1
00:00:00,040 --> 00:00:02,680
Speaker 1: Thanks for tuney into tech Stuff. If you don't recognize

2
00:00:02,680 --> 00:00:05,440
Speaker 1: my voice, my name is Oz Valoshian and I'm here

3
00:00:05,480 --> 00:00:08,639
Speaker 1: because the inimitable Jonathan Strickland has passed the baton to

4
00:00:08,720 --> 00:00:12,160
Speaker 1: Cara Price and myself to host Tech Stuff. The show

5
00:00:12,160 --> 00:00:15,040
Speaker 1: will remain your home for all things tech, and all

6
00:00:15,080 --> 00:00:18,400
Speaker 1: the old episodes will remain available in this feed. Thanks

7
00:00:18,400 --> 00:00:22,200
Speaker 1: for listening. Welcome to Tech Stuff. I'm Os Valoshian and

8
00:00:22,239 --> 00:00:23,080
Speaker 1: I'm Cara Price.

9
00:00:23,239 --> 00:00:26,960
Speaker 2: And this is the story our weekly episode, dropping each Wednesday,

10
00:00:27,160 --> 00:00:30,160
Speaker 2: sharing an in depth conversation with a fascinating person in

11
00:00:30,160 --> 00:00:33,159
Speaker 2: and around tech. This week, Oz tells me he's bringing

12
00:00:33,200 --> 00:00:35,400
Speaker 2: me a conversation with Meredith Whittaker.

13
00:00:37,760 --> 00:00:41,800
Speaker 1: That's right. Meredith leads the Signal Foundation, which is a

14
00:00:41,800 --> 00:00:47,360
Speaker 1: nonprofit that oversees the encrypted messaging app Signal, beloved by journalists,

15
00:00:47,360 --> 00:00:50,320
Speaker 1: deal makers, drug dealers, and privacy buffs alike.

16
00:00:50,720 --> 00:00:54,320
Speaker 2: It takes wanting to keep a really really really big

17
00:00:54,360 --> 00:00:57,080
Speaker 2: secret to get on Signal, but we should also think

18
00:00:57,080 --> 00:00:59,640
Speaker 2: about everything we say as a really big secret. If

19
00:00:59,640 --> 00:01:02,600
Speaker 2: you can who has access to all of our information.

20
00:01:02,280 --> 00:01:04,679
Speaker 1: One hundred percent, Well, you don't sometimes know what's a

21
00:01:04,720 --> 00:01:08,400
Speaker 1: really really big secret until it's too late. Meredith has

22
00:01:08,480 --> 00:01:11,640
Speaker 1: a fascinating background. She started at Google back in the

23
00:01:11,720 --> 00:01:14,640
Speaker 1: day when it famously boasted that don't be Evil slogan,

24
00:01:15,000 --> 00:01:17,320
Speaker 1: and she actually started as a temp in a kind

25
00:01:17,360 --> 00:01:20,880
Speaker 1: of customer service adjacent gig that was despite being a

26
00:01:20,920 --> 00:01:24,240
Speaker 1: Berkeley grad in literature and rhetoric. Eventually she worked her

27
00:01:24,280 --> 00:01:28,280
Speaker 1: way up into leadership, founding Google Open Research Group and

28
00:01:28,319 --> 00:01:32,640
Speaker 1: then separately founding something called the AI Now Institute at NYU.

29
00:01:32,880 --> 00:01:35,480
Speaker 1: And after she left Google, she became head of Signal

30
00:01:35,480 --> 00:01:39,000
Speaker 1: in twenty twenty two. But the circumstances of her departure

31
00:01:39,000 --> 00:01:40,880
Speaker 1: from Google were memorable, to say the least.

32
00:01:41,000 --> 00:01:42,399
Speaker 2: I think it's fair to say that she got on

33
00:01:42,440 --> 00:01:45,720
Speaker 2: everyone's radar back in twenty eighteen when she led the

34
00:01:45,760 --> 00:01:48,640
Speaker 2: walkouts at Google over the experience of women in tech

35
00:01:48,680 --> 00:01:52,760
Speaker 2: and at Google specifically over things like pay inequality and

36
00:01:52,760 --> 00:01:55,120
Speaker 2: the handling of sexual harassment at the company.

37
00:01:55,400 --> 00:01:59,640
Speaker 1: And then there's also Project Maven, which was Google's decision

38
00:01:59,640 --> 00:02:02,800
Speaker 1: to take part in a Department of Defense program to

39
00:02:03,000 --> 00:02:07,960
Speaker 1: use machine learning and data for battlefield target identification, which

40
00:02:08,040 --> 00:02:11,120
Speaker 1: was very very controversial internally and which in part is

41
00:02:11,120 --> 00:02:14,040
Speaker 1: what led her to her role today at signal. Merediths

42
00:02:14,080 --> 00:02:16,440
Speaker 1: won a prize in Germany last year called the Helmut

43
00:02:16,440 --> 00:02:19,280
Speaker 1: Schmidt Future Award, where she was honored to quote her

44
00:02:19,280 --> 00:02:22,960
Speaker 1: commitment to the development of AI technology oriented towards the

45
00:02:23,000 --> 00:02:25,639
Speaker 1: common good. So we talk about that too, and why

46
00:02:25,720 --> 00:02:29,640
Speaker 1: Meredith fundamentally describes herself as an optimist. But we start

47
00:02:29,639 --> 00:02:31,119
Speaker 1: by geeking out over literature.

