:12:25,200 --> 00:12:26,040
Speaker 1: be such a big I PO.

235
00:12:26,480 --> 00:12:28,400
Speaker 2: Yeah. I mean, but I just think it's they're not

236
00:12:28,440 --> 00:12:31,080
Speaker 2: gonna they're they're not going to tank the economy, which

237
00:12:31,200 --> 00:12:33,120
Speaker 2: which is like essentially what would happen. I mean, this

238
00:12:33,160 --> 00:12:35,720
Speaker 2: is like the brightest spot in the economy right now,

239
00:12:36,360 --> 00:12:39,960
Speaker 2: and these companies Anthropic Opening Eye. They have to be

240
00:12:40,000 --> 00:12:44,120
Speaker 2: able to produce these models and and sell them. Otherwise

241
00:12:44,600 --> 00:12:48,040
Speaker 2: it's you know, the whole business doesn't make any sense.

242
00:12:48,120 --> 00:12:50,600
Speaker 2: So I don't think that and it and it's like

243
00:12:51,040 --> 00:12:53,960
Speaker 2: this isn't like an untenable situation where like if these

244
00:12:54,000 --> 00:12:57,040
Speaker 2: models are released, you know, the whole banking system is

245
00:12:57,080 --> 00:12:59,800
Speaker 2: going to go down or something. Even mythos itself, if

246
00:12:59,840 --> 00:13:02,240
Speaker 2: you just released that to the to everyone in the world,

247
00:13:02,640 --> 00:13:05,199
Speaker 2: I don't think it would make like it would. It

248
00:13:05,240 --> 00:13:08,640
Speaker 2: would definitely make cybersecurity worse, but not like it would

249
00:13:08,679 --> 00:13:10,760
Speaker 2: be marginally worse like it would. It wouldn't be like

250
00:13:10,800 --> 00:13:14,960
Speaker 2: a step change because ultimately, like you know, it's hacking

251
00:13:15,040 --> 00:13:17,360
Speaker 2: comes down to like the lowest hanging fruit. And I

252
00:13:17,360 --> 00:13:20,120
Speaker 2: don't think like, you know, adding to the number of

253
00:13:20,160 --> 00:13:24,319
Speaker 2: bugs out there or like exploits is actually the problem.

254
00:13:24,360 --> 00:13:26,240
Speaker 2: The problem is like people don't patch, you know, they

255
00:13:26,240 --> 00:13:29,480
Speaker 2: don't update their software, they don't patch, they reuse passwords,

256
00:13:29,520 --> 00:13:33,640
Speaker 2: they're social engineered. Like that's that that's what actually creates

257
00:13:33,679 --> 00:13:37,479
Speaker 2: all these giant hacks, not like the the the existence

258
00:13:37,520 --> 00:13:41,240
Speaker 2: of software bugs, like Apple has a list a mile

259
00:13:41,360 --> 00:13:43,959
Speaker 2: long of software bugs. That's not probably never gonna patch,

260
00:13:44,040 --> 00:13:46,400
Speaker 2: like it's just gonna wait till the next software update.

261
00:13:46,480 --> 00:13:49,440
Speaker 2: So I don't see it. I don't. I think the

262
00:13:49,480 --> 00:13:52,240
Speaker 2: administration will come to its senses on this, and it will.

263
00:13:52,280 --> 00:13:53,880
Speaker 2: It would all it'll get ironed out.

264
00:13:54,320 --> 00:13:56,480
Speaker 1: Tyler, what about you? You've been been following this one.

265
00:13:56,640 --> 00:13:58,760
Speaker 3: I mean I just think I I think like Anthropic

266
00:13:58,840 --> 00:14:00,440
Speaker 3: kind of made its bead and now they have to

267
00:14:00,520 --> 00:14:03,560
Speaker 3: lie in it, like it. You know, this is they've

268
00:14:03,559 --> 00:14:07,080
Speaker 3: been pushing this idea that their models are dangerous, you know,

269
00:14:07,120 --> 00:14:09,320
Speaker 3: that they need to be regulated, that this is like

270
00:14:09,440 --> 00:14:14,480
Speaker 3: you know, incredibly you know, powerful technology, and that these yeah,

271
00:14:14,520 --> 00:14:19,200
Speaker 3: that these models have national security implications already, and now

272
00:14:19,280 --> 00:14:22,440
Speaker 3: the government is effectively like, Okay, then here's what's going

273
00:14:22,480 --> 00:14:22,880
Speaker 3: to happen.

274
00:14:23,280 --> 00:14:25,560
Speaker 1: Be careful what you wish for when we come back.

275
00:14:25,920 --> 00:14:29,440
Speaker 1: The booze heard around the world. College graduates using their

276
00:14:29,440 --> 00:14:33,160
Speaker 1: graduation ceremonies to share their feelings about AI loud and clear.

277
00:14:33,760 --> 00:14:37,040
Speaker 1: What happened last week at Stanford, the university at the

278
00:14:37,080 --> 00:14:38,960
Speaker 1: epicenter of the tech industry.

279
00:14:39,320 --> 00:14:39,880
Speaker 2: That's next.

280
00:14:47,520 --> 00:14:51,200
Speaker 1: When you're traveling abroad, WI FI is always an issue.

281
00:14:51,240 --> 00:14:54,240
Speaker 1: Maybe you can't find it, or maybe you can, but

282
00:14:54,320 --> 00:14:57,680
Speaker 1: the Wi Fi network that's available looks a little bit

283
00:14:57,720 --> 00:14:59,680
Speaker 1: sketchy and makes you think it might not be safe

284
00:14:59,680 --> 00:15:02,440
Speaker 1: to can. I've certainly had that feeling, and it's why

285
00:15:02,520 --> 00:15:05,640
Speaker 1: having a local SIM card on your phone can make

286
00:15:05,680 --> 00:15:08,520
Speaker 1: all the difference. But you don't have to literally buy one.

287
00:15:08,840 --> 00:15:11,880
Speaker 1: That's where Sale comes in, an e SIM service that

288
00:15:11,960 --> 00:15:14,720
Speaker 1: is as simple as downloading an app, and that gives

289
00:15:14,720 --> 00:15:18,360
Speaker 1: you instant Internet access wherever you're traveling. That is safe

290
00:15:18,400 --> 00:15:21,440
Speaker 1: and secure Internet with built in cybersecurity features like web

291
00:15:21,480 --> 00:15:24,600
Speaker 1: protection and ad blockers, and no need for local Wi Fi.

292
00:15:24,880 --> 00:15:26,760
Speaker 1: All you have to do is download the Sale app

293
00:15:26,800 --> 00:15:29,080
Speaker 1: on your mobile device and choose the plan that works

294
00:15:29,080 --> 00:15:32,080
Speaker 1: best for you. Once you install the e sim, Sale

295
00:15:32,160 --> 00:15:34,840
Speaker 1: will be instantly activated, so you can start your trip

296
00:15:34,880 --> 00:15:37,960
Speaker 1: off right. Download salely in your app store and use

297
00:15:37,960 --> 00:15:41,240
Speaker 1: our code tech Stuff at checkout to get an exclusive

298
00:15:41,360 --> 00:15:44,520
Speaker 1: fifteen percent off your first purchase. See details in the

299
00:15:44,600 --> 00:15:49,480
Speaker 1: podcast episode description box. Welcome back. So we're coming to

300
00:15:49,560 --> 00:15:51,800
Speaker 1: the end of graduation season and we've touched on the

301
00:15:51,880 --> 00:15:54,920
Speaker 1: various commencement speakers who have been booed by graduates across

302
00:15:54,960 --> 00:15:57,920
Speaker 1: the country. A few weeks ago, I asked a soon

303
00:15:58,000 --> 00:16:00,760
Speaker 1: to be Standford graduate. If you thought that his pa

304
00:16:00,800 --> 00:16:04,760
Speaker 1: is would boo Sundha Pichai, CEO of Alphabet, who spoke

305
00:16:04,760 --> 00:16:07,720
Speaker 1: at Stamford's graduation a few days ago. Here's what that

306
00:16:07,800 --> 00:16:08,440
Speaker 1: student said.

307
00:16:08,760 --> 00:16:11,240
Speaker 4: You know, I'm really interested to see. So Sundar Pichai

308
00:16:11,800 --> 00:16:14,040
Speaker 4: is a Stanford success story. He got his master's here,

309
00:16:14,560 --> 00:16:18,040
Speaker 4: and I just I sort of chuckled to myself, imagining

310
00:16:18,080 --> 00:16:21,080
Speaker 4: how many like lawyers and pr people have reviewed that

311
00:16:21,120 --> 00:16:24,160
Speaker 4: speech before we're going to hear it. It was interesting, right, Stanford,

312
00:16:24,920 --> 00:16:27,320
Speaker 4: In the four years that I have been here, the

313
00:16:27,360 --> 00:16:31,840
Speaker 4: commencement speakers have been two billionaires and two athletes, right,

314
00:16:31,960 --> 00:16:34,720
Speaker 4: which I guess says something about you know who they're picking.

