1
00:00:02,120 --> 00:00:05,840
Speaker 1: Welcome to Tech Stuff, a production of iHeart Podcasts and Kaleidoscope.

2
00:00:06,200 --> 00:00:08,920
Speaker 1: I'm Osvoloshin and today Kara Price and I will bring

3
00:00:08,960 --> 00:00:12,959
Speaker 1: you the headlines this week, well tariffs, obviously, but also

4
00:00:13,119 --> 00:00:16,159
Speaker 1: a dating game. Then on Tech Support, we'll talk to

5
00:00:16,200 --> 00:00:19,520
Speaker 1: The Wall Street Journal's Family and Tech columnist Julie Jargon

6
00:00:19,920 --> 00:00:22,919
Speaker 1: about a mother's worst fear, a cry for help over

7
00:00:22,960 --> 00:00:26,360
Speaker 1: the phone that sounded like her youngest daughter, all of

8
00:00:26,360 --> 00:00:40,320
Speaker 1: that on the Weekend Tech. It's Friday, April eleventh. Kara, Hello, Hello, Hello, ohs.

9
00:00:40,880 --> 00:00:43,040
Speaker 2: I think I'm going to start this thing called Kara's

10
00:00:43,080 --> 00:00:45,400
Speaker 2: Hats of the Week. And just for people who can't

11
00:00:45,440 --> 00:00:48,239
Speaker 2: see me, I'm wearing a hat today that says, and

12
00:00:48,280 --> 00:00:50,080
Speaker 2: this is from a show called Summer Heatsie. If you

13
00:00:50,120 --> 00:00:53,640
Speaker 2: ever watched Summerheatsie, I'm a naughty girl with a bad habit.

14
00:00:53,920 --> 00:00:55,639
Speaker 1: What is the bad habit in the.

15
00:00:55,600 --> 00:00:59,200
Speaker 2: Show, It's for drugs. If I had preempted today's top story,

16
00:00:59,200 --> 00:01:01,720
Speaker 2: it would have said, with a bad habit for tariffs.

17
00:01:01,800 --> 00:01:05,600
Speaker 1: Uh huh. That would have been fast fashion. Indeed, that's

18
00:01:05,600 --> 00:01:08,480
Speaker 1: a good one. But today's news is all about quote.

19
00:01:08,720 --> 00:01:10,479
Speaker 1: The most beautiful word in the dictionary.

20
00:01:10,560 --> 00:01:13,080
Speaker 2: Well, I don't know anything about the dictionary, but I

21
00:01:13,240 --> 00:01:16,480
Speaker 2: looked up the word tariff with chat GPT.

22
00:01:16,640 --> 00:01:18,959
Speaker 1: The Dictionary twenty twenty five correct.

23
00:01:18,600 --> 00:01:22,040
Speaker 2: Correct, and the thesaurus and also what we'll eventually write

24
00:01:22,120 --> 00:01:26,960
Speaker 2: someone's wedding vesse. I asked chat Gpt what a tariff is,

25
00:01:27,360 --> 00:01:30,479
Speaker 2: and she responded because she's a she yea in my book.

26
00:01:30,520 --> 00:01:33,280
Speaker 2: In your book, a tariff is a tax or fee

27
00:01:33,480 --> 00:01:37,000
Speaker 2: that a government imposes on imported or exported goods. But

28
00:01:37,160 --> 00:01:41,559
Speaker 2: most importantly, chat Gpt says tariffs can impact the price

29
00:01:41,560 --> 00:01:45,039
Speaker 2: of goods, trade relationships, and the global economy.

30
00:01:45,280 --> 00:01:49,960
Speaker 1: Oh my prophetic soul can and do.

31
00:01:50,680 --> 00:01:53,800
Speaker 2: Chat Gipt was onto something there because Trump's tariff announcement

32
00:01:54,040 --> 00:01:57,120
Speaker 2: definitely shook up the global economy and trade relationships, and

33
00:01:57,200 --> 00:02:00,000
Speaker 2: everything is still evolving. I mean, there's basically a new

34
00:02:00,240 --> 00:02:03,960
Speaker 2: update every hour. Just this week, tariffs on about ninety

35
00:02:04,000 --> 00:02:06,960
Speaker 2: countries went into effect. Then Trump issued a pause on

36
00:02:07,000 --> 00:02:09,400
Speaker 2: most of them. But the country who's been pummeled the

37
00:02:09,400 --> 00:02:12,840
Speaker 2: most is China. As of Thursday midday, Trump has increased

38
00:02:12,880 --> 00:02:15,760
Speaker 2: tariffs on the country's imports by one hundred and twenty

39
00:02:15,760 --> 00:02:19,120
Speaker 2: five percent. So with everything going on, the technology industry

40
00:02:19,160 --> 00:02:20,680
Speaker 2: has been feeling some effects.

41
00:02:21,040 --> 00:02:23,359
Speaker 1: There's a headline in the Washington Post earlier this week

42
00:02:23,440 --> 00:02:27,079
Speaker 1: that was delicious in its understated irony. Big tech bet

43
00:02:27,120 --> 00:02:29,440
Speaker 1: on Trump. It's still waiting for the payoff.

44
00:02:29,760 --> 00:02:31,800
Speaker 2: I'm just thinking about when we first started this version

45
00:02:31,800 --> 00:02:33,600
Speaker 2: of tech stuff. All of those tech bros were at

46
00:02:33,600 --> 00:02:36,359
Speaker 2: the inauguration, and now they're all scrambling to rethink their

47
00:02:36,360 --> 00:02:37,160
Speaker 2: supply chains.

48
00:02:37,320 --> 00:02:40,320
Speaker 1: But it's also about the threat of reciprocal tariffs. And

49
00:02:40,360 --> 00:02:43,800
Speaker 1: in that Washington Post story, the writers quip those tech

50
00:02:43,840 --> 00:02:47,000
Speaker 1: giants are in another front row, lol, not the front

51
00:02:47,080 --> 00:02:50,840
Speaker 1: row the inauguration as targets for US trade partners looking

52
00:02:50,840 --> 00:02:53,480
Speaker 1: for ways to strike back at the US economy.

53
00:02:53,919 --> 00:02:56,600
Speaker 2: Yeah. I heard at one point the EU is considering

54
00:02:56,639 --> 00:03:01,000
Speaker 2: tariffs on digital products like Netflix subscriptions and Google Cloud storage,

55
00:03:01,000 --> 00:03:05,079
Speaker 2: which I honestly didn't even know was possible. But another

56
00:03:05,120 --> 00:03:07,480
Speaker 2: area of concern for the tech industry is how these

57
00:03:07,520 --> 00:03:11,400
Speaker 2: tariffs will affect semiconductors, because they really do power the

58
00:03:11,440 --> 00:03:16,120
Speaker 2: modern world. Everything from consumer tech, data centers, even cars,

59
00:03:16,240 --> 00:03:17,880
Speaker 2: they all use semiconductors.

60
00:03:18,080 --> 00:03:21,560
Speaker 1: And here's where the kind of ironies continue to abound, because,

61
00:03:21,560 --> 00:03:25,000
Speaker 1: of course, you know, President Trump has made AI supremacy

62
00:03:25,080 --> 00:03:28,720
Speaker 1: a key element of his policy for this term, and

63
00:03:28,880 --> 00:03:34,480
Speaker 1: so technically the tariffs announce an exemption for semiconductors, but

64
00:03:34,960 --> 00:03:37,560
Speaker 1: as it turns out, many of the semiconductors imported are

65
00:03:37,600 --> 00:03:41,360
Speaker 1: actually bundled into other products like GPU chips and service

66
00:03:41,560 --> 00:03:44,680
Speaker 1: to train AI models. That's per wired to have a

67
00:03:44,720 --> 00:03:47,920
Speaker 1: story under the headline Trump's tariffs are threatening the US

68
00:03:48,000 --> 00:03:51,640
Speaker 1: semiconductor revival. Wide also points out that all of the

69
00:03:51,680 --> 00:03:56,160
Speaker 1: machinery and the underlying materials to manufacture semiconductors here in

70
00:03:56,200 --> 00:03:59,600
Speaker 1: the US will become far more expensive with these tariffs

71
00:03:59,720 --> 00:04:01,840
Speaker 1: make it less attractive to manufacture domestically.

72
00:04:02,080 --> 00:04:04,280
Speaker 2: Yeah, you know this is a little bit heady, but

73
00:04:04,400 --> 00:04:07,080
Speaker 2: needless to say, it is a consumer tech story. You know,

74
00:04:07,160 --> 00:04:10,040
Speaker 2: there are many articles circulating about how these tariffs could

75
00:04:10,120 --> 00:04:13,320
Speaker 2: affect the price of something as ubiquitous as the iPhone.

76
00:04:13,360 --> 00:04:15,840
Speaker 1: Have you been stoking up for your ebase sales.