48
00:02:31,560 --> 00:02:32,160
Speaker 2: Let's hear it.

49
00:02:33,040 --> 00:02:35,200
Speaker 1: So I was reading your bio and one of the

50
00:02:35,240 --> 00:02:37,959
Speaker 1: things that stuck out to me was your background as

51
00:02:38,000 --> 00:02:41,440
Speaker 1: a English literature and rhetoric major in college. I was

52
00:02:41,480 --> 00:02:44,280
Speaker 1: also in English major. As I looked at your kind

53
00:02:44,320 --> 00:02:49,040
Speaker 1: of story, there's this arc from arriving at Google as

54
00:02:49,040 --> 00:02:51,760
Speaker 1: a graduate in two thousand and six in the Don't

55
00:02:51,800 --> 00:02:55,280
Speaker 1: Be Evil error, seeing all of these technologies developed, many

56
00:02:55,280 --> 00:02:57,480
Speaker 1: of which went on to be kind of part of

57
00:02:57,520 --> 00:03:00,640
Speaker 1: the foundation of what we now call AI, leading a

58
00:03:00,680 --> 00:03:05,120
Speaker 1: protest movement and ultimately leaving and it made me think

59
00:03:05,160 --> 00:03:08,359
Speaker 1: of Pilgrim's Progress or a buildings Room.

60
00:03:08,400 --> 00:03:11,959
Speaker 3: Then yeah, I mean, maybe a Jeremiah is more accurate.

61
00:03:14,800 --> 00:03:17,840
Speaker 4: So that was a joke for the four English literature

62
00:03:18,120 --> 00:03:22,400
Speaker 4: majors who are listening. I think for me, I was

63
00:03:22,600 --> 00:03:26,440
Speaker 4: very sincere. I still am very sincere, and I came

64
00:03:26,480 --> 00:03:28,320
Speaker 4: in and took a lot of things at face value,

65
00:03:28,400 --> 00:03:32,000
Speaker 4: in part because I had no idea about tech. I

66
00:03:32,040 --> 00:03:35,240
Speaker 4: was stepping in and I was asking very basic questions.

67
00:03:34,960 --> 00:03:36,320
Speaker 1: You did in two thousand and six.

68
00:03:36,400 --> 00:03:39,720
Speaker 4: I mean, well, you know, it wasn't the money occupation

69
00:03:40,160 --> 00:03:43,160
Speaker 4: when I got into Google. It was it was actually,

70
00:03:43,240 --> 00:03:46,240
Speaker 4: it was a fairly lovely environment in some ways because

71
00:03:46,280 --> 00:03:50,520
Speaker 4: a lot of the people there were people who had

72
00:03:50,520 --> 00:03:53,800
Speaker 4: a special interest, like nerdy people like myself, but I

73
00:03:53,920 --> 00:03:56,320
Speaker 4: was a book nerd, and they were sort of math

74
00:03:56,400 --> 00:04:01,880
Speaker 4: and science nerds. And suddenly their special interest, their sweet

75
00:04:01,920 --> 00:04:04,960
Speaker 4: passion that they spent so much time thinking about, was

76
00:04:05,000 --> 00:04:09,000
Speaker 4: also profitable. It was also important for the world. It

77
00:04:09,040 --> 00:04:12,400
Speaker 4: put them in a position of being a meaningful contributor.

78
00:04:12,800 --> 00:04:15,800
Speaker 4: And there was a genuineness to that. And we're in

79
00:04:15,840 --> 00:04:18,840
Speaker 4: an era now, of course, where you know, no one's

80
00:04:18,920 --> 00:04:21,000
Speaker 4: mom is saying be a doctor or a lawyer. They're

81
00:04:21,040 --> 00:04:24,080
Speaker 4: saying be an engineer. And so it was, you know,

82
00:04:24,240 --> 00:04:26,799
Speaker 4: it was simply a very different world.

83
00:04:26,800 --> 00:04:29,000
Speaker 1: And what about don't be evil? When was that generation?

84
00:04:29,080 --> 00:04:29,200
Speaker 3: Then?

85
00:04:29,200 --> 00:04:31,040
Speaker 1: When was that abandoned? Was that part of the kind

86
00:04:31,080 --> 00:04:33,360
Speaker 1: of discourse when you went to Google, that.