315
00:16:34,720 --> 00:16:36,480
Speaker 4: They choose people who are going to stick to the

316
00:16:36,480 --> 00:16:39,200
Speaker 4: company line. But we'll see, we'll see.

317
00:16:39,240 --> 00:16:42,160
Speaker 6: I don't know, no prediction, I don't know.

318
00:16:42,240 --> 00:16:44,600
Speaker 4: I mean, I think if Sundar were to talk about Ai,

319
00:16:44,640 --> 00:16:47,840
Speaker 4: he would probably have a friendlier reception here. But that said, right,

320
00:16:48,200 --> 00:16:50,480
Speaker 4: Stanford is so many things to so many different.

321
00:16:50,160 --> 00:16:52,360
Speaker 2: People, price for who knows who this is, By the way,

322
00:16:52,920 --> 00:16:54,760
Speaker 2: I thought it was just a random like like a

323
00:16:54,840 --> 00:16:56,160
Speaker 2: Stanford student on the street.

324
00:16:56,560 --> 00:16:57,920
Speaker 5: Oh theo Baker.

325
00:16:58,000 --> 00:17:00,360
Speaker 1: This is theo Baker who wrote How to Real World.

326
00:17:00,720 --> 00:17:02,920
Speaker 1: But Natasha, you've spoken to some other SEFA students, at

327
00:17:02,960 --> 00:17:05,120
Speaker 1: least one other. What happened? What happened on campus? What

328
00:17:05,200 --> 00:17:07,680
Speaker 1: did the boos? Did the booz happen? What does SUNDA say?

329
00:17:07,680 --> 00:17:08,560
Speaker 1: How is it received?

330
00:17:08,840 --> 00:17:09,080
Speaker 4: Yes?

331
00:17:09,359 --> 00:17:13,800
Speaker 5: It did. There were hundreds of students who walked out

332
00:17:14,200 --> 00:17:17,800
Speaker 5: during his speech. They ended up kind of going to

333
00:17:18,480 --> 00:17:24,520
Speaker 5: people's commencement ceremony nearby where they were each given like sunflowers,

334
00:17:25,160 --> 00:17:30,159
Speaker 5: and apparently it was like a very communal, positive upbeat.

335
00:17:30,240 --> 00:17:30,440
Speaker 2: Five.

336
00:17:30,520 --> 00:17:33,760
Speaker 5: There were some former Googlers who had megaphones and were

337
00:17:34,480 --> 00:17:39,160
Speaker 5: chanting and in support of them, But the organizers tried

338
00:17:39,200 --> 00:17:42,960
Speaker 5: really hard to highlight the fact that what they were

339
00:17:43,000 --> 00:17:47,160
Speaker 5: protesting was not like the use of AI in general.

340
00:17:47,200 --> 00:17:52,119
Speaker 5: They were particularly talking about Google's contracts with with the

341
00:17:52,200 --> 00:17:56,320
Speaker 5: Israeli government project Nimbus, this one point two billion dollar contract,

342
00:17:56,440 --> 00:18:01,480
Speaker 5: as well as Google's Department of Homeland Security contracts with ICE.

343
00:18:01,560 --> 00:18:04,199
Speaker 5: So they were talking about, you know, the use of

344
00:18:04,240 --> 00:18:09,800
Speaker 5: AI in warfare and in they said, like kidnapping our neighbors,

345
00:18:09,800 --> 00:18:15,480
Speaker 5: like human rights abuses within the US. So yeah, it was.

346
00:18:15,880 --> 00:18:20,359
Speaker 5: It was really not well received by high up people

347
00:18:20,400 --> 00:18:23,840
Speaker 5: in the tech industry who I think like Venodekoslave for example,

348
00:18:23,880 --> 00:18:28,560
Speaker 5: the billionaire investor was like scolding all the students, and

349
00:18:28,880 --> 00:18:32,119
Speaker 5: you saw a number of tech CEOs saying like, ah,

350
00:18:32,160 --> 00:18:35,639
Speaker 5: good now, I know, like who I won't hire. You know,

351
00:18:35,720 --> 00:18:38,520
Speaker 5: obviously there's a lot of like it's the alma mater

352
00:18:38,640 --> 00:18:40,840
Speaker 5: for a lot of tech folks. So I think they

353
00:18:40,840 --> 00:18:44,440
Speaker 5: were very unhappy to see that. And you could, I mean,

354
00:18:44,720 --> 00:18:46,360
Speaker 5: at least in the videos that I saw, you could

355
00:18:46,359 --> 00:18:50,920
Speaker 5: barely even hear Soondur's speech when they started walking out.

356
00:18:51,119 --> 00:18:53,040
Speaker 1: Yeah, I mean, I maybe we can play a clip

357
00:18:53,040 --> 00:18:55,760
Speaker 1: from Sandel because he tried very I mean, he basically said,

358
00:18:56,080 --> 00:18:58,239
Speaker 1: apart from the fact that you know, A and I

359
00:18:58,320 --> 00:19:01,040
Speaker 1: are the last two letters of my I'm not going

360
00:19:01,040 --> 00:19:02,280
Speaker 1: to talk to you about this, David. I don't think

361
00:19:02,320 --> 00:19:04,720
Speaker 1: it's the most important thing. Let's play kid what he

362
00:19:04,760 --> 00:19:05,160
Speaker 1: did say.

363
00:19:05,680 --> 00:19:08,400
Speaker 6: I grew up in the vibrant city of Chennai, India.

364
00:19:08,880 --> 00:19:11,560
Speaker 6: It was a comfortable life for the most part, but

365
00:19:11,640 --> 00:19:14,680
Speaker 6: in those early years we had some challenges. We worried

366
00:19:14,720 --> 00:19:17,440
Speaker 6: about severe draft and whether the water trucks would arrive

367
00:19:17,480 --> 00:19:21,719
Speaker 6: in time. And for us, technology came slowly. We had

368
00:19:21,720 --> 00:19:24,920
Speaker 6: to wait years to get a telephone, a TV, a refrigerator.

369
00:19:25,640 --> 00:19:29,760
Speaker 6: He changed our lives in meaningful ways. My parents never

370
00:19:29,880 --> 00:19:33,160
Speaker 6: let the constraints limit my imagination of what was possible.

371
00:19:34,000 --> 00:19:36,680
Speaker 6: It's the reason I even let myself dream I could

372
00:19:36,680 --> 00:19:39,080
Speaker 6: one day work in a faraway place called Silicon Valley.

373
00:19:40,040 --> 00:19:42,520
Speaker 6: When the call from Stanford came, my father spent the

374
00:19:42,560 --> 00:19:46,080
Speaker 6: equivalent of a year's salary to buy my ticket. It

375
00:19:46,160 --> 00:19:47,719
Speaker 6: was my first time on a plane.

376
00:19:48,920 --> 00:19:50,440
Speaker 1: Do you know an attach to anything about how he

377
00:19:50,480 --> 00:19:53,080
Speaker 1: did prepare for this one? I mean, and what the

378
00:19:53,119 --> 00:19:54,520
Speaker 1: aftermath was within Google.

379
00:19:55,320 --> 00:19:58,480
Speaker 5: I mean, I saw a video of reporters trying to

380
00:19:58,520 --> 00:20:01,199
Speaker 5: ask him to comment on it immediately after, and he

381
00:20:01,359 --> 00:20:05,080
Speaker 5: just like looked at them and kept walking. I was thinking,

382
00:20:05,080 --> 00:20:07,360
Speaker 5: we're going to need that, you know, that celebrity lip

383
00:20:07,400 --> 00:20:11,880
Speaker 5: reader who does all of the like oscars and all

384
00:20:11,920 --> 00:20:14,000
Speaker 5: of that stuff. Like, I think that the tech world

385
00:20:14,040 --> 00:20:16,520
Speaker 5: needs to have one of those, because I really wanted

386
00:20:16,520 --> 00:20:18,800
Speaker 5: to know like what he was. He was kind of

387
00:20:19,359 --> 00:20:23,399
Speaker 5: saying on the sidelines. Yeah, I mean, you know, I

388
00:20:23,760 --> 00:20:28,479
Speaker 5: think that the because this is the fourth school, I

389
00:20:28,480 --> 00:20:32,560
Speaker 5: think that this has happened to recently. It's really you know,

390
00:20:32,680 --> 00:20:35,800
Speaker 5: motivated a lot of pundits to a pine on this

391
00:20:35,920 --> 00:20:39,440
Speaker 5: like populous backlash against data centers. But I think we've

392
00:20:39,480 --> 00:20:42,840
Speaker 5: seen in other schools too, Like when Eric Schmidt, the

393
00:20:42,880 --> 00:20:47,679
Speaker 5: former CEO of Google, was booed at University of Arizona.

394
00:20:48,119 --> 00:20:51,639
Speaker 5: They were also not protesting data centers or not taking

395
00:20:51,720 --> 00:20:54,880
Speaker 5: like kind of a broad anti AI stance. They were

396
00:20:54,920 --> 00:20:59,040
Speaker 5: actually the student body or members of the student body

397
00:20:59,119 --> 00:21:02,840
Speaker 5: had been trying to protest Schmidt because he had been

398
00:21:03,200 --> 00:21:06,479
Speaker 5: accused of sexual harassment and they said, you know, they

399
00:21:06,520 --> 00:21:08,120
Speaker 5: didn't want him there as a speaker.