77
00:04:15,600 --> 00:04:18,560
Speaker 2: Of my iPhone? Oh to sell iPhones? Oh yeah, it's

78
00:04:18,600 --> 00:04:20,640
Speaker 2: not a bad idea. Actually, I hadn't thought about it,

79
00:04:21,000 --> 00:04:22,120
Speaker 2: you know. And just to give a shout out to

80
00:04:22,160 --> 00:04:24,080
Speaker 2: four A form Meta who we love, you know, they

81
00:04:24,160 --> 00:04:28,440
Speaker 2: pointed out that on Apple's own supply chain website. The

82
00:04:28,520 --> 00:04:32,480
Speaker 2: big beautiful bold text that overlays the video of people

83
00:04:32,520 --> 00:04:37,760
Speaker 2: making iPhones in a factory says, designed by Apple in California,

84
00:04:37,800 --> 00:04:42,600
Speaker 2: made by people everywhere. And it's true, you know, as

85
00:04:42,640 --> 00:04:46,720
Speaker 2: a device exemplifies globalization. The materials for the batteries come

86
00:04:46,760 --> 00:04:50,280
Speaker 2: from one country, the display from another. Almost every part

87
00:04:50,279 --> 00:04:52,760
Speaker 2: of the iPhone comes from a different country, and then

88
00:04:52,760 --> 00:04:56,440
Speaker 2: they are predominantly assembled in China. So if things go

89
00:04:56,560 --> 00:04:59,400
Speaker 2: the way they're going and tariffs on Chinese imports remain,

90
00:04:59,720 --> 00:05:03,480
Speaker 2: these phones could get a lot more expensive. And I'm

91
00:05:03,520 --> 00:05:05,159
Speaker 2: going to get the burner phone that I plan on

92
00:05:05,200 --> 00:05:06,479
Speaker 2: getting the summer anyway.

93
00:05:06,200 --> 00:05:08,400
Speaker 1: Okay, good, Yeah, Well this will be another inducement for

94
00:05:08,480 --> 00:05:11,719
Speaker 1: us to get dumb phones, but that that might be

95
00:05:11,800 --> 00:05:13,719
Speaker 1: enough for us on Tarish this week. I feel this

96
00:05:13,720 --> 00:05:15,839
Speaker 1: is probably something going to be coming back to again

97
00:05:15,880 --> 00:05:17,279
Speaker 1: and again. So time for a game.

98
00:05:17,520 --> 00:05:20,279
Speaker 2: I have been sitting on my hands for this entire

99
00:05:20,360 --> 00:05:22,960
Speaker 2: show as we talk about tariffs, to play a game

100
00:05:23,040 --> 00:05:25,640
Speaker 2: with you that came out last week for April Fools.

101
00:05:26,080 --> 00:05:28,520
Speaker 2: I will catch people up a little bit. Last week,

102
00:05:28,600 --> 00:05:32,160
Speaker 2: Tinder launched an in app game called The Game Game.

103
00:05:32,279 --> 00:05:34,840
Speaker 1: The Game Game it sounds a bit like that seventies

104
00:05:34,880 --> 00:05:37,960
Speaker 1: dating show, the dating game with a sprinkle of my

105
00:05:38,080 --> 00:05:39,520
Speaker 1: hero Neil Strauss.

106
00:05:39,680 --> 00:05:41,960
Speaker 2: I think they were definitely going for seventies dating show.

107
00:05:41,960 --> 00:05:43,640
Speaker 2: I don't think they were going for a sprinkle of

108
00:05:43,680 --> 00:05:47,280
Speaker 2: Neil Strauss. But that's your drama. So basically, Open AI's Chat,

109
00:05:47,360 --> 00:05:51,080
Speaker 2: GPT four oh and Tinder partner together to create this

110
00:05:51,120 --> 00:05:54,960
Speaker 2: thing called the Game Game, which allows real Tinder users

111
00:05:55,480 --> 00:05:59,080
Speaker 2: to enter pretend scenarios and talk to AI characters.

112
00:05:59,080 --> 00:06:02,640
Speaker 1: This was a meet cute between chat ChiPT and Tinder,

113
00:06:03,360 --> 00:06:04,359
Speaker 1: good one if you will.

114
00:06:04,440 --> 00:06:07,320
Speaker 2: It was a me cute between altman and Tinder. Absolutely,

115
00:06:07,720 --> 00:06:10,680
Speaker 2: But there's also a competitive part of the game, which

116
00:06:10,680 --> 00:06:13,120
Speaker 2: is that as you talk into your phone, you try

117
00:06:13,160 --> 00:06:15,520
Speaker 2: your best to flirt with the AI, and you get

118
00:06:15,560 --> 00:06:19,960
Speaker 2: points for how suave or empathetic or interesting your responses are.

119
00:06:20,120 --> 00:06:23,480
Speaker 2: And here's how it works. You entered a preferred scenario

120
00:06:23,880 --> 00:06:28,080
Speaker 2: like I'm on a train and my shoes untied and

121
00:06:28,120 --> 00:06:31,800
Speaker 2: a man says, miss, your shoes untied and I say, sir,

122
00:06:32,480 --> 00:06:34,440
Speaker 2: I'm into women. That would be the end of that.

123
00:06:34,640 --> 00:06:36,880
Speaker 2: But no, if I were straight, he would say, mam,

124
00:06:36,920 --> 00:06:39,320
Speaker 2: your shoes untied and I'd look up and it would

125
00:06:39,320 --> 00:06:46,600
Speaker 2: get no. But our producer Tory actually played it and

126
00:06:46,720 --> 00:06:50,280
Speaker 2: she talked to a character named Nathan, who was interested

127
00:06:50,279 --> 00:06:53,920
Speaker 2: in technology and had a Southern accent. But after their

128
00:06:53,960 --> 00:06:58,080
Speaker 2: conversation ended, Tinder told Tory that her replies were charming,

129
00:06:58,440 --> 00:07:01,200
Speaker 2: but that her conversation could have fload a little better

130
00:07:01,400 --> 00:07:04,760
Speaker 2: as she jumped between topics too quickly. Now, as a producer,

131
00:07:04,800 --> 00:07:05,640
Speaker 2: that's a great quality.

132
00:07:05,920 --> 00:07:08,440
Speaker 1: Yeah, exactly, short attention span. I think Tory may have

133
00:07:08,440 --> 00:07:11,680
Speaker 1: had it right and the and the app wrong, but

134
00:07:12,160 --> 00:07:14,280
Speaker 1: I would love to hear you try it.

135
00:07:15,240 --> 00:07:19,040
Speaker 2: My phone is ready and Tinder is downloaded. So let's

136
00:07:21,120 --> 00:07:27,320
Speaker 2: the game game. So what happens first is the prompt

137
00:07:27,360 --> 00:07:30,720
Speaker 2: comes up in the game. It's in the Tinder app.

138
00:07:30,920 --> 00:07:34,880
Speaker 1: Well, I mean, let's not pass over this soundtrack and

139
00:07:35,000 --> 00:07:35,800
Speaker 1: loving it.

140
00:07:35,800 --> 00:07:36,520
Speaker 2: It's fantastic.

141
00:07:36,440 --> 00:07:38,240
Speaker 1: This is like old New York.

142
00:07:39,240 --> 00:07:45,239
Speaker 2: We're going to Cafe Carlisle for a drink and a smoke. Okay,

143
00:07:45,240 --> 00:07:47,360
Speaker 2: I'm done, but let me read you the prompt so

144
00:07:47,400 --> 00:07:48,160
Speaker 2: we can start the game.

145
00:07:48,240 --> 00:07:48,560
Speaker 1: Okay.

146
00:07:50,120 --> 00:07:51,840
Speaker 2: And by the way, it lets you re roll, so

147
00:07:51,880 --> 00:07:53,960
Speaker 2: if you don't like the person you're introduced to, it

148
00:07:53,960 --> 00:07:55,960
Speaker 2: gives you another prompt, which I don't think should happen.

149
00:07:56,040 --> 00:07:59,200
Speaker 2: I think you should happen. But that's how Tinder works,

150
00:07:59,200 --> 00:08:01,520
Speaker 2: That's how chatchy be T four zero works. At a

151
00:08:01,520 --> 00:08:05,960
Speaker 2: bustling dog park, your curious dog eagerly runs up to Penelope,

152
00:08:06,320 --> 00:08:10,200
Speaker 2: a customer service representative, and starts playing with her dog.

153
00:08:10,560 --> 00:08:13,880
Speaker 1: How does your dog know what Penelope's job is? Oh,

154
00:08:13,920 --> 00:08:14,720
Speaker 1: you find that out later.

155
00:08:14,960 --> 00:08:18,320
Speaker 2: Hello, get my dog's like? Whoa, she's in customer service.