87
00:04:33,440 --> 00:04:35,760
Speaker 4: Was generated before my time, and it was core to

88
00:04:35,800 --> 00:04:39,640
Speaker 4: the discourse at Google. I think it was removed by

89
00:04:39,800 --> 00:04:43,200
Speaker 4: lawyers in like twenty fifteen, twenty sixteen, I don't actually

90
00:04:43,240 --> 00:04:46,400
Speaker 4: remember the date, but look, it's a slogan.

91
00:04:47,440 --> 00:04:47,640
Speaker 3: You know.

92
00:04:47,680 --> 00:04:51,440
Speaker 4: The structure of Google was still up into the right

93
00:04:51,560 --> 00:04:55,960
Speaker 4: forever so profits and growth worthy objective functions. That's the

94
00:04:56,000 --> 00:05:02,200
Speaker 4: sort of perpetual motion machine that Google adhere. However, I

95
00:05:02,279 --> 00:05:06,359
Speaker 4: do think don't Be Evil had some power in that

96
00:05:06,520 --> 00:05:10,000
Speaker 4: it was a touchstone that we all shared where socially

97
00:05:10,040 --> 00:05:12,880
Speaker 4: awkward people are people who might not quite understand how

98
00:05:12,920 --> 00:05:15,400
Speaker 4: to formulate a discomfort could kind of point to it

99
00:05:15,440 --> 00:05:17,720
Speaker 4: and say like, hey, this is making me think that

100
00:05:17,760 --> 00:05:20,480
Speaker 4: we should examine this because we said don't be evil.

101
00:05:21,040 --> 00:05:23,480
Speaker 4: And most of this, you know, throughout Google's history was

102
00:05:23,520 --> 00:05:27,479
Speaker 4: at a time where let's say, the horizons that would

103
00:05:27,600 --> 00:05:31,760
Speaker 4: challenge that were very far in the distance. So you

104
00:05:31,760 --> 00:05:34,960
Speaker 4: could still make a lot of quarterly returns doing a

105
00:05:35,000 --> 00:05:38,320
Speaker 4: lot of things that were understood and we can debate

106
00:05:38,360 --> 00:05:42,800
Speaker 4: that as as good right, and things like building a

107
00:05:42,920 --> 00:05:46,560
Speaker 4: search engine for the Chinese market that would include surveillance

108
00:05:46,600 --> 00:05:49,200
Speaker 4: functions for the government was not something you even had

109
00:05:49,200 --> 00:05:51,880
Speaker 4: to consider because there was so much low hanging fruit

110
00:05:51,960 --> 00:05:55,080
Speaker 4: in terms of continuing to grow, continuing to profit. And

111
00:05:55,120 --> 00:05:57,480
Speaker 4: so I think that combination of like it was, it

112
00:05:57,520 --> 00:06:00,760
Speaker 4: was sincerely held, and that the actual temptation to do

113
00:06:00,839 --> 00:06:04,239
Speaker 4: the things that really got into that sort of hot

114
00:06:04,279 --> 00:06:08,200
Speaker 4: water potential evil were not on the table at that

115
00:06:08,240 --> 00:06:10,400
Speaker 4: point because there was so much else that could be done.

116
00:06:10,440 --> 00:06:12,920
Speaker 4: And you know, of course, the ratchet turns, the ratchet turns,

117
00:06:12,960 --> 00:06:16,520
Speaker 4: the ratchet turns, and then suddenly those horizons are much closer.

118
00:06:17,360 --> 00:06:23,880
Speaker 4: The company is swollen by orders of magnitude, and that

119
00:06:24,080 --> 00:06:28,880
Speaker 4: little slogan is quietly removed by a team of lawyers

120
00:06:29,880 --> 00:06:32,760
Speaker 4: who probably bought some summer houses with the fees.

121
00:06:34,000 --> 00:06:36,800
Speaker 1: Did you believe that slogan? And was there a moment

122
00:06:36,920 --> 00:06:40,120
Speaker 1: along the way where you definitively started not to believe it.

123
00:06:42,360 --> 00:06:45,200
Speaker 4: I liked having that slogan as a tool. Let me

124
00:06:45,240 --> 00:06:46,200
Speaker 4: say that, right.

125
00:06:46,560 --> 00:06:49,400
Speaker 3: It was useful. It did work in rooms.

126
00:06:49,480 --> 00:06:53,000
Speaker 4: It you know, allowed us to frame debates in a

127
00:06:53,040 --> 00:06:58,040
Speaker 4: way that had at at least as their reference port

128
00:06:58,560 --> 00:07:04,840
Speaker 4: a core and ethical consideration. Like I was young, there

129
00:07:04,880 --> 00:07:07,920
Speaker 4: was about a million different learning curves I was scaling

130
00:07:07,920 --> 00:07:10,040
Speaker 4: at once. Yes, what does it mean to have a

131
00:07:10,080 --> 00:07:10,600
Speaker 4: real job.

132
00:07:10,680 --> 00:07:10,880
Speaker 2: Right.

133
00:07:11,160 --> 00:07:12,680
Speaker 3: I didn't come from this class.