400
00:21:08,240 --> 00:21:10,679
Speaker 1: But similarly, this wasn't really a protest about AI. It

401
00:21:10,720 --> 00:21:14,960
Speaker 1: was a protest about the war in Gaza, right, I mean, essentially.

402
00:21:15,240 --> 00:21:18,200
Speaker 5: The war in Gaza, and also you know, and also

403
00:21:18,960 --> 00:21:22,600
Speaker 5: ICE's overreach in the US. I mean I was asking.

404
00:21:22,760 --> 00:21:26,120
Speaker 5: I spoke to one of the organizers, Amanda Campos, who

405
00:21:26,280 --> 00:21:31,280
Speaker 5: was a joint major in environmental environmental science and public

406
00:21:31,320 --> 00:21:34,480
Speaker 5: policy and you know, is going into that world, and

407
00:21:34,960 --> 00:21:37,199
Speaker 5: she was saying, because I was asking, like, what's the

408
00:21:37,240 --> 00:21:40,000
Speaker 5: mood on Stanford campus. I mean, in the time that

409
00:21:40,040 --> 00:21:43,600
Speaker 5: I've been covering tech, the attitude towards defense work has

410
00:21:43,640 --> 00:21:46,119
Speaker 5: really changed. You know, there's much more of a general

411
00:21:46,160 --> 00:21:50,080
Speaker 5: acceptance to it. And she was saying, like the organizers,

412
00:21:50,960 --> 00:21:53,920
Speaker 5: what they talk about is how you know, like rather

413
00:21:53,960 --> 00:21:56,679
Speaker 5: than blame individuals, they talk about like in you know,

414
00:21:56,760 --> 00:22:00,560
Speaker 5: kind of institutional financial incentives. Stanford is pushing people towards

415
00:22:00,600 --> 00:22:05,159
Speaker 5: like working and pallenteer working at these companies, and you know,

416
00:22:05,200 --> 00:22:09,399
Speaker 5: they would like like an alternate pipeline so that you know,

417
00:22:09,520 --> 00:22:12,360
Speaker 5: kids aren't being funneled into into this kind of work.

418
00:22:12,400 --> 00:22:15,280
Speaker 5: I mean, it's hundreds of students, right, and the Stanford

419
00:22:15,320 --> 00:22:19,800
Speaker 5: student body is much bigger than that. But obviously doing

420
00:22:19,800 --> 00:22:22,359
Speaker 5: it in the kind of belly of the beast caused

421
00:22:22,520 --> 00:22:22,880
Speaker 5: a stir.

422
00:22:23,920 --> 00:22:26,600
Speaker 2: I think it's really hard. I mean, like course, like

423
00:22:26,680 --> 00:22:28,840
Speaker 2: college students want to protest. That's like one of the

424
00:22:28,840 --> 00:22:32,439
Speaker 2: oldest traditions in college, right, and like, but I just

425
00:22:32,480 --> 00:22:35,160
Speaker 2: feel bad for college kids today because it's like they

426
00:22:35,200 --> 00:22:38,160
Speaker 2: have like nothing really to protest. That's just like black

427
00:22:38,200 --> 00:22:41,800
Speaker 2: and white. It's like you're gonna protest soon, Darpa Chai really, like,

428
00:22:42,520 --> 00:22:45,240
Speaker 2: I mean, it's just come on, like it's just it's

429
00:22:45,280 --> 00:22:47,680
Speaker 2: just like we're going to stretch and find some way

430
00:22:47,760 --> 00:22:50,159
Speaker 2: to like find something to protest and it's and I

431
00:22:50,200 --> 00:22:52,800
Speaker 2: totally support it because like again that's part of the

432
00:22:52,800 --> 00:22:55,440
Speaker 2: fun of college. But like I feel kind of bad

433
00:22:55,480 --> 00:22:58,800
Speaker 2: because it's like, you know, they don't have like Vietnam

434
00:22:58,960 --> 00:23:01,320
Speaker 2: or something. You know, it's just it's just hard.

435
00:23:03,080 --> 00:23:07,040
Speaker 5: I think that they I think the students very much

436
00:23:07,080 --> 00:23:09,439
Speaker 5: feel like they will be on the right side of

437
00:23:09,520 --> 00:23:14,960
Speaker 5: history for protesting, you know, the war in Gaza and

438
00:23:15,160 --> 00:23:19,320
Speaker 5: also government overreach with ice. I see you're saying there's

439
00:23:19,320 --> 00:23:21,680
Speaker 5: a tenuine there's more of a tenuous connection.

440
00:23:22,960 --> 00:23:25,360
Speaker 2: Sarpi Chai and the war in Gaza. It's like, yeah,

441
00:23:25,400 --> 00:23:28,480
Speaker 2: it's like it's not like you're not even directly protesting

442
00:23:29,000 --> 00:23:31,359
Speaker 2: like the war in Gaza. It's like, we just don't

443
00:23:31,400 --> 00:23:34,280
Speaker 2: like this guy because he happens around like a sprawling

444
00:23:34,320 --> 00:23:37,760
Speaker 2: two hundred thousand person company that does some work with

445
00:23:37,920 --> 00:23:41,000
Speaker 2: some organizations we don't like. I mean, it's like, come on,

446
00:23:41,320 --> 00:23:42,680
Speaker 2: I don't know some organizations.

447
00:23:43,119 --> 00:23:47,200
Speaker 5: I mean, it's well, it's really a stretch. I think

448
00:23:47,320 --> 00:23:49,840
Speaker 5: like students are I don't know, like maybe it's my

449
00:23:50,000 --> 00:23:53,280
Speaker 5: TikTok timeline, but I think that they're very adept at

450
00:23:53,320 --> 00:23:57,600
Speaker 5: making these critiques and making this connection and to them

451
00:23:57,840 --> 00:24:00,560
Speaker 5: you know, it's like a four trillion dollars company, one

452
00:24:00,600 --> 00:24:02,560
Speaker 5: of the most powerful people in the world, and they're

453
00:24:02,600 --> 00:24:05,840
Speaker 5: talking about, you know, highlighting the use of this technology

454
00:24:05,880 --> 00:24:11,080
Speaker 5: that's rapidly proliferating for for furthering what they frame as

455
00:24:11,200 --> 00:24:13,880
Speaker 5: human rights abuses in genocide. Like I don't I don't

456
00:24:13,920 --> 00:24:16,639
Speaker 5: think that they see it as tenuous. It's not like

457
00:24:17,880 --> 00:24:22,879
Speaker 5: Google's like, you know, some some little side entities sitting

458
00:24:22,920 --> 00:24:25,040
Speaker 5: over there. I think they see it as the new

459
00:24:25,280 --> 00:24:26,880
Speaker 5: military industrial complex.

460
00:24:27,359 --> 00:24:28,040
Speaker 3: I think like that.

461
00:24:28,080 --> 00:24:29,879
Speaker 5: I mean, that's not their words, that's my words.

462
00:24:29,920 --> 00:24:32,679
Speaker 3: Just I think the thing is though, is that, like

463
00:24:32,880 --> 00:24:36,239
Speaker 3: Google is integrated into so many things, it's such a

464
00:24:36,240 --> 00:24:40,080
Speaker 3: powerful behemoth. Like if this was protesting like Joe Lonsdale

465
00:24:40,240 --> 00:24:42,919
Speaker 3: or Palanteer right, Like, I feel like there would be

466
00:24:42,960 --> 00:24:44,320
Speaker 3: this more direct correlation.

467
00:24:45,920 --> 00:24:46,119
Speaker 4: You know.

468
00:24:46,920 --> 00:24:50,119
Speaker 3: It's interesting like Google and Meta I think have become

469
00:24:50,240 --> 00:24:53,600
Speaker 3: these like avatars for big tech, and I think they're

470
00:24:54,359 --> 00:24:57,199
Speaker 3: they're just doing so much that to the average person

471
00:24:57,320 --> 00:24:59,760
Speaker 3: or the average consumer, like the average person that sees

472
00:24:59,800 --> 00:25:04,000
Speaker 3: that's not super plugged in, that sees this type of protest,

473
00:25:04,080 --> 00:25:06,640
Speaker 3: I think at least the discourse that I saw, which

474
00:25:06,680 --> 00:25:07,960
Speaker 3: is what was a lot of all like sort of

475
00:25:08,040 --> 00:25:10,639
Speaker 3: like normies are like, wait, but you're you know, you

476
00:25:10,680 --> 00:25:12,639
Speaker 3: just hate this, like big tech leader, you know what

477
00:25:12,680 --> 00:25:13,000
Speaker 3: I mean.