156
00:08:18,360 --> 00:08:22,360
Speaker 2: You should chase her down. No, and so my dog

157
00:08:22,600 --> 00:08:25,559
Speaker 2: goes up to Penelope's dog and starts playing with her dog.

158
00:08:26,000 --> 00:08:30,960
Speaker 2: This unexpected encounter creates the perfect excuse to strike up

159
00:08:31,000 --> 00:08:35,000
Speaker 2: a conversation with the one and only Penelope from Customer Service.

160
00:08:35,400 --> 00:08:39,480
Speaker 2: Let's go, I'm not rerolling, I'm taking my first card.

161
00:08:40,840 --> 00:08:44,080
Speaker 3: Well, hey, there looks like your dog's got quite a

162
00:08:44,120 --> 00:08:47,000
Speaker 3: bit of energy, don't they. My pup here is pretty

163
00:08:47,000 --> 00:08:49,920
Speaker 3: friendly too. Seems they're getting along just fine.

164
00:08:50,200 --> 00:08:58,360
Speaker 2: What's their name? My dog's name is Astrid Pastrid.

165
00:08:58,480 --> 00:09:02,400
Speaker 3: Huh, that's a fun name. Seems like an estrid. It's

166
00:09:02,400 --> 00:09:05,800
Speaker 3: a real social butterfly. Does she usually make friends this quick?

167
00:09:06,280 --> 00:09:09,000
Speaker 2: Honestly? No, your dog is the first dog she's ever

168
00:09:09,040 --> 00:09:14,280
Speaker 2: taken any interest in her? Well, ain't that something special?

169
00:09:14,679 --> 00:09:16,280
Speaker 3: Looks like they found the money today.

170
00:09:16,720 --> 00:09:17,880
Speaker 1: Must be a good match.

171
00:09:18,440 --> 00:09:21,319
Speaker 3: Maybe Astrid knows a kindred spirit when she sees one.

172
00:09:21,920 --> 00:09:24,800
Speaker 2: So how often do you come to the park. I've

173
00:09:24,840 --> 00:09:27,439
Speaker 2: tried to go to the dog park more now because

174
00:09:27,600 --> 00:09:30,600
Speaker 2: Astroid's gotten kind of crazy inside, so have I I mean,

175
00:09:30,600 --> 00:09:33,080
Speaker 2: post COVID it's really hard to get outside even now.

176
00:09:34,640 --> 00:09:37,120
Speaker 2: Why I hear you. It's been tough all around.

177
00:09:37,679 --> 00:09:40,440
Speaker 3: The fresh air and space can do wonders for both

178
00:09:40,480 --> 00:09:43,600
Speaker 3: you and Astro. Charleston's got a lot of beauty if

179
00:09:43,600 --> 00:09:44,800
Speaker 3: you take time to notice it.

180
00:09:45,440 --> 00:09:45,599
Speaker 1: You know.

181
00:09:45,720 --> 00:09:48,800
Speaker 2: Pela. Sorry, Penelte, I'm so sorry to cut you off.

182
00:09:49,640 --> 00:09:53,200
Speaker 3: No worries at all, I understand it. I don't want

183
00:09:53,200 --> 00:09:55,679
Speaker 3: to chat about the park or just need a friendly here.

184
00:09:56,040 --> 00:09:56,480
Speaker 3: I'm here.

185
00:09:57,040 --> 00:10:01,360
Speaker 2: Take care of yourself and Astrid. Penelope, would you be

186
00:10:01,400 --> 00:10:08,480
Speaker 2: interested in either giving me or getting Did she go away?

187
00:10:07,400 --> 00:10:17,920
Speaker 2: Oh you're still here, Penelope? Sorry, so Penelope, I want

188
00:10:18,000 --> 00:10:19,720
Speaker 2: another go But he just like girl.

189
00:10:19,600 --> 00:10:22,400
Speaker 1: You have no no. I like the way you dropped

190
00:10:22,400 --> 00:10:23,200
Speaker 1: your voice a little bit.

191
00:10:24,120 --> 00:10:26,040
Speaker 2: Wait, do you want to try it? We have to

192
00:10:26,120 --> 00:10:28,040
Speaker 2: have things to do on the show, but that's really

193
00:10:28,080 --> 00:10:31,920
Speaker 2: how it goes. CHATPG is kind of incredible. So guys,

194
00:10:32,760 --> 00:10:34,880
Speaker 2: that is the game game.

195
00:10:35,080 --> 00:10:37,800
Speaker 1: I love the game game. I'm also curious as to

196
00:10:37,880 --> 00:10:41,679
Speaker 1: why both your and Torri's match as a Southern accent is.

197
00:10:42,200 --> 00:10:46,600
Speaker 1: Is this a subtle kind of white loticification of society.

198
00:10:46,120 --> 00:10:49,120
Speaker 2: Where chat GPT four roh is just like, you know what,

199
00:10:49,160 --> 00:10:52,160
Speaker 2: they're getting Southern girls and that's about it, or southern boys.

200
00:10:52,160 --> 00:10:54,640
Speaker 2: In the case of Tory, Nathan was southern. I live

201
00:10:54,679 --> 00:10:56,200
Speaker 2: in New York and one of the things that I

202
00:10:56,240 --> 00:11:00,679
Speaker 2: saw come up and I was talking to Penelope was

203
00:11:01,520 --> 00:11:04,440
Speaker 2: plus plus empathetic. Now, if I was talking to New York,

204
00:11:04,440 --> 00:11:06,640
Speaker 2: I would have been like, yo, girl, what's up. I'm

205
00:11:06,640 --> 00:11:09,160
Speaker 2: talking to Penelope. I'm like, well, girl, would you like

206
00:11:09,240 --> 00:11:10,880
Speaker 2: to meet me at the park again?

207
00:11:12,200 --> 00:11:15,040
Speaker 1: I thought that you're lying about how Astroid had never

208
00:11:15,120 --> 00:11:17,719
Speaker 1: approached any other dogs. Was I mean that was you

209
00:11:17,760 --> 00:11:20,760
Speaker 1: gotta make them feel speak.

210
00:11:21,400 --> 00:11:25,000
Speaker 2: I hope this podcast never comes out. I love that

211
00:11:25,160 --> 00:11:27,200
Speaker 2: more than anything. I will be playing that all day

212
00:11:27,360 --> 00:11:28,800
Speaker 2: and I think I will, by the end of it

213
00:11:28,840 --> 00:11:32,880
Speaker 2: have a Southern accent, just to get a little bit

214
00:11:32,920 --> 00:11:34,839
Speaker 2: more serious about this story. You know, the Washington Post

215
00:11:35,040 --> 00:11:37,080
Speaker 2: reached out to the vice president of Product Growth and

216
00:11:37,120 --> 00:11:40,200
Speaker 2: Revenue at Tinder, and she said that the game is

217
00:11:40,280 --> 00:11:43,040
Speaker 2: meant to be silly and that the company quote leaned

218
00:11:43,080 --> 00:11:47,120
Speaker 2: into the campiness. Apparently, though, She went on to call

219
00:11:47,240 --> 00:11:50,440
Speaker 2: gen Z a socially anxious generation, and while the game

220
00:11:50,520 --> 00:11:53,319
Speaker 2: might be cringe, it's a generation that might look past

221
00:11:53,400 --> 00:11:56,400
Speaker 2: that if it indeed leads to a real connection.

222
00:11:56,960 --> 00:11:58,800
Speaker 1: I had to say, I mean, it was definitely fun

223
00:11:58,960 --> 00:12:02,680
Speaker 1: watching you play. I have never before myself, and I

224
00:12:02,679 --> 00:12:04,520
Speaker 1: didn't just time either, but I've never had a conversation

225
00:12:04,640 --> 00:12:06,920
Speaker 1: directly using my voice with an ali before. Was that

226
00:12:06,960 --> 00:12:08,040
Speaker 1: Was that a first for you or.

227
00:12:08,880 --> 00:12:11,400
Speaker 2: Only when I tried to scam my cousin? Actually?

228
00:12:11,480 --> 00:12:11,880
Speaker 1: Yeah, yeah?

229
00:12:12,040 --> 00:12:14,600
Speaker 2: So crazy? Is the pressure cooker that that just created

230
00:12:14,679 --> 00:12:16,840
Speaker 2: for me? Felt like there was literally a gun to

231
00:12:16,880 --> 00:12:18,359
Speaker 2: my head that was like flirt.

232
00:12:19,360 --> 00:12:21,880
Speaker 1: That's what it felt like. It's getting hot in here.

233
00:12:21,920 --> 00:12:24,520
Speaker 1: So we're going to take a quick break when we

234
00:12:24,559 --> 00:12:38,320
Speaker 1: come back some more headlines. Now to pivot back to

235
00:12:38,360 --> 00:12:41,640
Speaker 1: the headlines, We've got a few more for today, continuing

236
00:12:41,679 --> 00:12:45,600
Speaker 1: the theme of sex, deaths, and money. Well, no death, thankfully,

237
00:12:46,000 --> 00:12:48,520
Speaker 1: but we've had sex in the form of flirting now

238
00:12:48,559 --> 00:12:50,040
Speaker 1: for money taxes.