134
00:07:13,160 --> 00:07:16,440
Speaker 4: I didn't have familiarity with these cultures or environments because

135
00:07:16,480 --> 00:07:19,840
Speaker 4: I'd literally never been in an environment that looked anything

136
00:07:20,000 --> 00:07:23,640
Speaker 4: like you know what my family still jokingly refers to

137
00:07:23,720 --> 00:07:24,640
Speaker 4: as a business job.

138
00:07:24,920 --> 00:07:25,120
Speaker 3: Right.

139
00:07:25,800 --> 00:07:28,320
Speaker 4: I didn't know what was normal and what was not normal.

140
00:07:28,400 --> 00:07:30,880
Speaker 4: So for me, Google was normal, and so I just

141
00:07:30,920 --> 00:07:32,280
Speaker 4: want to be care Like. I don't know if I

142
00:07:32,400 --> 00:07:34,800
Speaker 4: believed it or not. I found it useful and I

143
00:07:34,800 --> 00:07:39,440
Speaker 4: saw it do work in the world, and I liked

144
00:07:39,520 --> 00:07:42,200
Speaker 4: the idea, Like, it sounds great that we're giving people

145
00:07:42,280 --> 00:07:44,760
Speaker 4: ads that match their interests. What could be wrong with

146
00:07:44,800 --> 00:07:48,600
Speaker 4: that We're organizing the world's information. Wow, all of this

147
00:07:48,680 --> 00:07:49,120
Speaker 4: is cool.

148
00:07:49,360 --> 00:07:53,160
Speaker 3: Like I was like, Okay, what is tech? What is Google? Hey,

149
00:07:53,240 --> 00:07:56,320
Speaker 3: sit me down and tell me what is information? I

150
00:07:56,360 --> 00:07:58,640
Speaker 3: was a master at asking the dumbest questions, but sometimes

151
00:07:58,680 --> 00:08:01,200
Speaker 3: they would unlock whole world for me, and then I

152
00:08:01,240 --> 00:08:05,160
Speaker 3: would realize, like, oh, information is just websites. Okay, but

153
00:08:05,240 --> 00:08:07,440
Speaker 3: you can't make a website if you don't have a computer,

154
00:08:09,040 --> 00:08:12,480
Speaker 3: and you have to be able to code HTML markup

155
00:08:12,520 --> 00:08:16,040
Speaker 3: to do that. Okay, So it's more limited than I thought. Okay,

156
00:08:16,080 --> 00:08:18,040
Speaker 3: well what is an index? Oh, it's ranking it. But

157
00:08:18,200 --> 00:08:19,600
Speaker 3: isn't that a value judgment?

158
00:08:20,040 --> 00:08:20,160
Speaker 2: Well?

159
00:08:20,200 --> 00:08:21,960
Speaker 4: I guess it's not if it's an algorithm, but like

160
00:08:22,000 --> 00:08:24,800
Speaker 4: who makes the algorithm? And so it was butting my

161
00:08:24,880 --> 00:08:30,040
Speaker 4: head up against what are kind of dumb, very basic

162
00:08:30,200 --> 00:08:34,640
Speaker 4: questions and sort of sensing an insufficiency of the responses,

163
00:08:34,679 --> 00:08:39,319
Speaker 4: and like, you know, over almost twenty years now, continuing

164
00:08:39,360 --> 00:08:41,760
Speaker 4: that process of trying to, you know, just figure out

165
00:08:41,920 --> 00:08:43,360
Speaker 4: like what are we doing here?

166
00:08:44,600 --> 00:08:48,520
Speaker 3: And how do I understand it? And that was basically

167
00:08:48,559 --> 00:08:49,240
Speaker 3: my education.

168
00:08:49,760 --> 00:08:54,000
Speaker 1: So fust forward, you know, thirteen years from two thousand

169
00:08:54,000 --> 00:08:56,440
Speaker 1: and six to twenty nineteen. How did you get to

170
00:08:56,480 --> 00:09:00,720
Speaker 1: that point where it became impossible for you to stay?

171
00:09:01,559 --> 00:09:03,839
Speaker 3: I don't have a pat answer, But how did I

172
00:09:03,880 --> 00:09:06,320
Speaker 3: get to that point? Well?

173
00:09:06,520 --> 00:09:10,360
Speaker 4: Again, I was a sincere person and I had done

174
00:09:10,360 --> 00:09:11,720
Speaker 4: a lot in my career. I had a lot of

175
00:09:11,720 --> 00:09:14,600
Speaker 4: opportunities and then I took them, and you know, I

176
00:09:14,600 --> 00:09:16,640
Speaker 4: built a research group inside of Google.

177
00:09:17,400 --> 00:09:18,840
Speaker 3: I was involved in.

178
00:09:18,800 --> 00:09:24,800
Speaker 4: Efforts to spread privacy and security technologies across the tech ecosystem.