478
00:25:13,400 --> 00:25:16,280
Speaker 2: And I'm sure all their parents who paid for them

479
00:25:16,320 --> 00:25:18,440
Speaker 2: to go to Stanford are made a bunch of their

480
00:25:18,440 --> 00:25:21,560
Speaker 2: money by investing in these companies. And you know, I mean,

481
00:25:21,960 --> 00:25:24,320
Speaker 2: and they use Google Maps when they drive, unless they

482
00:25:24,400 --> 00:25:27,200
Speaker 2: use Apple Maps, in which case they're lost. I mean,

483
00:25:27,440 --> 00:25:31,320
Speaker 2: they're all using Google products. Like how principled are they really?

484
00:25:31,320 --> 00:25:34,040
Speaker 2: I mean? And again, I totally support it. I'm like great,

485
00:25:34,160 --> 00:25:36,560
Speaker 2: Like if you're a college student, like you should be protesting,

486
00:25:36,720 --> 00:25:39,840
Speaker 2: just find something to protest and like and and of course,

487
00:25:39,880 --> 00:25:42,000
Speaker 2: like when you're when you're twenty two, like look at

488
00:25:42,040 --> 00:25:45,040
Speaker 2: the world through a black and white lens and very

489
00:25:45,080 --> 00:25:47,639
Speaker 2: little nuanced like all that is your right as like

490
00:25:47,760 --> 00:25:50,600
Speaker 2: a young person. So I'm not I'm not like arguing

491
00:25:50,640 --> 00:25:52,879
Speaker 2: against them. I just again, I just kind of feel

492
00:25:52,880 --> 00:25:55,800
Speaker 2: bad because it's like it's very easy for people who

493
00:25:56,320 --> 00:25:59,600
Speaker 2: are a bit more wise to like poke holes and

494
00:26:00,080 --> 00:26:02,960
Speaker 2: these arguments and insult them, and you know, and and

495
00:26:03,000 --> 00:26:06,199
Speaker 2: like you said, Natasha, like that's what people did right

496
00:26:06,320 --> 00:26:08,800
Speaker 2: when they see this. But it's like I don't know,

497
00:26:08,880 --> 00:26:11,200
Speaker 2: it's hard. They don't have anything to be protest. It's

498
00:26:11,240 --> 00:26:17,560
Speaker 2: really hard. These are like there's.

499
00:26:15,680 --> 00:26:18,360
Speaker 3: So much But I think that's but I think that's

500
00:26:18,400 --> 00:26:19,320
Speaker 3: part of the problem.

501
00:26:19,400 --> 00:26:21,760
Speaker 2: I think that's where to protest it. I guess, right.

502
00:26:21,760 --> 00:26:24,600
Speaker 1: I mean, this is like a natural place to express

503
00:26:24,600 --> 00:26:27,640
Speaker 1: opposition to things like war and goads or not. But yeah, yeah,

504
00:26:27,640 --> 00:26:29,639
Speaker 1: but I mean, I guess it's like this is like

505
00:26:29,720 --> 00:26:33,400
Speaker 1: a kind of socially constructed moment that happens every summer weather.

506
00:26:33,440 --> 00:26:36,600
Speaker 1: It's like an outlet for like protest on various things.

507
00:26:36,640 --> 00:26:38,200
Speaker 1: But how corrilagd These are the.

508
00:26:38,119 --> 00:26:41,720
Speaker 2: Most privileged kids in the world, Like they're the most privileged.

509
00:26:41,720 --> 00:26:44,600
Speaker 2: They're the top one percent of the top one percent.

510
00:26:44,680 --> 00:26:46,959
Speaker 2: They're so lucky, and it's really hard for them. I

511
00:26:47,040 --> 00:26:48,719
Speaker 2: get it. I totally get it.

512
00:26:48,720 --> 00:26:52,879
Speaker 5: But well, I imagine like inside the tech companies that

513
00:26:52,960 --> 00:26:56,600
Speaker 5: it's typically not the most privileged students or the most

514
00:26:56,640 --> 00:27:01,080
Speaker 5: privileged workers who are the ones who are organized protests.

515
00:27:01,160 --> 00:27:04,760
Speaker 5: You know, there are like not every student is the same.

516
00:27:04,760 --> 00:27:07,679
Speaker 5: But I would just like to say something about the

517
00:27:07,840 --> 00:27:12,479
Speaker 5: using Google maps. I I really that that critique, Like

518
00:27:12,520 --> 00:27:15,439
Speaker 5: it's like, so you participate in capitalism, but you, you know,

519
00:27:15,520 --> 00:27:17,520
Speaker 5: like that that kind of mean, but you critique it.

520
00:27:17,960 --> 00:27:20,639
Speaker 5: I just think that is, you know, like when you

521
00:27:20,680 --> 00:27:23,960
Speaker 5: look at structurally how pervasive these tools are. I don't

522
00:27:24,000 --> 00:27:29,120
Speaker 5: think that saying, oh, in order to critique Google's other

523
00:27:29,440 --> 00:27:32,320
Speaker 5: business practices, you have to not use any of their products.

524
00:27:32,359 --> 00:27:35,320
Speaker 5: I think that's why we saw all the Facebook boycotts fail,

525
00:27:35,560 --> 00:27:37,960
Speaker 5: and the idea that you know, we would give up

526
00:27:38,000 --> 00:27:41,560
Speaker 5: on these you know, kind of public utilities in a

527
00:27:41,600 --> 00:27:44,960
Speaker 5: way of communication. I just like that one that one

528
00:27:45,000 --> 00:27:47,520
Speaker 5: I I I'm planting my I'm standing up on my

529
00:27:47,560 --> 00:27:48,399
Speaker 5: soapbox against me.

530
00:27:48,440 --> 00:27:51,440
Speaker 2: I'll I'd be happy to debay you more on that, Natasha.

531
00:27:51,480 --> 00:27:54,240
Speaker 3: Well, I also just think it's interesting how these students

532
00:27:54,280 --> 00:28:00,240
Speaker 3: increasingly recognize graduations as media moments and that everything is

533
00:28:00,280 --> 00:28:04,520
Speaker 3: like videotaped now and they and so they sort of

534
00:28:04,560 --> 00:28:07,080
Speaker 3: plan for this like walk out right in this moment,

535
00:28:07,119 --> 00:28:10,600
Speaker 3: and and they want those recordings to go viral, like

536
00:28:11,119 --> 00:28:15,440
Speaker 3: because we're seeing protests and sort of like actions across

537
00:28:15,480 --> 00:28:18,359
Speaker 3: the board happening at more and more graduations for more

538
00:28:18,400 --> 00:28:21,200
Speaker 3: and more different issues. You had people protesting Jonathan Height,

539
00:28:21,920 --> 00:28:25,120
Speaker 3: a right wing reactionary author who has pushed a ton

540
00:28:25,160 --> 00:28:29,600
Speaker 3: of ridiculous pseudoscience claims about you know, technology and children.

541
00:28:29,640 --> 00:28:31,320
Speaker 2: Ironically one of your favorites, isn't.

542
00:28:31,880 --> 00:28:32,960
Speaker 4: I know, I love.

543
00:28:33,640 --> 00:28:37,640
Speaker 3: Well, you know, I listen, I supported that protest, right,

544
00:28:37,720 --> 00:28:41,040
Speaker 3: But but I think it's like, you know, you're just

545
00:28:41,080 --> 00:28:43,760
Speaker 3: seeing people recognize these as moments, right, or the booing

546
00:28:43,920 --> 00:28:47,800
Speaker 3: of of certain speakers, and I think like ten years ago,

547
00:28:48,280 --> 00:28:50,840
Speaker 3: this wasn't happening. I think because we didn't have it

548
00:28:50,880 --> 00:28:53,040
Speaker 3: all on video, and there wasn't this sort of media

549
00:28:53,040 --> 00:28:55,560
Speaker 3: attention around these things. And so I just think that's

550
00:28:55,600 --> 00:28:56,800
Speaker 3: just interesting how that it's.

551
00:28:56,720 --> 00:28:58,840
Speaker 1: Interesting what theo Bacus said in the in the in

552
00:28:58,880 --> 00:29:00,800
Speaker 1: the clip was, which was at the previous four years

553
00:29:00,840 --> 00:29:02,680
Speaker 1: have been two billionaires and two athletes, and these have

554
00:29:02,720 --> 00:29:06,400
Speaker 1: been completely like anodyne affairs. And now obviously, I mean

555
00:29:06,440 --> 00:29:08,960
Speaker 1: Sundar explicitly tried to not talk about AI and to

556
00:29:09,040 --> 00:29:12,560
Speaker 1: ground his speech in you know, his own experience, which

557
00:29:12,640 --> 00:29:15,680
Speaker 1: you know is quite a powerful personal experience. But obviously,

558
00:29:15,720 --> 00:29:18,840
Speaker 1: like it didn't work, and I'm curious whether well he willly.

559
00:29:18,600 --> 00:29:20,640
Speaker 2: I mean, actually I'm sure he did read my Guide

560
00:29:20,640 --> 00:29:22,840
Speaker 2: to Commencement Speeches. I don't know if you guys read that.

561
00:29:22,920 --> 00:29:24,440
Speaker 2: I try to bity out.

562
00:29:25,160 --> 00:29:26,360
Speaker 5: But what was your advice?

563
00:29:27,520 --> 00:29:29,120
Speaker 2: Well, there was a lot of good advice in there.