239
00:12:52,559 --> 00:12:55,120
Speaker 2: We know it's tax month, and one of the stories

240
00:12:55,160 --> 00:12:57,560
Speaker 2: has to do with two things you never want to

241
00:12:57,600 --> 00:13:02,600
Speaker 2: hear put together, which is IRS and hackathon. And of

242
00:13:02,679 --> 00:13:07,080
Speaker 2: course what does this start with the Department of Government

243
00:13:07,160 --> 00:13:11,200
Speaker 2: Efficiency is planning to stage a hackathon event. I sound

244
00:13:11,280 --> 00:13:14,160
Speaker 2: sad because I am is planning to stage a hackathon

245
00:13:14,200 --> 00:13:18,280
Speaker 2: event with the best engineers at the Internal Revenue Service.

246
00:13:18,600 --> 00:13:22,000
Speaker 2: According to Wired, DOE is planning to host dozens of

247
00:13:22,040 --> 00:13:24,679
Speaker 2: them in DC to build a mega API.

248
00:13:24,880 --> 00:13:28,400
Speaker 1: A mega API, that's actually what I read process.

249
00:13:28,679 --> 00:13:31,840
Speaker 2: It is a MEGAAPI essentially, which would make it easier

250
00:13:31,880 --> 00:13:36,199
Speaker 2: to access taxpayer data across different applications and cloud platforms.

251
00:13:36,240 --> 00:13:38,600
Speaker 1: We don't yet have a lot of details on the hackathon,

252
00:13:38,760 --> 00:13:41,040
Speaker 1: but I do hope they keep it tight because the

253
00:13:41,080 --> 00:13:45,200
Speaker 1: idea of highly sensitive tax data moving freely between what

254
00:13:45,240 --> 00:13:48,719
Speaker 1: maybe third party applications is a little frightening. There is

255
00:13:48,760 --> 00:13:52,200
Speaker 1: a broader controversy roiling the IRS. Several officials, including the

256
00:13:52,240 --> 00:13:56,360
Speaker 1: acting Commissioner, are quitting over the Trump administration's insistence that

257
00:13:56,480 --> 00:14:01,880
Speaker 1: the agency disclosed taxpayer information to Immigration and Customs Enforcement.

258
00:14:02,160 --> 00:14:05,719
Speaker 1: The IRIS has typically kept taxpay information confidential, even from

259
00:14:05,840 --> 00:14:10,920
Speaker 1: other government agencies, and that includes information submitted by undocumented immigrants.

260
00:14:11,360 --> 00:14:13,960
Speaker 1: But in a new agreement which appeared redacted in a

261
00:14:14,000 --> 00:14:17,480
Speaker 1: court filing, ICE officials can now ask the IRS for

262
00:14:17,520 --> 00:14:21,120
Speaker 1: information about people they're investigating or who've been ordered to

263
00:14:21,200 --> 00:14:22,280
Speaker 1: leave the US.

264
00:14:22,360 --> 00:14:24,760
Speaker 2: And in a story that takes us elsewhere into a

265
00:14:24,800 --> 00:14:27,360
Speaker 2: topic I am personally obsessed with, which is right to

266
00:14:27,400 --> 00:14:28,280
Speaker 2: repair laws?

267
00:14:28,480 --> 00:14:29,200
Speaker 1: What does that mean?

268
00:14:29,600 --> 00:14:32,160
Speaker 2: Right to repair law is basically like laws that say

269
00:14:32,160 --> 00:14:35,640
Speaker 2: that companies have to provide information to people who buy

270
00:14:35,680 --> 00:14:38,800
Speaker 2: things that teach them how to repair it, so one

271
00:14:38,880 --> 00:14:41,320
Speaker 2: you're not just buying new things every time they come out,

272
00:14:41,360 --> 00:14:44,640
Speaker 2: and two that you're able to actually know how to,

273
00:14:44,800 --> 00:14:47,360
Speaker 2: for example, repair a tractor. I was drawn to this

274
00:14:47,400 --> 00:14:50,840
Speaker 2: headline from The Verge with the perfect subhead quote, India's

275
00:14:50,920 --> 00:14:54,280
Speaker 2: repair culture gives new life to dead tech.

276
00:14:54,440 --> 00:14:56,400
Speaker 1: So we had sex, some money, and we do indeed have.

277
00:14:57,920 --> 00:15:00,640
Speaker 2: Oh dead technology, which is about the least sexy thing

278
00:15:00,640 --> 00:15:04,800
Speaker 2: on the planet. There's a rise of Frankenstein laptops in India. Now,

279
00:15:04,800 --> 00:15:06,800
Speaker 2: when I say Frankenstein laptops, what do you think?

280
00:15:07,320 --> 00:15:11,320
Speaker 1: Uh, gosh, I guess I think about laptops assembled from

281
00:15:11,320 --> 00:15:12,040
Speaker 1: all different parts.

282
00:15:12,400 --> 00:15:14,320
Speaker 2: I thought you were gonna say laptops with two bolts

283
00:15:14,360 --> 00:15:18,000
Speaker 2: on the side of that, But yes, they're basically resurrected

284
00:15:18,040 --> 00:15:21,080
Speaker 2: computers made with parts from trash older laptops and other

285
00:15:21,160 --> 00:15:24,480
Speaker 2: e waste ewtes, meaning trash that is of the electronic

286
00:15:24,520 --> 00:15:27,600
Speaker 2: variety at a fraction of the price. These laptops are

287
00:15:27,600 --> 00:15:31,400
Speaker 2: a good option for students, freelancers, or really anyone who

288
00:15:31,480 --> 00:15:34,000
Speaker 2: needs to be a part of India's growing digital economy

289
00:15:34,120 --> 00:15:36,640
Speaker 2: but may not be able to afford to participate otherwise.

290
00:15:36,680 --> 00:15:38,960
Speaker 1: Yeah, I think I read that you can basically get

291
00:15:39,080 --> 00:15:41,800
Speaker 1: a functional laptop from one of these one of these

292
00:15:41,800 --> 00:15:45,280
Speaker 1: Frankenstein laptops for around one hundred US dollars, which is

293
00:15:45,320 --> 00:15:47,880
Speaker 1: like an eighth of the price of any decent new laptop.

294
00:15:47,920 --> 00:15:49,120
Speaker 1: So it's pretty cool story.

295
00:15:49,440 --> 00:15:52,160
Speaker 2: But these Frankenstein tinkers don't have it so easy. There

296
00:15:52,200 --> 00:15:56,080
Speaker 2: are actually many global tech giants who restrict access to

297
00:15:56,120 --> 00:15:59,600
Speaker 2: spare parts or use proprietary hardware, which means people are

298
00:15:59,600 --> 00:16:03,440
Speaker 2: going through piles of sometimes toxic trash to get the parts,

299
00:16:03,640 --> 00:16:06,360
Speaker 2: and India's government is beginning to discuss right to repair

300
00:16:06,440 --> 00:16:09,120
Speaker 2: laws to address this, but progress has been slow.

301
00:16:09,840 --> 00:16:12,200
Speaker 1: Final story for this week is about a question I

302
00:16:12,320 --> 00:16:16,160
Speaker 1: find quite fascinating. What will be the iPhone of ai?

303
00:16:16,720 --> 00:16:19,000
Speaker 1: Will there be a kind of AI product that becomes

304
00:16:19,040 --> 00:16:21,880
Speaker 1: so ubiquitous that we forget what life was like before

305
00:16:21,880 --> 00:16:26,440
Speaker 1: it existed? Well, the iPhone designer himself, Johnny Ive or

306
00:16:26,640 --> 00:16:29,960
Speaker 1: Sir Johnny I've is working on it. Over a year ago,

307
00:16:30,000 --> 00:16:33,120
Speaker 1: he and Sam Altman, the CEO of Open Ai, began

308
00:16:33,160 --> 00:16:36,600
Speaker 1: discussing a device that might bring to life voice enabled

309
00:16:36,800 --> 00:16:41,280
Speaker 1: AI assistance, partly inspired by Altman's well documented fascination with

310
00:16:41,360 --> 00:16:44,920
Speaker 1: the movie Her. So. Alman and I have this startup together,

311
00:16:45,080 --> 00:16:48,600
Speaker 1: Ioproducts that's raised hundreds of millions of dollars and is

312
00:16:48,640 --> 00:16:51,880
Speaker 1: working on some device concepts, including a quote phone without

313
00:16:51,880 --> 00:16:55,280
Speaker 1: a screen, although some sources insist that it's in fact

314
00:16:55,480 --> 00:17:00,240
Speaker 1: not a phone, so the mystery remains. But this into

315
00:17:00,240 --> 00:17:02,480
Speaker 1: a story in the Information, which is also reporting that

316
00:17:02,560 --> 00:17:07,760
Speaker 1: Open Ai executives are considering acquiring a startup Ioproducts. This

317
00:17:07,800 --> 00:17:09,960
Speaker 1: would be a move that could potentially bring the AI

318
00:17:10,080 --> 00:17:13,960
Speaker 1: giant into more direct competition with Apple. It's not clear

319
00:17:14,080 --> 00:17:17,040
Speaker 1: where the negotiation is at the moment, but another of

320
00:17:17,080 --> 00:17:22,440
Speaker 1: these types of XAI X deals perhaps brewing, Although while

321
00:17:22,440 --> 00:17:24,520
Speaker 1: Altman worked close to you with ive on the project.