179
00:09:24,880 --> 00:09:28,040
Speaker 4: I was a champion for privacy within Google. I had

180
00:09:28,040 --> 00:09:32,360
Speaker 4: built you know, open source measurement infrastructures to help inform

181
00:09:32,400 --> 00:09:35,920
Speaker 4: the net neutrality debate. I had always tried to be

182
00:09:36,040 --> 00:09:41,040
Speaker 4: on the side of doing social good, to.

183
00:09:40,559 --> 00:09:42,240
Speaker 3: Be very bland about it.

184
00:09:42,520 --> 00:09:46,160
Speaker 4: Yeah, And I had been able to do very very

185
00:09:46,200 --> 00:09:50,840
Speaker 4: well doing that. And at the point where I became frustrated,

186
00:09:51,679 --> 00:09:57,080
Speaker 4: I had already established the AI Now Institute, co founding

187
00:09:57,160 --> 00:10:01,800
Speaker 4: that at NYU, looking at some of the present day

188
00:10:02,760 --> 00:10:06,520
Speaker 4: real term harms and social implications of AI technologies. This

189
00:10:06,720 --> 00:10:12,200
Speaker 4: was twenty sixteen. I had become known through the academic

190
00:10:12,280 --> 00:10:15,720
Speaker 4: world and the technical world, and particularly within Google itself

191
00:10:16,360 --> 00:10:19,920
Speaker 4: as an authority and a speaker and an expert on

192
00:10:20,000 --> 00:10:23,680
Speaker 4: those issues. So talking about some of the harms that

193
00:10:24,120 --> 00:10:30,160
Speaker 4: these technologies could cause, criticizing within, essentially criticizing within. I

194
00:10:30,200 --> 00:10:33,080
Speaker 4: saw myself as a resource, and actually many many people did.

195
00:10:33,080 --> 00:10:35,600
Speaker 4: I would be brought in when teams were struggling with things,

196
00:10:35,640 --> 00:10:39,119
Speaker 4: I would advise them. I would often give academic lectures

197
00:10:39,640 --> 00:10:42,440
Speaker 4: that went against the party line at Google, but we're

198
00:10:42,920 --> 00:10:47,960
Speaker 4: very well cited, were very empirically documented, and I had

199
00:10:48,080 --> 00:10:51,640
Speaker 4: felt that I was making some headway right.

200
00:10:51,679 --> 00:10:54,040
Speaker 3: People were listening to me. I was at the table,

201
00:10:54,559 --> 00:10:55,240
Speaker 3: so to speak.

202
00:10:55,400 --> 00:11:00,240
Speaker 4: And then in twenty seventeen Fall twenty seventeen, and I

203
00:11:00,320 --> 00:11:04,520
Speaker 4: learned about a military contract that Google had secretively signed

204
00:11:04,520 --> 00:11:08,480
Speaker 4: to build for the US drone program. And that was

205
00:11:08,760 --> 00:11:11,920
Speaker 4: at a time and still today, when those systems were

206
00:11:11,960 --> 00:11:14,800
Speaker 4: not purpose built right, they were not safe.

207
00:11:14,840 --> 00:11:18,440
Speaker 3: There were ethical concerns we had. Sergey Brin, the co

208
00:11:18,520 --> 00:11:20,800
Speaker 3: founder of Google, had made.

209
00:11:20,640 --> 00:11:23,120
Speaker 4: Remarks that were very clear in the past about keeping

210
00:11:23,120 --> 00:11:26,840
Speaker 4: away from military contracting. Given Google's business model, it's public interest.

211
00:11:26,920 --> 00:11:28,199
Speaker 4: It serves the world, not just.

212
00:11:28,160 --> 00:11:30,360
Speaker 1: The US business model, meaning that it knows a lot

213
00:11:30,400 --> 00:11:32,960
Speaker 1: about people. Essentially, yeah that it is.

214
00:11:33,040 --> 00:11:36,160
Speaker 4: Ultimately it's a surveillance company. It collects huge amounts of

215
00:11:36,280 --> 00:11:39,760
Speaker 4: data about people, and then it creates models from that

216
00:11:39,880 --> 00:11:45,119
Speaker 4: data that it sells to advertisers, and advertisers can leverage

217
00:11:45,160 --> 00:11:50,079
Speaker 4: Google's models and ultimately the surveillance that they're built on

218
00:11:50,480 --> 00:11:53,680
Speaker 4: in order to precision target their message and thus reach

219
00:11:53,800 --> 00:11:56,800
Speaker 4: more customers. And that remains the core of the business model,

220
00:11:56,800 --> 00:11:58,960
Speaker 4: that remains the core of the Internet economy.

221
00:11:59,480 --> 00:12:01,920
Speaker 3: And putting that type.

222
00:12:01,640 --> 00:12:05,360
Speaker 4: Of information at the service of one or another government

223
00:12:05,720 --> 00:12:08,400
Speaker 4: is in my view, it is dangerous. And the first

224
00:12:08,400 --> 00:12:10,480
Speaker 4: thing I did is I think start a signal thread

225
00:12:10,480 --> 00:12:12,040
Speaker 4: with some people and say what should we do?