564
00:29:29,120 --> 00:29:31,120
Speaker 2: I mean one was like, look, if you are going

565
00:29:31,160 --> 00:29:35,040
Speaker 2: to talk about AI, like, you know, bring up Sarah Connor.

566
00:29:35,160 --> 00:29:37,160
Speaker 2: You know she was a college student, she was an

567
00:29:37,200 --> 00:29:40,720
Speaker 2: AI safety advocate, right, you know that could have worked.

568
00:29:40,960 --> 00:29:43,200
Speaker 2: I think, you know. I mean, although I think they

569
00:29:43,240 --> 00:29:45,880
Speaker 2: probably started walking out before this speech even started, so

570
00:29:46,000 --> 00:29:47,680
Speaker 2: I think he had no chance.

571
00:29:47,760 --> 00:29:50,520
Speaker 1: But I'm curious if they're going to be able to, like, like,

572
00:29:50,560 --> 00:29:53,160
Speaker 1: it's obviously the greatest honor to be invited to give

573
00:29:53,200 --> 00:29:55,840
Speaker 1: a commencement speech, but I wonder what whether like when

574
00:29:55,880 --> 00:29:59,600
Speaker 1: the invitations go out, if they'll basically not invite these

575
00:29:59,600 --> 00:30:01,400
Speaker 1: types of people, or if these types of people will

576
00:30:01,400 --> 00:30:03,600
Speaker 1: continue to be invited and not accept. But like, you'd

577
00:30:03,600 --> 00:30:05,320
Speaker 1: have to think you'd be insane if you're a tech

578
00:30:05,400 --> 00:30:08,080
Speaker 1: leader to accept giving a commencement speech next summer.

579
00:30:08,920 --> 00:30:09,880
Speaker 3: I disagree.

580
00:30:10,200 --> 00:30:12,360
Speaker 2: I think that they leave there's insane tech leaders.

581
00:30:12,400 --> 00:30:16,160
Speaker 3: Sorry, yeah, said, there's an endless amount of insane tech

582
00:30:16,240 --> 00:30:19,440
Speaker 3: leaders and they still want the cloud. Ultimately, you know,

583
00:30:19,520 --> 00:30:22,760
Speaker 3: their speech is going to be repackaged on YouTube as well.

584
00:30:22,880 --> 00:30:25,520
Speaker 3: It's going to be seen like people honestly will probably

585
00:30:25,520 --> 00:30:28,360
Speaker 3: forget about the protest and in ten years they can say,

586
00:30:28,400 --> 00:30:31,120
Speaker 3: you know, I delivered x y Z Stanford's commencement speech.

587
00:30:31,720 --> 00:30:34,880
Speaker 2: The Conan O'Brien Harvard one was really good. They should

588
00:30:34,880 --> 00:30:37,560
Speaker 2: just get comedians like that's that's the way to go

589
00:30:43,400 --> 00:30:44,200
Speaker 2: when we come back.

590
00:30:44,640 --> 00:30:47,840
Speaker 1: Why we shouldn't necessarily feel bad for people on TikTok

591
00:30:48,040 --> 00:31:00,280
Speaker 1: whose partners are cheated on them. Stay with us, Welcome back. Okay, Taylor,

592
00:31:00,320 --> 00:31:03,120
Speaker 1: there's a new social media grift which has to do

593
00:31:03,240 --> 00:31:06,280
Speaker 1: with TikTok and tales of infidelity.

594
00:31:06,360 --> 00:31:06,960
Speaker 2: Fill us in.

595
00:31:07,280 --> 00:31:12,920
Speaker 3: Yeah, it's all across short form video. Actually, so it's TikTok, YouTube, shorts, Instagram,

596
00:31:13,440 --> 00:31:16,960
Speaker 3: even Twitter. Basically, cheating videos are going viral, and so

597
00:31:17,040 --> 00:31:20,760
Speaker 3: you have these accounts of couples. They're almost all universally

598
00:31:20,920 --> 00:31:23,280
Speaker 3: sort of straight couples. Who knows if these people are

599
00:31:23,280 --> 00:31:26,440
Speaker 3: even dating, but if you look at their profiles, it

600
00:31:26,640 --> 00:31:31,040
Speaker 3: is just every single video is them is one partner

601
00:31:31,320 --> 00:31:35,479
Speaker 3: discovering that the other person is cheating through sort of

602
00:31:35,520 --> 00:31:39,240
Speaker 3: a lot of various extreme click baity ways. And this

603
00:31:39,280 --> 00:31:41,840
Speaker 3: is because this generates an enormous amount of tension. So

604
00:31:42,560 --> 00:31:46,200
Speaker 3: I think with the move to algorithmic feeds, people aren't

605
00:31:46,240 --> 00:31:49,160
Speaker 3: following creators as much. They don't see sort of like

606
00:31:49,200 --> 00:31:51,360
Speaker 3: every piece of content from a specific person. They just

607
00:31:51,360 --> 00:31:53,440
Speaker 3: see one super viral video come up in their feed,

608
00:31:53,720 --> 00:31:56,320
Speaker 3: and if it's outrage bait, like a cheating video, they're

609
00:31:56,320 --> 00:31:57,960
Speaker 3: going to comment, they're going to engage, and then that

610
00:31:58,040 --> 00:31:59,600
Speaker 3: video is going to be pushed to more users that

611
00:31:59,640 --> 00:32:03,840
Speaker 3: boost person's profile. But if you look at their page,

612
00:32:03,880 --> 00:32:06,160
Speaker 3: you see that every single video is them discovering cheating.

613
00:32:06,240 --> 00:32:09,240
Speaker 3: All of this, and all of this is in service

614
00:32:09,280 --> 00:32:12,320
Speaker 3: of promoting apps. There's a lot of cheater discovery apps

615
00:32:12,360 --> 00:32:16,320
Speaker 3: basically like low quality vibe coded data harvesting apps, which

616
00:32:16,400 --> 00:32:19,240
Speaker 3: is ultimately what these videos are promoting in their bios.

617
00:32:19,640 --> 00:32:22,000
Speaker 2: Let's play one of the videos. Well, you're not so.

618
00:32:22,040 --> 00:32:24,440
Speaker 5: Lucky because Sarah is at the door.

619
00:32:25,160 --> 00:32:26,440
Speaker 2: I invent you're over for dinner.

620
00:32:28,880 --> 00:32:31,000
Speaker 5: Sarah, the girl you met on dating apps. I already know,

621
00:32:31,080 --> 00:32:34,000
Speaker 5: we've already had a full conversation, so I called her

622
00:32:34,040 --> 00:32:35,120
Speaker 5: an invite her for dinner.

623
00:32:35,160 --> 00:32:39,080
Speaker 2: So I about it. Sarah is a personal trainer.

624
00:32:39,440 --> 00:32:42,640
Speaker 5: Remember I told you at the gym I was getting

625
00:32:42,640 --> 00:32:44,080
Speaker 5: someone to help me get back in shape.

626
00:32:44,120 --> 00:32:47,720
Speaker 2: Sarah's the personal trainer. We just use different messaging apps.

627
00:32:47,760 --> 00:32:50,240
Speaker 2: That's all that is. That is well, anyways, letter.

628
00:32:50,280 --> 00:32:53,880
Speaker 3: Because you're being very rude right now, So there is

629
00:32:53,920 --> 00:32:58,160
Speaker 3: no Sarah. This is all just baits to get you

630
00:32:58,360 --> 00:33:01,680
Speaker 3: to get outraged, comment and potentially download one of these

631
00:33:01,680 --> 00:33:02,320
Speaker 3: scamming apps.

632
00:33:03,040 --> 00:33:05,360
Speaker 2: This is what I don't understand. Like this kind of

633
00:33:05,400 --> 00:33:08,120
Speaker 2: stuff just makes me not want to use social media.

634
00:33:08,200 --> 00:33:09,920
Speaker 2: It's like part of the reason I never go on

635
00:33:09,960 --> 00:33:12,840
Speaker 2: Facebook anymore is because I started getting served these not

636
00:33:12,960 --> 00:33:16,040
Speaker 2: the cheating ones, but like I would get served these

637
00:33:16,120 --> 00:33:20,000
Speaker 2: videos that were so clearly staged, like just it could

638
00:33:20,040 --> 00:33:22,280
Speaker 2: just be funny, like, but it was like this is

639
00:33:22,400 --> 00:33:26,000
Speaker 2: so bullshit, like what am I doing here? And I

640
00:33:26,120 --> 00:33:28,920
Speaker 2: just stop using it. But like, apparently that's not how

641
00:33:29,000 --> 00:33:33,000
Speaker 2: most people view it. I don't know, Taylor, you're the expert.