322
00:17:24,960 --> 00:17:26,760
Speaker 1: Is not clear what his economic stake in it.

323
00:17:26,800 --> 00:17:29,879
Speaker 2: Maybe if the new phone is not a phone. It

324
00:17:29,920 --> 00:17:31,960
Speaker 2: begs the question how the next thing that we cover

325
00:17:32,200 --> 00:17:35,960
Speaker 2: actually is going to happen in a no phone phone universe.

326
00:17:36,800 --> 00:17:40,159
Speaker 1: You're right, and our next segment is all about scammers,

327
00:17:40,280 --> 00:17:44,720
Speaker 1: and specifically scam callers who famously use phones and the

328
00:17:44,800 --> 00:17:48,400
Speaker 1: tech they're using to be more convincing and successful than ever.

329
00:17:48,960 --> 00:17:51,080
Speaker 2: Yeah, And one of the things that I can't stop

330
00:17:51,160 --> 00:17:53,120
Speaker 2: talking about on the show and talked a lot about

331
00:17:53,119 --> 00:17:58,000
Speaker 2: on Sleepwalkers is how much technological progress and innovation happens

332
00:17:58,040 --> 00:18:01,040
Speaker 2: in the sort of seedier parts of society. And then

333
00:18:01,080 --> 00:18:05,520
Speaker 2: it's after everyone hears these sensational stories about criminal ingenuity

334
00:18:05,680 --> 00:18:08,320
Speaker 2: that the tech is more widely adopted by the general public.

335
00:18:08,600 --> 00:18:11,480
Speaker 2: But it's actually the illicit use that forges the way.

336
00:18:11,600 --> 00:18:13,640
Speaker 1: Yeah. I remember, back in twenty nineteen, when we first

337
00:18:13,680 --> 00:18:16,239
Speaker 1: started covering this stuff together, there was a study that

338
00:18:16,280 --> 00:18:19,080
Speaker 1: revealed that more than ninety five percent of all deep

339
00:18:19,119 --> 00:18:22,400
Speaker 1: fake videos on the Internet were non consensual porn.

340
00:18:22,600 --> 00:18:25,000
Speaker 2: Well, I actually didn't even know that three D printing

341
00:18:25,200 --> 00:18:28,159
Speaker 2: was a consumer tech until I heard that blueprints for

342
00:18:28,240 --> 00:18:31,640
Speaker 2: three D printed ghost guns were circulating the internet.

343
00:18:32,119 --> 00:18:34,400
Speaker 1: Together. We actually ran an experiment a few years ago

344
00:18:34,600 --> 00:18:38,040
Speaker 1: to create a deep fake of your voice and scam

345
00:18:38,119 --> 00:18:40,439
Speaker 1: your cousin, and it took us about a week to

346
00:18:40,480 --> 00:18:42,680
Speaker 1: make that clone with the help of a company called

347
00:18:42,760 --> 00:18:46,240
Speaker 1: liar Bird that was subsequently acquired by Descript, the software

348
00:18:46,280 --> 00:18:49,159
Speaker 1: that we use every week to make our podcast. We

349
00:18:49,200 --> 00:18:51,200
Speaker 1: didn't actually get to the scamming apart, but we did

350
00:18:51,240 --> 00:18:54,600
Speaker 1: briefly trick Kara's cousin. That was back in twenty nineteen.

351
00:18:55,119 --> 00:18:57,640
Speaker 1: Since then, the state of the art and the kind

352
00:18:57,680 --> 00:19:01,840
Speaker 1: of social risks haveally advanced. And here to tell us

353
00:19:01,880 --> 00:19:04,720
Speaker 1: more is Julie Jargon, the family and tech columnist at

354
00:19:04,760 --> 00:19:06,960
Speaker 1: the Wall Street Journal. Julie, welcome to Tech Stuff.

355
00:19:07,440 --> 00:19:08,280
Speaker 4: Thank you for having me.

356
00:19:08,640 --> 00:19:12,359
Speaker 2: So, just to begin, your article tells the story of

357
00:19:12,400 --> 00:19:14,520
Speaker 2: a woman who gets a terrifying call. Can you tell

358
00:19:14,600 --> 00:19:18,760
Speaker 2: us a little bit more about what happened in this exchange?

359
00:19:18,960 --> 00:19:21,479
Speaker 4: Yeah, absolutely So. There was a woman in Colorado by

360
00:19:21,480 --> 00:19:23,720
Speaker 4: the name of Linda Rohan, and she was just at

361
00:19:23,760 --> 00:19:26,640
Speaker 4: home one night making herself dinner, and her phone rang

362
00:19:26,760 --> 00:19:29,199
Speaker 4: her cell phone and the caller ID showed that the

363
00:19:29,240 --> 00:19:30,960
Speaker 4: call was from a local number, so she thought it

364
00:19:31,040 --> 00:19:33,840
Speaker 4: might be someone she should talk to, so she picked

365
00:19:33,880 --> 00:19:36,280
Speaker 4: it up and immediately heard a voice of a young

366
00:19:36,280 --> 00:19:39,840
Speaker 4: woman that she thought sounded exactly like the youngest of

367
00:19:39,840 --> 00:19:44,280
Speaker 4: her three adult daughters, a panicked, you know message, Mom,

368
00:19:44,320 --> 00:19:46,879
Speaker 4: I'm okay, but something awful has happened and she's sobbing

369
00:19:46,920 --> 00:19:50,080
Speaker 4: and saying she needs help. And that immediately put this

370
00:19:50,119 --> 00:19:53,760
Speaker 4: woman on high alert. And then apparently a man took

371
00:19:53,800 --> 00:19:57,320
Speaker 4: the phone and mentioned the name of her daughter by

372
00:19:57,600 --> 00:20:00,000
Speaker 4: you know by name, and said that she had witnessed

373
00:20:00,200 --> 00:20:02,919
Speaker 4: this drug deal and she screamed and it scared the

374
00:20:02,920 --> 00:20:05,439
Speaker 4: buyers away, and so now he was out all this money,

375
00:20:05,480 --> 00:20:08,560
Speaker 4: and he had pulled this girl into his van and

376
00:20:08,720 --> 00:20:10,280
Speaker 4: now was demanding money.

377
00:20:10,480 --> 00:20:13,520
Speaker 2: We know that this wasn't Linda's real daughter. Where was

378
00:20:13,600 --> 00:20:15,800
Speaker 2: Linda's daughter actually during this.

379
00:20:15,880 --> 00:20:18,600
Speaker 4: She was in her apartment the whole time, safe at home.

380
00:20:18,800 --> 00:20:21,240
Speaker 2: And can you talk a little bit about how this happened.

381
00:20:21,640 --> 00:20:23,960
Speaker 4: I think what happens with these kind of callers is

382
00:20:24,040 --> 00:20:28,560
Speaker 4: they operate on fear and a sense of urgency. And

383
00:20:28,880 --> 00:20:33,199
Speaker 4: this scammer had an elaborate story that he kind of

384
00:20:33,640 --> 00:20:37,040
Speaker 4: kept this woman through this whole time. He told her

385
00:20:37,080 --> 00:20:39,560
Speaker 4: that he needed money in order to free her daughter.

386
00:20:40,040 --> 00:20:43,520
Speaker 4: He told her to go to Walmart and wire money.

387
00:20:43,560 --> 00:20:46,000
Speaker 4: And so she gets in her car and finds the

388
00:20:46,040 --> 00:20:49,520
Speaker 4: nearest Walmart, and he timed how long it took her

389
00:20:49,560 --> 00:20:51,800
Speaker 4: to get there. When she went to the Walmart, he

390
00:20:51,880 --> 00:20:53,600
Speaker 4: wanted to be on speaker the whole time, so he

391
00:20:53,680 --> 00:20:56,840
Speaker 4: had her conceal her phone in her shirt so he

392
00:20:56,880 --> 00:21:00,200
Speaker 4: could hear the conversation. And you know, I think he'd

393
00:21:00,200 --> 00:21:02,119
Speaker 4: made some kind of threats to her and you know,

394
00:21:02,240 --> 00:21:05,119
Speaker 4: her daughter. And when she got to the Walmart, she

395
00:21:05,160 --> 00:21:07,119
Speaker 4: couldn't do the wire transfer because she didn't have a

396
00:21:07,119 --> 00:21:09,240
Speaker 4: debit card. So he told her to go home and

397
00:21:09,480 --> 00:21:12,520
Speaker 4: do it online. And he said, you've got sixteen minutes.