226
00:12:12,559 --> 00:12:14,800
Speaker 1: This was Project Maven. It was to do with helping

227
00:12:14,800 --> 00:12:18,920
Speaker 1: the US government target enemy combatants on the battlefield.

228
00:12:18,920 --> 00:12:23,160
Speaker 4: Though well, it was building surveillance and targeting AI for

229
00:12:23,679 --> 00:12:24,600
Speaker 4: the drone program.

230
00:12:24,679 --> 00:12:26,760
Speaker 1: And what was the result of me used so talking

231
00:12:26,760 --> 00:12:29,679
Speaker 1: to people on a signal. What happened next?

232
00:12:30,200 --> 00:12:33,120
Speaker 4: You know, it grew and it grew, and then I

233
00:12:33,160 --> 00:12:37,560
Speaker 4: wrote an open letter. That letter got signed by thousands

234
00:12:37,559 --> 00:12:40,520
Speaker 4: of people. Eventually someone leaked it to the New York Times.

235
00:12:41,120 --> 00:12:46,000
Speaker 4: It snowballed, it grew into a movement, and ultimately Google

236
00:12:46,160 --> 00:12:49,479
Speaker 4: dropped that contract. Now you know, these efforts are ongoing.

237
00:12:49,960 --> 00:12:53,280
Speaker 4: I want to be clear that this was a fairly

238
00:12:53,320 --> 00:12:57,280
Speaker 4: different time. I think now there's a lot more comfort

239
00:12:57,320 --> 00:13:00,440
Speaker 4: with military technology, but I think those issue shoes have

240
00:13:00,559 --> 00:13:03,720
Speaker 4: yet to be addressed. The issues of the danger of

241
00:13:04,280 --> 00:13:10,480
Speaker 4: yoking the interests of any one country irrespective to a

242
00:13:10,679 --> 00:13:13,560
Speaker 4: massive surveillance apparatus, the types of which we haven't seen

243
00:13:13,600 --> 00:13:16,199
Speaker 4: in human history. That's a real concern, and I think

244
00:13:16,200 --> 00:13:19,760
Speaker 4: that is a concern across political spectrum. We should all

245
00:13:19,800 --> 00:13:23,000
Speaker 4: be concerned about that, and frankly, I think it speaks

246
00:13:23,040 --> 00:13:26,959
Speaker 4: to the need to question that form altogether. The form

247
00:13:27,000 --> 00:13:30,440
Speaker 4: of sort of the surveillance business model and the collection

248
00:13:30,600 --> 00:13:36,239
Speaker 4: of massive amounts of information that can be used to harm, control, oppress,

249
00:13:36,480 --> 00:13:36,959
Speaker 4: et cetera.

250
00:13:37,240 --> 00:13:40,840
Speaker 1: How did you end up finally deciding what's time to leave?

251
00:13:41,880 --> 00:13:45,280
Speaker 4: There's not one answer to that, you know, I realized

252
00:13:45,280 --> 00:13:47,679
Speaker 4: pretty quickly that I had poked a bear. I had

253
00:13:47,679 --> 00:13:50,000
Speaker 4: a huge amount of support inside the company.

254
00:13:50,040 --> 00:13:53,600
Speaker 3: I still do. I don't think anyone's saying I'm wrong.

255
00:13:53,679 --> 00:13:55,720
Speaker 1: I mean, there are twenty thousand employees at one point

256
00:13:55,840 --> 00:13:59,000
Speaker 1: right involved in Yeah, and it's a huge no. I mean,

257
00:14:00,120 --> 00:14:01,319
Speaker 1: the employees are there.

258
00:14:01,800 --> 00:14:05,800
Speaker 4: At that point. I think it was like two hundred thousand. Yeah,

259
00:14:05,840 --> 00:14:07,160
Speaker 4: So it's it's significant.

260
00:14:07,200 --> 00:14:11,440
Speaker 3: It was the largest labor action in tech history, which

261
00:14:11,520 --> 00:14:15,439
Speaker 3: speaks not to the militancy of people who work in tech,

262
00:14:15,960 --> 00:14:18,480
Speaker 3: but I think speaks to the fact that this was

263
00:14:18,520 --> 00:14:21,280
Speaker 3: really a rupture of a lot of discomfort that had

264
00:14:21,280 --> 00:14:24,360
Speaker 3: been building for a long time. Right, Like, people.

265
00:14:24,040 --> 00:14:27,680
Speaker 4: Who often got into this industry because of their ethical compass,

266
00:14:27,680 --> 00:14:30,360
Speaker 4: because they really, you know, wanted to quote unquote change

267
00:14:30,440 --> 00:14:34,440
Speaker 4: the world, were feeling betrayed and feeling like their hands

268
00:14:34,440 --> 00:14:36,400
Speaker 4: were no longer on the rudder in a way that

269
00:14:36,440 --> 00:14:37,720
Speaker 4: they felt comfortable.

270
00:14:37,320 --> 00:14:43,600
Speaker 2: With coming up. We discussed why signal broke through the noise,

271
00:14:44,120 --> 00:14:44,560
Speaker 2: stay with.