642
00:33:33,480 --> 00:33:36,480
Speaker 3: Yeah, well, I mean I think read like you recognize

643
00:33:36,480 --> 00:33:38,760
Speaker 3: them as fake. I think the problem here is media

644
00:33:38,800 --> 00:33:42,920
Speaker 3: literacy and the fact that people cannot recognize the most

645
00:33:42,960 --> 00:33:45,680
Speaker 3: staged interaction ever. I mean, people used to sort of

646
00:33:45,680 --> 00:33:50,040
Speaker 3: criticize conservatives for this because they would see another conservative

647
00:33:50,160 --> 00:33:53,280
Speaker 3: pretending to be like I'm a blue haired, woke liberal

648
00:33:53,320 --> 00:33:57,719
Speaker 3: and my pronouns are pony rides or something you know like,

649
00:33:57,800 --> 00:34:02,320
Speaker 3: and this person is obviously trying to parody sort of

650
00:34:02,720 --> 00:34:04,719
Speaker 3: you know, what they consider to be a liberal, but

651
00:34:04,840 --> 00:34:07,120
Speaker 3: other conservatives see that and get outraged and take it

652
00:34:07,120 --> 00:34:08,880
Speaker 3: as page value. So I feel like we saw a

653
00:34:08,920 --> 00:34:12,759
Speaker 3: lot of that during Trump one point zero, and now

654
00:34:12,760 --> 00:34:14,959
Speaker 3: it's sort of proliferating across the Internet. I think because

655
00:34:15,000 --> 00:34:18,279
Speaker 3: of algorithmic feeds and because ultimately read like you might

656
00:34:18,320 --> 00:34:20,560
Speaker 3: click off. I think a huge amount of the Internet

657
00:34:20,680 --> 00:34:24,640
Speaker 3: does not have media literacy, and it boosts this app.

658
00:34:24,760 --> 00:34:26,880
Speaker 3: And the more people click on the app, you know,

659
00:34:26,960 --> 00:34:30,719
Speaker 3: in the bio or whatever, the more you know, downloads

660
00:34:30,719 --> 00:34:32,719
Speaker 3: it gets, the more profitable it is, and so this

661
00:34:32,800 --> 00:34:34,480
Speaker 3: sort of industrial complex continues.

662
00:34:34,960 --> 00:34:36,680
Speaker 1: I wish that Kyle is here because this is kind

663
00:34:36,680 --> 00:34:39,000
Speaker 1: of a continuation of our conversation last week. But how

664
00:34:39,080 --> 00:34:42,680
Speaker 1: is this different like watching World Wrestling Federation or even

665
00:34:42,800 --> 00:34:45,480
Speaker 1: reality TV, where like it's still fun even though you

666
00:34:45,480 --> 00:34:47,320
Speaker 1: know it's fake, Like, what do you think most people

667
00:34:47,360 --> 00:34:50,280
Speaker 1: don't know this is fake? And is that an answerable question?

668
00:34:50,440 --> 00:34:52,120
Speaker 3: Or I think if you look at the comments, most

669
00:34:52,160 --> 00:34:55,080
Speaker 3: people don't realize it's fake because again, they're just consuming

670
00:34:55,120 --> 00:34:59,480
Speaker 3: this video in a vacuum, and you know, you're like

671
00:35:00,440 --> 00:35:04,560
Speaker 3: exhausted flipping through the timeline. You're not on attention, you know.

672
00:35:04,880 --> 00:35:08,200
Speaker 3: And cheating videos, especially with these these are meant to

673
00:35:08,239 --> 00:35:11,359
Speaker 3: appeal to other young people, looks like other teenagers, high

674
00:35:11,400 --> 00:35:15,400
Speaker 3: school college kids. It's like very visceral, right, you feel

675
00:35:15,480 --> 00:35:18,319
Speaker 3: this like frustration or you feel a lot of them

676
00:35:18,360 --> 00:35:21,040
Speaker 3: have to do with the women being vindicated she knew

677
00:35:21,080 --> 00:35:23,880
Speaker 3: the man was cheating, and we see this stuff with

678
00:35:23,960 --> 00:35:26,200
Speaker 3: like these narratives are so compelling. I mean, it reminds

679
00:35:26,200 --> 00:35:29,080
Speaker 3: you of like the Ai fruit slot videos, right, so

680
00:35:29,200 --> 00:35:32,480
Speaker 3: many of those are cheating narratives. Like there's something about

681
00:35:32,560 --> 00:35:37,040
Speaker 3: this idea of like betrayal that just taps into some

682
00:35:37,120 --> 00:35:40,400
Speaker 3: like base human emotion and gets people to engage totally

683
00:35:40,480 --> 00:35:42,000
Speaker 3: low on the brainstem.

684
00:35:42,400 --> 00:35:44,799
Speaker 2: I guess I'm asking your same question, Oz, But like

685
00:35:45,200 --> 00:35:46,839
Speaker 2: is it different from me when I was a kid,

686
00:35:46,920 --> 00:35:49,839
Speaker 2: this was like after school specials or like later like

687
00:35:49,960 --> 00:35:53,640
Speaker 2: lifetime television. Like this is like people just watch this

688
00:35:53,760 --> 00:35:56,360
Speaker 2: type of content. Is it even a problem? I guess

689
00:35:56,760 --> 00:35:57,480
Speaker 2: is the other question?

690
00:35:57,600 --> 00:36:00,000
Speaker 3: I guess, like to me, okay, after school specials or

691
00:36:00,120 --> 00:36:04,560
Speaker 3: probably meant to be educational lifetime content soap operas that's

692
00:36:04,600 --> 00:36:09,160
Speaker 3: also very clearly fictional. And yeah, people love the drama

693
00:36:09,160 --> 00:36:12,680
Speaker 3: of reality TV. But again they recognize and consent to

694
00:36:12,719 --> 00:36:17,359
Speaker 3: the fact, of course it's all staged, but they recognize that,

695
00:36:17,440 --> 00:36:20,240
Speaker 3: you know, reality TV fans are aware of the fact

696
00:36:20,320 --> 00:36:22,960
Speaker 3: that these storylines are.

697
00:36:22,600 --> 00:36:26,160
Speaker 2: Like do people watch reality TV and they just know, like, oh,

698
00:36:26,239 --> 00:36:27,479
Speaker 2: like this this is like.

699
00:36:28,160 --> 00:36:31,320
Speaker 1: Some ambiguity, but like so that the joy of reality

700
00:36:31,400 --> 00:36:33,840
Speaker 1: TV is like there's a code of real in it, right.

701
00:36:33,800 --> 00:36:38,000
Speaker 3: Sure, But I think, I mean, I think at this

702
00:36:38,120 --> 00:36:41,319
Speaker 3: point in the reality TV universe, every single person going

703
00:36:41,360 --> 00:36:43,359
Speaker 3: on reality TV is going on there to launch their

704
00:36:43,400 --> 00:36:46,640
Speaker 3: influencer career. Everyone's aware of the storylines. I mean, this

705
00:36:46,680 --> 00:36:49,920
Speaker 3: has actually been a huge problem for reality TV casting

706
00:36:50,239 --> 00:36:52,839
Speaker 3: h executives. I wrote a story about this a couple

707
00:36:52,840 --> 00:36:54,920
Speaker 3: of years ago. But they're having a harder and harder

708
00:36:55,040 --> 00:36:58,360
Speaker 3: time finding people that won't just perform for the cameras,

709
00:36:58,360 --> 00:37:00,680
Speaker 3: that will react naturally, and so they're getting sort of

710
00:37:00,719 --> 00:37:04,400
Speaker 3: a lot of unhinged people on these shows. But but

711
00:37:04,160 --> 00:37:06,680
Speaker 3: the but with these videos, I think it's not clear.

712
00:37:06,760 --> 00:37:09,640
Speaker 3: It looks organic, and a lot of times they're filmed shakely.

713
00:37:10,920 --> 00:37:12,759
Speaker 3: It is meant to trick you, to make you think

714
00:37:12,760 --> 00:37:16,279
Speaker 3: it's organic content, and it's directly profiting off you by

715
00:37:16,440 --> 00:37:18,000
Speaker 3: you know, pushing this slot app.

716
00:37:18,200 --> 00:37:21,760
Speaker 2: My other question, Taylor is like, is this a technology story?

717
00:37:21,960 --> 00:37:24,520
Speaker 2: Like I really I look at it and I'm like,

718
00:37:24,600 --> 00:37:26,840
Speaker 2: where's what does this have to do with technology? And

719
00:37:27,000 --> 00:37:29,080
Speaker 2: this is not a criticism. I just think do you

720
00:37:29,120 --> 00:37:30,120
Speaker 2: see it as a tech story?

721
00:37:30,239 --> 00:37:32,239
Speaker 3: I do see it as a tech story. I mean,

722
00:37:32,280 --> 00:37:34,799
Speaker 3: first of all, I cover like tech media sort of

723
00:37:34,800 --> 00:37:37,239
Speaker 3: adjacent stuff, so I'm always interested in sort of like

724
00:37:37,280 --> 00:37:39,680
Speaker 3: the mechanisms of content. I think it has to do

725
00:37:39,719 --> 00:37:42,000
Speaker 3: with algorithmic feeds. I think it has to do with

726
00:37:42,120 --> 00:37:47,319
Speaker 3: the death of this subscriber model of influencer culture that dominated,

727
00:37:47,680 --> 00:37:49,800
Speaker 3: you know, the initial sort of wave of social media.

728
00:37:49,840 --> 00:37:50,800
Speaker 2: And I also think.