398
00:21:12,840 --> 00:21:15,080
Speaker 4: If you stop anywhere, I'm going to know, because he

399
00:21:15,160 --> 00:21:16,800
Speaker 4: knew how long it had taken her to drive there

400
00:21:16,800 --> 00:21:18,040
Speaker 4: in the first place. And he kept her on the

401
00:21:18,080 --> 00:21:20,720
Speaker 4: phone this whole time, talking to her, trying to keep

402
00:21:20,720 --> 00:21:24,359
Speaker 4: her calm, asking her questions. This whole scenario played out

403
00:21:24,480 --> 00:21:27,800
Speaker 4: for a long time, and she made not one, but

404
00:21:27,880 --> 00:21:33,520
Speaker 4: two money transfers online in order to obtain her daughter's freedom.

405
00:21:34,080 --> 00:21:37,680
Speaker 4: And once it was finally over, she called her daughter

406
00:21:38,000 --> 00:21:40,480
Speaker 4: and found that her daughter was safe in her apartments.

407
00:21:40,800 --> 00:21:43,520
Speaker 1: It was such a striking story because it had this

408
00:21:43,640 --> 00:21:46,600
Speaker 1: kind of cinematic quality. I mean's actually like a movie.

409
00:21:46,680 --> 00:21:49,880
Speaker 1: The guy is playing a version of her daughter's voice,

410
00:21:50,320 --> 00:21:53,320
Speaker 1: making her literally drive from a to b, having her

411
00:21:53,440 --> 00:21:57,200
Speaker 1: conceal a phone in her clothes. I mean, there's fifteen

412
00:21:57,280 --> 00:21:59,840
Speaker 1: twenty thirty forty five minutes, all the while she thinks

413
00:21:59,840 --> 00:22:04,160
Speaker 1: that her daughter has been abducted by a drug dealer.

414
00:22:04,280 --> 00:22:05,800
Speaker 1: I mean, what did that do to the mother? And

415
00:22:05,800 --> 00:22:07,199
Speaker 1: when you were into viewing her and what does she

416
00:22:07,560 --> 00:22:08,760
Speaker 1: reflect about the experience?

417
00:22:09,480 --> 00:22:11,480
Speaker 4: Yeah, she described it as something that she can still

418
00:22:11,640 --> 00:22:15,560
Speaker 4: feel viscerally like. She retold the story to me three

419
00:22:15,600 --> 00:22:18,120
Speaker 4: times over the course of a few different conversations with her,

420
00:22:18,280 --> 00:22:20,480
Speaker 4: as I went through the story again and again with her,

421
00:22:20,920 --> 00:22:22,840
Speaker 4: and I could tell each time I talk to her

422
00:22:22,840 --> 00:22:24,560
Speaker 4: that she felt really nervous and worked up about it,

423
00:22:24,560 --> 00:22:26,520
Speaker 4: even though she knows it was all a scam, even

424
00:22:26,560 --> 00:22:28,880
Speaker 4: though she knows that her daughter was never in any

425
00:22:28,920 --> 00:22:33,000
Speaker 4: actual danger, but this whole ordeal was so terrifying to her.

426
00:22:33,600 --> 00:22:35,880
Speaker 4: And then that's of course why these scammers are so effective,

427
00:22:36,200 --> 00:22:38,920
Speaker 4: that they prey on the fear of people thinking that

428
00:22:38,960 --> 00:22:41,000
Speaker 4: they have a loved one, especially a child who might

429
00:22:41,000 --> 00:22:43,520
Speaker 4: be in some sort of danger. So, even though she's

430
00:22:43,720 --> 00:22:46,680
Speaker 4: now more than a month removed from the situation, still

431
00:22:46,680 --> 00:22:50,760
Speaker 4: in the retelling she feels very like physically nervous and scared.

432
00:22:50,960 --> 00:22:53,119
Speaker 1: Well, she's not surprising because the tension was kind of

433
00:22:53,160 --> 00:22:55,440
Speaker 1: resting up. And then just when she thought that she'd

434
00:22:55,440 --> 00:22:57,920
Speaker 1: made the payment everything was okay, there was a kind

435
00:22:57,920 --> 00:23:00,159
Speaker 1: of not the tone of the screw right right.

436
00:23:00,240 --> 00:23:01,679
Speaker 4: She thought it was kind of over. She'd made one

437
00:23:01,720 --> 00:23:05,119
Speaker 4: transfer of a thousand dollars, and then there was a

438
00:23:05,119 --> 00:23:08,720
Speaker 4: commotion and the man on the phone came back and said, well,

439
00:23:08,760 --> 00:23:10,439
Speaker 4: you know, my boss is angry that it took so

440
00:23:10,560 --> 00:23:13,919
Speaker 4: long to transfer this money, so we need more. My

441
00:23:13,960 --> 00:23:17,119
Speaker 4: boss is mad and he thinks he could sell your

442
00:23:17,200 --> 00:23:19,520
Speaker 4: daughter for thirty thousand dollars. And then at that point

443
00:23:19,560 --> 00:23:22,040
Speaker 4: she hears her daughter in the background screaming like no, no,

444
00:23:22,160 --> 00:23:25,960
Speaker 4: you know, please help me, and this woman Linda wanted

445
00:23:26,000 --> 00:23:28,280
Speaker 4: to talk to her daughter. She pleaded with this man

446
00:23:28,640 --> 00:23:30,639
Speaker 4: to let her talk to her daughter again, and he

447
00:23:30,680 --> 00:23:32,480
Speaker 4: said no, but you know, we can end this now

448
00:23:32,520 --> 00:23:35,679
Speaker 4: if he send another thousand dollars. So then she wired

449
00:23:36,240 --> 00:23:40,720
Speaker 4: another thousand dollars through a different wire service, and at

450
00:23:40,720 --> 00:23:42,359
Speaker 4: that point it was finally over.

451
00:23:43,200 --> 00:23:46,800
Speaker 1: It's this kind of incredible intersection of both a new technology,

452
00:23:46,920 --> 00:23:50,119
Speaker 1: I like the ubiquity of deep fake voices, and a

453
00:23:50,160 --> 00:23:53,440
Speaker 1: tremendously sophisticated psychological hack, right, I mean it has both

454
00:23:53,480 --> 00:23:54,640
Speaker 1: elements exactly.

455
00:23:54,640 --> 00:23:56,600
Speaker 4: And these kind of imposter scams have been going on

456
00:23:56,640 --> 00:23:58,679
Speaker 4: for a long time. I mean years ago, we'd be

457
00:23:58,760 --> 00:24:01,679
Speaker 4: hearing about grandpa parents getting calls from someone that was

458
00:24:01,760 --> 00:24:03,959
Speaker 4: claiming to be their grandson. But they usually didn't have

459
00:24:04,000 --> 00:24:06,639
Speaker 4: a name, they didn't have you know, the voice was

460
00:24:06,680 --> 00:24:09,600
Speaker 4: like just any young man, you know. And so it's

461
00:24:09,720 --> 00:24:14,040
Speaker 4: kind of using the same type of social engineering, but

462
00:24:14,200 --> 00:24:16,960
Speaker 4: ramped up in a more technological way that makes it

463
00:24:17,040 --> 00:24:18,080
Speaker 4: all the more believable.

464
00:24:24,359 --> 00:24:26,680
Speaker 2: When we come back, we'll hear about the way generative

465
00:24:26,720 --> 00:24:44,240
Speaker 2: AI makes imposter scams so convincing. Welcome back, So, Julie,

466
00:24:44,320 --> 00:24:47,119
Speaker 2: I have a lot of friends whose grandparents this has

467
00:24:47,160 --> 00:24:50,600
Speaker 2: happened to, and it preys on this sort of psychology

468
00:24:50,600 --> 00:24:54,000
Speaker 2: of oh my god, my grandchild is in trouble, let

469
00:24:54,080 --> 00:24:57,439
Speaker 2: me help them, without really thinking about how possible it

470
00:24:57,480 --> 00:24:59,920
Speaker 2: is that this is actually going on. This to your point,

471
00:25:00,119 --> 00:25:03,679
Speaker 2: is like an extremely ratcheted up version of this, and

472
00:25:03,720 --> 00:25:08,320
Speaker 2: it begs the question how exactly does something like this work?