272
00:14:44,600 --> 00:14:53,400
Speaker 1: Us, so you poked the bear and it became time

273
00:14:53,440 --> 00:14:56,080
Speaker 1: to leave. I mean, not for nothing. The kind of

274
00:14:56,360 --> 00:14:58,880
Speaker 1: one of the wedge issues was this use of Google's

275
00:14:58,880 --> 00:15:02,920
Speaker 1: technology for military purposes. And I think twenty twenty four

276
00:15:03,040 --> 00:15:05,080
Speaker 1: was the year. I mean, of course we will remember,

277
00:15:05,720 --> 00:15:08,480
Speaker 1: you know, the strikes, the drone strikes during the Obama years.

278
00:15:08,480 --> 00:15:11,720
Speaker 1: And this is not per se a new phenomenon, but

279
00:15:11,840 --> 00:15:16,240
Speaker 1: something about Ukraine and Gaza has really elevated into mainstream

280
00:15:16,280 --> 00:15:20,560
Speaker 1: consciousness what it means to have autonomous weapon systems.

281
00:15:20,880 --> 00:15:24,080
Speaker 4: Yeah, I think the environment has changed a lot. Look

282
00:15:24,160 --> 00:15:27,040
Speaker 4: on one side, there's a very real argument that I'm

283
00:15:27,040 --> 00:15:33,480
Speaker 4: deeply sympathetic to, like military infrastructure and technology is often outdated.

284
00:15:33,520 --> 00:15:39,760
Speaker 4: They're amortizing yanky old systems that are not up to snuff.

285
00:15:40,160 --> 00:15:44,400
Speaker 4: Logistics and processes are threadbare, all of that being true,

286
00:15:45,040 --> 00:15:47,880
Speaker 4: and some of the new technologies would make that a

287
00:15:47,880 --> 00:15:53,840
Speaker 4: lot easier. On the other hand, I think we're often

288
00:15:53,920 --> 00:15:57,000
Speaker 4: not looking at these dangers. What does it mean to

289
00:15:57,040 --> 00:16:00,360
Speaker 4: automate this process? What does it mean to predicate kate?

290
00:16:00,840 --> 00:16:06,560
Speaker 4: Many of these technologies on surveillance data on infrastructure that

291
00:16:06,680 --> 00:16:09,840
Speaker 4: is ultimately controlled by a handful of companies based in

292
00:16:09,880 --> 00:16:13,240
Speaker 4: the US. How do we make sure the very real

293
00:16:13,360 --> 00:16:18,560
Speaker 4: dangers of such reliance are actually surfaced and quick fixes

294
00:16:18,720 --> 00:16:24,440
Speaker 4: to bad practices accrued over many, many years of grifty

295
00:16:24,520 --> 00:16:29,320
Speaker 4: military contracting don't come at the expense of, you know,

296
00:16:29,880 --> 00:16:33,400
Speaker 4: civilian life. Don't come at the expense of handing the

297
00:16:33,480 --> 00:16:36,920
Speaker 4: keys to military operations to for profit companies based in

298
00:16:36,920 --> 00:16:37,360
Speaker 4: the US.

299
00:16:38,120 --> 00:16:41,280
Speaker 1: And you talked about Project Lavender and your Helmut Schmidt speech,

300
00:16:41,840 --> 00:16:44,120
Speaker 1: which I found fascinating. Why do you think it was

301
00:16:44,120 --> 00:16:45,240
Speaker 1: important to start there?

302
00:16:47,080 --> 00:16:51,600
Speaker 4: Well, Lavender, Gospel and Where's Daddy are three systems that

303
00:16:52,440 --> 00:16:58,440
Speaker 4: were revealed to be used in Gaza by the Israeli

304
00:16:58,560 --> 00:17:07,120
Speaker 4: military to identify targets and then, you know, basically kill them.

305
00:17:07,960 --> 00:17:11,520
Speaker 4: I looked at those because they are happening now, and

306
00:17:11,600 --> 00:17:16,800
Speaker 4: because they were the type of danger that the thousands

307
00:17:16,840 --> 00:17:20,399
Speaker 4: of people who were pushing back on Google's role in

308
00:17:20,440 --> 00:17:26,119
Speaker 4: military contracting were raising as a hypothetical, and because I

309
00:17:26,160 --> 00:17:29,760
Speaker 4: think it shows the way that the logics of the

310
00:17:29,800 --> 00:17:35,640
Speaker 4: Obama era drone war have been supercharged and exacerbated by

311
00:17:35,880 --> 00:17:41,000
Speaker 4: the introduction of these automated AI systems. So for folks

312
00:17:41,040 --> 00:17:44,159
Speaker 4: who might not remember, the drone war was the theater

313
00:17:44,680 --> 00:17:48,119
Speaker 4: to use the military term in which the signature strike

314
00:17:48,240 --> 00:17:53,760
Speaker 4: was introduced, and the signature strike was it's killing people

315
00:17:53,840 --> 00:17:57,639
Speaker 4: based on their data fingerprint, not based on any real

316
00:17:57,720 --> 00:18:00,800
Speaker 4: knowledge about who they are, what they've done, or anything

317
00:18:00,840 --> 00:18:04,800
Speaker 4: that would resemble what we think of as evidence. So,

318
00:18:05,560 --> 00:18:11,680
Speaker 4: because someone visited five locations that are flagged as terrorist locations,

319
00:18:11,760 --> 00:18:15,880
Speaker 4: because they have family here, because they are in these

320
00:18:16,000 --> 00:18:20,280
Speaker 4: four group chats, whatever it is, I'm making a possible data.