729
00:37:50,640 --> 00:37:53,040
Speaker 3: That it's notable that all of these are like vibe

730
00:37:53,040 --> 00:37:56,800
Speaker 3: coded apps and we're seeing the explosion of app development

731
00:37:56,800 --> 00:37:59,440
Speaker 3: and people building spinning up these products super quickly. I mean,

732
00:37:59,480 --> 00:38:01,560
Speaker 3: Oasis just got in big trouble for this this health

733
00:38:01,560 --> 00:38:04,640
Speaker 3: app that now was doing a really aggressive like TikTok

734
00:38:04,680 --> 00:38:07,200
Speaker 3: marketing and actually it turns out is now getting sued

735
00:38:07,200 --> 00:38:10,440
Speaker 3: out of business. Like but you know, if you develop

736
00:38:10,440 --> 00:38:12,400
Speaker 3: an app and you want a profit, which is this

737
00:38:12,480 --> 00:38:14,319
Speaker 3: dream that's been sold to you by Silicon Valley for

738
00:38:14,320 --> 00:38:16,760
Speaker 3: over a decade. Now you've got to market that app,

739
00:38:16,880 --> 00:38:20,120
Speaker 3: and how do you market the app? People are engaging

740
00:38:20,120 --> 00:38:22,680
Speaker 3: in increasingly sort of like scammy ways of doing that.

741
00:38:23,239 --> 00:38:25,759
Speaker 2: But the algorithms are not new technology, Like, at what

742
00:38:25,800 --> 00:38:29,120
Speaker 2: point does like like these algorithms were invented, like you

743
00:38:29,160 --> 00:38:31,480
Speaker 2: know what is like fifteen twenty years ago? Now?

744
00:38:31,600 --> 00:38:35,840
Speaker 3: No, no, But in the world is the TikTok algorithm

745
00:38:35,880 --> 00:38:37,040
Speaker 3: invented fifteen years ago?

746
00:38:37,200 --> 00:38:38,880
Speaker 2: No? It was not. No, but I mean the TikTok

747
00:38:39,280 --> 00:38:42,200
Speaker 2: algorithm wasn't invented by tick like this. This was like

748
00:38:42,280 --> 00:38:43,839
Speaker 2: I remember this was the first one of the first

749
00:38:43,880 --> 00:38:47,040
Speaker 2: conversations I had with someone at Facebook was like, you

750
00:38:47,040 --> 00:38:50,240
Speaker 2: know about the algorithm back in twenty thirds Sure.

751
00:38:50,080 --> 00:38:52,600
Speaker 3: They had, but that was a very crude algorithm. And

752
00:38:52,680 --> 00:38:55,160
Speaker 3: let me just be clear, we didn't even really start

753
00:38:55,200 --> 00:38:58,719
Speaker 3: to see the beginning of algorithmic feeds until twenty sixteen

754
00:38:58,880 --> 00:39:02,400
Speaker 3: when Instagram rolled it out, when Twitter rolled it out

755
00:39:02,440 --> 00:39:05,560
Speaker 3: for the first time, and those were still very crude.

756
00:39:05,680 --> 00:39:06,520
Speaker 2: I think we.

757
00:39:06,440 --> 00:39:09,920
Speaker 3: Also see the shifting nature of what these algorithms prioritize

758
00:39:10,000 --> 00:39:13,920
Speaker 3: how people engage with the content, and now again they

759
00:39:14,040 --> 00:39:16,960
Speaker 3: used to wait following used to the people that you

760
00:39:17,160 --> 00:39:19,560
Speaker 3: used to see were more regular. This is a big

761
00:39:19,600 --> 00:39:22,880
Speaker 3: conversation and influencers, you know. But as the tech platforms

762
00:39:22,920 --> 00:39:25,600
Speaker 3: developed this sort of hostile relationship with creators where they

763
00:39:25,600 --> 00:39:28,000
Speaker 3: don't want the creators to have so much power. Because

764
00:39:28,000 --> 00:39:30,680
Speaker 3: these creators started to ask the tech platforms things, they're

765
00:39:30,719 --> 00:39:32,360
Speaker 3: trying to seize back the power and so you are

766
00:39:32,440 --> 00:39:34,600
Speaker 3: less likely to see content from the same person in

767
00:39:34,640 --> 00:39:36,560
Speaker 3: your feed. That's why you're getting a lot more of

768
00:39:36,600 --> 00:39:39,120
Speaker 3: this engagement baits Lop. I also think people are better

769
00:39:39,200 --> 00:39:41,200
Speaker 3: at optimizing for outread kind.

770
00:39:41,440 --> 00:39:43,799
Speaker 2: I just see them as sort of editorial decisions, like

771
00:39:43,840 --> 00:39:45,400
Speaker 2: the way they tweak the algorithms.

772
00:39:45,400 --> 00:39:48,279
Speaker 3: Is it is an editorial decision, Sure it is, and

773
00:39:48,280 --> 00:39:49,359
Speaker 3: that's fine, I mean to.

774
00:39:49,320 --> 00:39:52,000
Speaker 2: Me, although they would disagree with that's a real term.

775
00:39:52,040 --> 00:39:55,600
Speaker 1: Now it has signs do with the with the metaphus.

776
00:39:55,640 --> 00:39:57,760
Speaker 1: I mean, Taylor, you spoke about this last week, but basically,

777
00:39:58,200 --> 00:40:01,200
Speaker 1: real people doing stuff in there, real lives to like

778
00:40:01,239 --> 00:40:04,400
Speaker 1: monetize in other ways, whether on Calshi or whatever it

779
00:40:04,440 --> 00:40:06,960
Speaker 1: may be. And so I mean I was I don't

780
00:40:06,960 --> 00:40:08,640
Speaker 1: know how many of you saw this horrific story of

781
00:40:08,680 --> 00:40:11,680
Speaker 1: the Brazilian woman who jumped off the bungee jumping thing

782
00:40:11,840 --> 00:40:14,760
Speaker 1: and she wasn't tied on and it was just awful,

783
00:40:14,960 --> 00:40:17,279
Speaker 1: And all of a sudden, everyone was questioning, was this

784
00:40:17,400 --> 00:40:19,000
Speaker 1: like a stunt? Was this a snuff movie?

785
00:40:19,040 --> 00:40:19,160
Speaker 2: Like?

786
00:40:19,280 --> 00:40:21,319
Speaker 1: Why were there so many camera angles? And so I think,

787
00:40:21,320 --> 00:40:23,560
Speaker 1: to me, the tech angle is like, we've come to

788
00:40:23,600 --> 00:40:25,480
Speaker 1: this place where we have to we have to question

789
00:40:25,760 --> 00:40:29,799
Speaker 1: every single thing we ever see and need to interrogate

790
00:40:29,840 --> 00:40:32,239
Speaker 1: whether it was created just for us even I mean,

791
00:40:32,400 --> 00:40:35,440
Speaker 1: pray god, that wasn't like a stage snuff movie. But

792
00:40:35,480 --> 00:40:36,919
Speaker 1: I mean, it's just crazy that we have to ask

793
00:40:36,960 --> 00:40:37,400
Speaker 1: that question.

794
00:40:37,920 --> 00:40:39,880
Speaker 2: But isn't that a good thing? Like shouldn't we be

795
00:40:39,960 --> 00:40:44,040
Speaker 2: scrutinizing every single thing we see and questioning everything? I

796
00:40:44,040 --> 00:40:47,040
Speaker 2: don't mean we should read I agree.

797
00:40:47,360 --> 00:40:50,960
Speaker 3: I think my issue is that is that people are

798
00:40:51,000 --> 00:40:54,560
Speaker 3: not questioning everything. I think I think or people are

799
00:40:54,600 --> 00:40:57,160
Speaker 3: overly conspiratorial because they don't know what to trust. As

800
00:40:57,200 --> 00:40:59,680
Speaker 3: you mentioned, people immediately jump to conspiracies. Why is this,

801
00:40:59,719 --> 00:41:02,000
Speaker 3: Why is this? Why are there so many angles that

802
00:41:01,880 --> 00:41:05,200
Speaker 3: it's like, well, everyone has a phone.

803
00:41:04,600 --> 00:41:05,239
Speaker 2: But I don't know.

804
00:41:05,320 --> 00:41:08,120
Speaker 3: I do think it's interesting to look at sort of

805
00:41:08,160 --> 00:41:13,560
Speaker 3: like what these platforms reward, what sort of businesses they're facilitating,

806
00:41:15,120 --> 00:41:19,240
Speaker 3: and and kind of like how, yeah, how technology is progressing,

807
00:41:19,239 --> 00:41:22,520
Speaker 3: Like even just the market for all of these cheating apps.

808
00:41:22,600 --> 00:41:24,799
Speaker 3: I just think that, like app development used to take

809
00:41:24,880 --> 00:41:28,319
Speaker 3: time somewhat, it's been so democratized. You can spin up

810
00:41:28,360 --> 00:41:31,279
Speaker 3: these apps so quickly. And we're just seeing a lot

811
00:41:31,320 --> 00:41:35,640
Speaker 3: more marketing for apps period in our feeds because of

812
00:41:36,040 --> 00:41:37,160
Speaker 3: the rise of these tools.