473
00:25:08,640 --> 00:25:11,720
Speaker 2: How has it gotten so much more advanced? And I

474
00:25:11,760 --> 00:25:14,520
Speaker 2: think most importantly for this show tech stuff is like,

475
00:25:14,920 --> 00:25:18,639
Speaker 2: how were these people able to replicate Linda's daughter's voice?

476
00:25:19,680 --> 00:25:21,600
Speaker 4: Well, what we don't know here is whether they in

477
00:25:21,680 --> 00:25:25,840
Speaker 4: fact cloned her voice from some publicly available audio, you know,

478
00:25:25,880 --> 00:25:29,000
Speaker 4: whether her daughter had YouTube video out there or some

479
00:25:29,119 --> 00:25:31,639
Speaker 4: other type of audio or video that they could have

480
00:25:32,400 --> 00:25:34,679
Speaker 4: grabbed her voice from. She's twenty six, so chances are

481
00:25:34,720 --> 00:25:37,480
Speaker 4: she she could have. I didn't find any social media

482
00:25:37,520 --> 00:25:39,800
Speaker 4: accounts that I could access for her. But there are

483
00:25:39,800 --> 00:25:43,080
Speaker 4: other ways that you can approximate the sound of someone's voice.

484
00:25:43,200 --> 00:25:45,520
Speaker 4: There are a bunch of apps that are free or

485
00:25:45,720 --> 00:25:49,000
Speaker 4: very inexpensive on the different app stores that allow you

486
00:25:49,080 --> 00:25:51,920
Speaker 4: to change your voice. Fifty year old man could change

487
00:25:51,920 --> 00:25:54,879
Speaker 4: his voice to sound like a twenty year old woman,

488
00:25:55,119 --> 00:25:57,120
Speaker 4: you know, and you can change the dialect to the accent,

489
00:25:57,600 --> 00:26:00,639
Speaker 4: and those can be pretty convincing. And the experts I

490
00:26:00,680 --> 00:26:04,040
Speaker 4: talked to, both psychologists and cybersecurity experts, said that you

491
00:26:04,040 --> 00:26:07,400
Speaker 4: know when you're in this moment of fear, and you've

492
00:26:07,440 --> 00:26:10,439
Speaker 4: already gotten this idea in your mind that your daughter

493
00:26:10,600 --> 00:26:12,960
Speaker 4: is calling you. The first thing they're saying is mom,

494
00:26:13,600 --> 00:26:16,880
Speaker 4: your mind immediately switches to one of your children. And

495
00:26:17,000 --> 00:26:21,000
Speaker 4: so if they're able to approximate a voice of a

496
00:26:21,160 --> 00:26:25,240
Speaker 4: twenty year old woman, then your mind might immediately think

497
00:26:25,280 --> 00:26:28,040
Speaker 4: that that is your daughter, when it may not be

498
00:26:28,720 --> 00:26:30,800
Speaker 4: her actual voice or clone of her voice. So in

499
00:26:30,840 --> 00:26:34,239
Speaker 4: this case, we'll never know how they either got a

500
00:26:34,280 --> 00:26:36,640
Speaker 4: clone of her voice or whether they use some sort

501
00:26:36,680 --> 00:26:39,119
Speaker 4: of generative AI to create a voice that sounded like

502
00:26:39,160 --> 00:26:41,399
Speaker 4: it could be her daughter. And then a couple of

503
00:26:41,400 --> 00:26:44,320
Speaker 4: the tip offs here is that she wasn't able to

504
00:26:44,400 --> 00:26:47,560
Speaker 4: interact with the daughter. There were just these clips of

505
00:26:47,640 --> 00:26:51,120
Speaker 4: sound playing. She didn't have a conversation.

506
00:26:51,800 --> 00:26:53,879
Speaker 2: This is exactly what we did with my cousin and

507
00:26:53,920 --> 00:26:55,800
Speaker 2: what she had said to me that was so interesting

508
00:26:55,840 --> 00:26:58,840
Speaker 2: about this is the thing that tricked her was not

509
00:26:58,920 --> 00:27:01,679
Speaker 2: that I had this like incredible deep fake, but it

510
00:27:01,720 --> 00:27:04,800
Speaker 2: was the context. So she didn't really question the fact

511
00:27:04,920 --> 00:27:07,199
Speaker 2: that it was me, not because it really sounded like me,

512
00:27:07,560 --> 00:27:10,240
Speaker 2: but because of the context of our conversation. She called

513
00:27:10,240 --> 00:27:11,919
Speaker 2: me and I picked up, So why shouldn't she think

514
00:27:11,960 --> 00:27:14,160
Speaker 2: it's me right exactly.

515
00:27:14,359 --> 00:27:16,560
Speaker 1: So what is the scale of this problem?

516
00:27:16,920 --> 00:27:20,719
Speaker 4: It's really huge. The Federal Trade Commission said that the

517
00:27:21,280 --> 00:27:26,640
Speaker 4: number one category of fraud last year was imposter scams.

518
00:27:26,840 --> 00:27:29,359
Speaker 4: So that doesn't mean that they're all AI generated, but

519
00:27:29,960 --> 00:27:35,160
Speaker 4: scams in which people are calling, texting, emailing, whatever, impersonating

520
00:27:35,359 --> 00:27:38,840
Speaker 4: someone that someone knows with some sort of story and

521
00:27:38,880 --> 00:27:39,919
Speaker 4: a request for money.

522
00:27:40,760 --> 00:27:44,600
Speaker 2: This is something that is incredibly advanced for people who

523
00:27:44,640 --> 00:27:47,520
Speaker 2: we'd often call petty criminals. Does that mean that the

524
00:27:47,560 --> 00:27:51,920
Speaker 2: technology has become so ubiquitous that it's very accessible by

525
00:27:51,920 --> 00:27:54,359
Speaker 2: people we would call petty criminals. It's no longer the

526
00:27:54,359 --> 00:27:55,919
Speaker 2: thing of like, oh, I'm going to get on the

527
00:27:55,960 --> 00:27:58,600
Speaker 2: subway and pickpocket someone for the amount of money that

528
00:27:58,640 --> 00:28:00,320
Speaker 2: you might be able to get for this kind of scam.

529
00:28:00,400 --> 00:28:02,439
Speaker 2: So I mean not that I think you have a

530
00:28:02,480 --> 00:28:06,000
Speaker 2: criminal mind, Julie, but I'm wondering, from your perspective as

531
00:28:06,000 --> 00:28:08,360
Speaker 2: someone who's now reported on this, is this the kind

532
00:28:08,400 --> 00:28:12,240
Speaker 2: of crime that people who are looking to scam people

533
00:28:12,240 --> 00:28:15,000
Speaker 2: are engaging it? Is it because it's so easy?

534
00:28:15,080 --> 00:28:17,720
Speaker 4: Yeah, it has become a lot easier because of the

535
00:28:17,760 --> 00:28:21,040
Speaker 4: abiquity of these tools that can do voice clones or

536
00:28:21,600 --> 00:28:25,000
Speaker 4: AA generated voice approximations of people. All you have to

537
00:28:25,000 --> 00:28:28,879
Speaker 4: do is google it and you'll find dozens of online

538
00:28:28,880 --> 00:28:32,280
Speaker 4: tools that are either free or very very inexpensive. Or

539
00:28:32,359 --> 00:28:35,200
Speaker 4: go on the app store and download a voice changing app.

540
00:28:35,240 --> 00:28:39,440
Speaker 4: Though it's widely accessible, it's inexpensive, and all it takes

541
00:28:39,520 --> 00:28:42,920
Speaker 4: is one person who sends two thousand dollars and however

542
00:28:42,960 --> 00:28:45,680
Speaker 4: long this scenario went on, maybe thirty forty five minutes

543
00:28:45,760 --> 00:28:48,840
Speaker 4: or whatever, you know, they got two thousand dollars. So

544
00:28:49,240 --> 00:28:52,120
Speaker 4: you get a few victims over the course of some

545
00:28:52,360 --> 00:28:56,360
Speaker 4: period of time, and the payout can be pretty sizable.

546
00:28:57,080 --> 00:29:00,160
Speaker 1: Yeah, I think. I mean there's a financial cost to

547
00:29:00,200 --> 00:29:02,840
Speaker 1: your point about you know, the aftermath for Linda. There's

548
00:29:02,840 --> 00:29:06,280
Speaker 1: also this tremendous emotional costs, I mean, the trauma of it.

549
00:29:06,400 --> 00:29:09,360
Speaker 1: I saw a documentary the other day produced by Bloomberg

550
00:29:09,600 --> 00:29:13,960
Speaker 1: about young teens who are basically the victims of sextortion scam.

551
00:29:14,040 --> 00:29:16,160
Speaker 1: So somebody pretends to be somebody in their community and

552
00:29:16,200 --> 00:29:18,960
Speaker 1: gets them to send nude photos and maybe they're looking

553
00:29:19,000 --> 00:29:21,320
Speaker 1: to get I think hundreds of dollars, but in some

554
00:29:21,360 --> 00:29:24,520
Speaker 1: cases this pushes the teams to suicide and so it's

555
00:29:24,600 --> 00:29:26,479
Speaker 1: not like, yes, you have a sense of violation if

556
00:29:26,480 --> 00:29:28,880
Speaker 1: you get pick pocket on the subway, but this goes

557
00:29:28,920 --> 00:29:32,440
Speaker 1: to your core of your deepest fears, and I guess

558
00:29:32,480 --> 00:29:34,280
Speaker 1: maybe that's one of the reasons why your story went

559
00:29:34,440 --> 00:29:36,760
Speaker 1: so viral. But what can we do and what can

560
00:29:36,800 --> 00:29:38,840
Speaker 1: listeners do? What can readers do? What is the way

561
00:29:38,840 --> 00:29:41,120
Speaker 1: to make ourselves a bit more robust in the face

562
00:29:41,160 --> 00:29:41,360
Speaker 1: of this.

563
00:29:41,640 --> 00:29:43,440
Speaker 4: Well, I do have a column coming out this weekend

564
00:29:43,480 --> 00:29:45,040
Speaker 4: that will have tips, so I don't want to pre

565
00:29:45,120 --> 00:29:49,240
Speaker 4: empt to that, but there are things you can do, So.

566
00:29:49,240 --> 00:29:51,920
Speaker 1: Stay tuned, read all about it. Yeah, yeah, real about.

567
00:29:51,640 --> 00:29:53,720
Speaker 4: It when it comes out. But yeah, I mean, I

568
00:29:53,720 --> 00:29:56,360
Speaker 4: think just awareness, for one, is a major thing. And

569
00:29:56,400 --> 00:29:59,680
Speaker 4: I think that's why so many people responded to this,

570
00:29:59,720 --> 00:30:04,080
Speaker 4: Because you'd talk to anybody and someone knows someone to

571
00:30:04,120 --> 00:30:07,000
Speaker 4: whom this has happened or something you know very similar,

572
00:30:07,520 --> 00:30:10,600
Speaker 4: and that shows the scale of the problem. And I

573
00:30:10,640 --> 00:30:13,680
Speaker 4: think what's unfortunate is that people who are victimized by

574
00:30:13,680 --> 00:30:16,360
Speaker 4: these scams feel an incredible sense of shame and embarrassment

575
00:30:16,520 --> 00:30:19,040
Speaker 4: about it. You know, after their mind has calmed down,

576
00:30:19,120 --> 00:30:21,240
Speaker 4: they can easily go back and see the red flags

577
00:30:21,240 --> 00:30:24,040
Speaker 4: and they can you know, even when Linda was experiencing

578
00:30:24,040 --> 00:30:26,920
Speaker 4: this from the beginning, it cossed her mind this could

579
00:30:26,960 --> 00:30:29,280
Speaker 4: be a scam, but she felt like the stakes were

580
00:30:29,320 --> 00:30:32,360
Speaker 4: too high to just hang up and call her daughter

581
00:30:32,400 --> 00:30:34,080
Speaker 4: at that point, because she thought, you know, there was

582
00:30:34,080 --> 00:30:36,560
Speaker 4: that one kernel of like what if, what if my

583
00:30:36,640 --> 00:30:38,880
Speaker 4: daughter really has witnessed the drug deal and is in

584
00:30:38,920 --> 00:30:41,520
Speaker 4: the back of this person's van, and now her life

585
00:30:41,600 --> 00:30:42,280
Speaker 4: is in my hands.

586
00:30:42,400 --> 00:30:44,200
Speaker 2: And you don't want to be the mother that avoided

587
00:30:44,200 --> 00:30:47,400
Speaker 2: this because you think that you're being sort of techno savvy, right,

588
00:30:47,440 --> 00:30:49,120
Speaker 2: And all of a sudden, yeah, And that's when it

589
00:30:49,160 --> 00:30:52,600
Speaker 2: really speaks to the sort of core emotional piece of

590
00:30:52,640 --> 00:30:55,320
Speaker 2: these kind of scams. To Oz's point, it's not just pickpocketing.

591
00:30:55,640 --> 00:30:57,320
Speaker 2: You know, pickpocketing, you can say, oh, I should have

592
00:30:57,400 --> 00:30:59,880
Speaker 2: closed up my jacket better, But this is something that

593
00:31:00,120 --> 00:31:03,000
Speaker 2: is just so much more complicated than that. And also

594
00:31:03,080 --> 00:31:05,720
Speaker 2: I think has dual use. I just wanted to bring

595
00:31:05,760 --> 00:31:08,840
Speaker 2: that up quickly, like a lot of these technologies are

596
00:31:08,880 --> 00:31:12,120
Speaker 2: not just created for bad right, so it's not something

597
00:31:12,160 --> 00:31:14,920
Speaker 2: that can just be kind of wiped out. Deep fake

598
00:31:14,960 --> 00:31:19,880
Speaker 2: technology also has some really interesting applications that I think

599
00:31:19,960 --> 00:31:23,080
Speaker 2: we all benefit from. Now, so it becomes that sort

600
00:31:23,120 --> 00:31:26,040
Speaker 2: of complicated intersection of like some people are using this

601
00:31:26,480 --> 00:31:29,800
Speaker 2: to take advantage of people and scam them, and other

602
00:31:29,800 --> 00:31:34,080
Speaker 2: people are using it to make some really interesting practical applications.

603
00:31:34,120 --> 00:31:36,800
Speaker 2: So I don't know, I'm curious to see your column,

604
00:31:36,800 --> 00:31:41,080
Speaker 2: but it's definitely less simple than just get rid of

605
00:31:41,120 --> 00:31:41,920
Speaker 2: this technology.

606
00:31:42,120 --> 00:31:44,280
Speaker 4: Well, it's not going away, that's for sure. There are

607
00:31:44,680 --> 00:31:48,479
Speaker 4: obviously good uses of generative AI and it's definitely here

608
00:31:48,520 --> 00:31:51,160
Speaker 4: to stay. And my worry is as it gets better

609
00:31:51,200 --> 00:31:54,040
Speaker 4: and better, especially with video, I just wonder at some

610
00:31:54,160 --> 00:31:58,040
Speaker 4: point will people be able to receive faith time calls,

611
00:31:58,120 --> 00:32:02,520
Speaker 4: right video calls where they feel like they're seeing interacting

612
00:32:02,600 --> 00:32:05,920
Speaker 4: with someone who looks just like their child. Now, how

613
00:32:07,040 --> 00:32:08,720
Speaker 4: do you then tell that that's not real?

614
00:32:09,880 --> 00:32:11,400
Speaker 1: Julie, Thank you so much for your time today.

615
00:32:11,560 --> 00:32:14,240
Speaker 2: Thank you, Julie. We'll look forward to that column this Saturday.

616
00:32:14,480 --> 00:32:15,560
Speaker 4: Yeah, thank you for having me.

617
00:32:22,680 --> 00:32:24,280
Speaker 2: That's it for this week for TEXTA.

618
00:32:24,560 --> 00:32:27,800
Speaker 1: I'm Kara Price and I'm mos Vloschen. This episode was

619
00:32:27,800 --> 00:32:32,240
Speaker 1: produced by Eliza Dennis, Victoria Dominguez, and Adriana Tapia. It

620
00:32:32,280 --> 00:32:35,480
Speaker 1: was executive produced by me, Kara Price and Kate Osborne

621
00:32:35,560 --> 00:32:40,520
Speaker 1: Kaleidoscope and Katria Novelfi Hot Podcasts. The engineer is Bihit

622
00:32:40,560 --> 00:32:43,880
Speaker 1: Fraser and Kyle Murdoll makes this episode and he also

623
00:32:43,880 --> 00:32:44,640
Speaker 1: wrote our theme song.

624
00:32:44,880 --> 00:32:47,600
Speaker 2: Join us next Wednesday for Textuff the Story, when we

625
00:32:47,640 --> 00:32:50,880
Speaker 2: will share an in depth conversation with Jenstatsky, creator and

626
00:32:50,920 --> 00:32:54,000
Speaker 2: writer of the hit HBO Max show hacks Well, chat

627
00:32:54,000 --> 00:32:56,280
Speaker 2: about if AI is coming for her job and what

628
00:32:56,320 --> 00:32:58,800
Speaker 2: it's like to make TV. Knowing you're likely battling for

629
00:32:58,840 --> 00:32:59,760
Speaker 2: attention with a.

630
00:32:59,680 --> 00:33:03,440
Speaker 1: Second, please rate, review, and reach out to us at

631
00:33:03,480 --> 00:33:06,720
Speaker 1: tech Stuff podcast at gmail dot com. We love hearing

632
00:33:06,760 --> 00:33:07,120
Speaker 1: from you