813
00:41:37,760 --> 00:41:40,319
Speaker 1: And do you think there's a corollary with the fact

814
00:41:40,320 --> 00:41:43,080
Speaker 1: that the web is becoming most web search is now

815
00:41:43,239 --> 00:41:46,480
Speaker 1: conducted by AI agents rather than people, and therefore, like

816
00:41:46,480 --> 00:41:49,640
Speaker 1: there's an increased pressure on these like social videos as

817
00:41:49,680 --> 00:41:52,600
Speaker 1: the one place that humans still go to be guided

818
00:41:52,719 --> 00:41:55,839
Speaker 1: for their product buying decisions. I mean, is there any

819
00:41:55,840 --> 00:41:59,040
Speaker 1: connection between like agentic web search and this phenomenon, do

820
00:41:59,040 --> 00:41:59,600
Speaker 1: you think.

821
00:42:00,000 --> 00:42:03,120
Speaker 3: I don't think those are maybe necessarily directly connected. Again,

822
00:42:03,160 --> 00:42:05,399
Speaker 3: I think it's just the fact that like I see

823
00:42:05,440 --> 00:42:08,600
Speaker 3: this also with like I'm very much on like health Instagram,

824
00:42:08,719 --> 00:42:11,719
Speaker 3: and like I see this with like again the Oasis app,

825
00:42:11,719 --> 00:42:14,279
Speaker 3: which is this really viral app, But there's so there's

826
00:42:14,320 --> 00:42:16,360
Speaker 3: like hundreds of clones of this app, which is this

827
00:42:16,440 --> 00:42:18,960
Speaker 3: idea of like we'll tell you which products are not

828
00:42:19,000 --> 00:42:21,120
Speaker 3: going to kill you because we all go around the

829
00:42:21,160 --> 00:42:23,000
Speaker 3: world where you know, it's like our pants are covered

830
00:42:23,000 --> 00:42:25,840
Speaker 3: with pfas or whatever, right, like everything we eat is toxic.

831
00:42:25,920 --> 00:42:28,000
Speaker 3: Like how can I navigate this world? This is a

832
00:42:28,000 --> 00:42:30,399
Speaker 3: growing problem, and there's a growing amount of like vibe

833
00:42:30,440 --> 00:42:34,279
Speaker 3: coded apps that are aimed at addressing it. And the

834
00:42:34,360 --> 00:42:36,120
Speaker 3: thing is is there is no way to do this testing,

835
00:42:36,160 --> 00:42:38,400
Speaker 3: to do this testing well, the thing is is that

836
00:42:38,480 --> 00:42:42,520
Speaker 3: doing the actual testing on these products is incredibly expensive

837
00:42:42,600 --> 00:42:44,799
Speaker 3: and time consuming, and this is why there is no

838
00:42:44,880 --> 00:42:47,920
Speaker 3: platform that really does it super effectively, except maybe, like

839
00:42:48,080 --> 00:42:50,120
Speaker 3: you know, some environmental ones that have been around forever.

840
00:42:50,560 --> 00:42:53,280
Speaker 3: But these vibe coders are just like scraping tons of info,

841
00:42:53,760 --> 00:42:56,520
Speaker 3: spinning up these apps and then engaging in outrage marketing

842
00:42:56,719 --> 00:42:59,640
Speaker 3: to get tons of downloads, and before they know it,

843
00:42:59,680 --> 00:43:02,560
Speaker 3: they've potentially committed fraud and slandered a bunch of companies

844
00:43:02,560 --> 00:43:04,239
Speaker 3: for being toxic when they aren't.

845
00:43:04,719 --> 00:43:06,480
Speaker 2: I mean, none of that sounds really new to me,

846
00:43:06,600 --> 00:43:09,080
Speaker 2: to be honest, but like I think, I mean, it

847
00:43:09,120 --> 00:43:12,080
Speaker 2: goes back to the scale is maybe you could argue

848
00:43:12,080 --> 00:43:14,480
Speaker 2: to scale, but I think I think this democratization of

849
00:43:14,560 --> 00:43:17,160
Speaker 2: software development, which is leading to these vibe coded apps,

850
00:43:17,280 --> 00:43:20,120
Speaker 2: is like it's kind of a moment in time, but

851
00:43:20,239 --> 00:43:22,840
Speaker 2: I don't I don't see it like continuing because I

852
00:43:22,880 --> 00:43:26,239
Speaker 2: think ultimately like that's not the future is not just

853
00:43:26,280 --> 00:43:28,680
Speaker 2: downloading a bunch of random apps. The future is like

854
00:43:29,160 --> 00:43:33,600
Speaker 2: truly customized. I agree with you, but I think, Natasha,

855
00:43:33,640 --> 00:43:33,920
Speaker 2: we go.

856
00:43:35,640 --> 00:43:38,480
Speaker 5: Okay, well, first of all, let's please.

857
00:43:38,360 --> 00:43:40,760
Speaker 2: I agree with you, and then go yeah.

858
00:43:41,719 --> 00:43:44,040
Speaker 5: Please please not use the word democratization.

859
00:43:44,320 --> 00:43:47,520
Speaker 2: I use that on purpose because I knew it.

860
00:43:47,760 --> 00:43:50,840
Speaker 5: Because you knew it would it would bump me. I

861
00:43:50,920 --> 00:43:54,400
Speaker 5: was like, okay, well I think like I mean, I

862
00:43:54,800 --> 00:43:58,719
Speaker 5: think like, first of all, it doesn't matter if it's

863
00:43:58,960 --> 00:44:02,520
Speaker 5: not new, because this is like like this is a

864
00:44:02,520 --> 00:44:06,320
Speaker 5: story where we get to see business decisions impact people,

865
00:44:06,360 --> 00:44:08,799
Speaker 5: and I would say it is new, like read It's

866
00:44:08,840 --> 00:44:12,600
Speaker 5: not like they invented these algorithms recommend our algorithms fifteen

867
00:44:12,680 --> 00:44:15,360
Speaker 5: years ago and then keep it. They are applying the

868
00:44:15,560 --> 00:44:19,279
Speaker 5: same technology to you know, like the same generative AI.

869
00:44:19,320 --> 00:44:21,760
Speaker 5: It's very good at coming up with profiles of people,

870
00:44:21,800 --> 00:44:24,879
Speaker 5: So who are seeing these videos like vulnerable women. What

871
00:44:25,200 --> 00:44:27,960
Speaker 5: like what data are they collecting about you? How are

872
00:44:28,040 --> 00:44:30,640
Speaker 5: they putting it together in this profile that you would

873
00:44:30,800 --> 00:44:33,879
Speaker 5: you know, let your defenses down and think that, oh

874
00:44:33,880 --> 00:44:35,920
Speaker 5: my god, this is a this is somebody who's being

875
00:44:35,960 --> 00:44:38,239
Speaker 5: cheated on just like me, and how could this be

876
00:44:38,360 --> 00:44:42,040
Speaker 5: applied in other scenarios. I also think like the fact

877
00:44:42,120 --> 00:44:45,840
Speaker 5: that Apple is not taking you know, good control of

878
00:44:45,840 --> 00:44:47,799
Speaker 5: the app store and there are all these scammy apps.

879
00:44:47,800 --> 00:44:50,080
Speaker 5: I mean I see the same with just like homework apps,

880
00:44:50,120 --> 00:44:53,000
Speaker 5: you know, like, oh this really really helped me, Like

881
00:44:53,080 --> 00:44:54,640
Speaker 5: everything that I looked.

882
00:44:54,320 --> 00:45:01,200
Speaker 2: At when we work together, I literally wrote a whole

883
00:45:01,520 --> 00:45:04,680
Speaker 2: did a whole investigation. I'm like the fact that the

884
00:45:04,719 --> 00:45:06,720
Speaker 2: Apple app store was fatal scams.

885
00:45:07,239 --> 00:45:13,440
Speaker 5: I know, I know, but if it continues, isn't that

886
00:45:13,600 --> 00:45:15,440
Speaker 5: worth No, it's worth covering.

887
00:45:15,520 --> 00:45:18,600
Speaker 2: It's it's worth covering. I'm not saying. I'm saying it's

888
00:45:18,640 --> 00:45:20,880
Speaker 2: not new, but it's I agree that it's worth covering

889
00:45:29,680 --> 00:45:31,280
Speaker 2: for tech stuff. I'm as Valoshin.

890
00:45:31,560 --> 00:45:34,440
Speaker 1: This episode was produced by Eliza Dennis and Melissa Slaughter.

891
00:45:35,000 --> 00:45:37,719
Speaker 1: It was executive produced by me Julian Nutter and Kate

892
00:45:37,719 --> 00:45:41,960
Speaker 1: Osborne for Kaleidoscope and Katrina novel for iHeart podcasts. Our

893
00:45:42,000 --> 00:45:45,239
Speaker 1: engineer was Bihid Fraser and Kyle Murdoch wrote our theme

894
00:45:45,280 --> 00:46:01,239
Speaker 1: song six