321
00:18:20,320 --> 00:18:23,000
Speaker 4: But you could model what a terrorist looks like in

322
00:18:23,040 --> 00:18:26,679
Speaker 4: any way you want. And then if if my data

323
00:18:26,720 --> 00:18:32,080
Speaker 4: patterns look similar enough to that model, you deem me

324
00:18:32,280 --> 00:18:35,920
Speaker 4: to be a terrorist and you kill me. And you

325
00:18:36,000 --> 00:18:38,640
Speaker 4: know it is exactly the logic of ad targeting.

326
00:18:39,040 --> 00:18:39,240
Speaker 3: Right.

327
00:18:39,560 --> 00:18:44,160
Speaker 4: I scroll through Instagram. I see an ad for athletic greens.

328
00:18:44,960 --> 00:18:46,760
Speaker 4: Because you know they have my credit card data. You

329
00:18:46,760 --> 00:18:49,440
Speaker 4: buy healthy stuff. You're this age group, you live in

330
00:18:49,480 --> 00:18:52,520
Speaker 4: New York, whatever it is, we assume you will buy

331
00:18:52,920 --> 00:18:55,200
Speaker 4: athletic greens or be likely to click on this ad

332
00:18:55,240 --> 00:19:00,840
Speaker 4: and will get paid. It's that but for death and

333
00:19:00,920 --> 00:19:07,080
Speaker 4: the lavender, Where's Daddy? And gospel systems are basically that

334
00:19:07,760 --> 00:19:10,600
Speaker 4: for death supercharged.

335
00:19:12,840 --> 00:19:17,480
Speaker 1: And is it a coincidence the resemblance between how AD

336
00:19:17,480 --> 00:19:21,680
Speaker 1: targeting works and how these signature strikes on the battlefield

337
00:19:22,119 --> 00:19:26,080
Speaker 1: or in war work. Was the military inspired in some

338
00:19:26,200 --> 00:19:29,080
Speaker 1: way by the ad industry or what's the kind of

339
00:19:29,080 --> 00:19:30,119
Speaker 1: source of the connections.

340
00:19:31,840 --> 00:19:34,439
Speaker 4: I wouldn't be surprised. But then again, you know, it

341
00:19:34,480 --> 00:19:36,639
Speaker 4: wasn't just the ad industry. You know, there was an

342
00:19:36,680 --> 00:19:40,840
Speaker 4: Obama era and kind of a neoliberal faith in data.

343
00:19:41,359 --> 00:19:44,720
Speaker 4: Now there wasn't much of a definition of what data is,

344
00:19:44,760 --> 00:19:48,560
Speaker 4: but the idea that data was more objective, less fallible,

345
00:19:49,160 --> 00:19:53,200
Speaker 4: more scientific. These are almost like mythologized versions of data,

346
00:19:53,960 --> 00:19:56,919
Speaker 4: and that if we relied on the data instead of

347
00:19:57,240 --> 00:20:02,560
Speaker 4: subjective decision making, we would re each determinations that were better,

348
00:20:02,640 --> 00:20:07,359
Speaker 4: We would make the right choices. Now, part of my

349
00:20:07,359 --> 00:20:10,680
Speaker 4: background is I came up doing large scale measurement system

350
00:20:10,720 --> 00:20:12,359
Speaker 4: so I was in the business of making data.

351
00:20:12,680 --> 00:20:14,439
Speaker 1: That's interesting. I mean, I just want to pause on that,

352
00:20:14,480 --> 00:20:18,160
Speaker 1: because that's an interesting way. I've never heard it expressed before.

353
00:20:18,160 --> 00:20:21,080
Speaker 1: But essentially making data equals measuring stuff.

354
00:20:21,920 --> 00:20:25,960
Speaker 4: Yeah, data is a very rough proxy for a complex reality.

355
00:20:26,200 --> 00:20:27,600
Speaker 3: Right, So you.

356
00:20:27,520 --> 00:20:32,320
Speaker 4: Figure out, like, what's the way we're gonna create those proxies.

357
00:20:32,359 --> 00:20:34,040
Speaker 4: How are we going to measure it? Right, whether we

358
00:20:34,119 --> 00:20:38,679
Speaker 4: measure human emotion, or measure the timing of street lights,

359
00:20:38,800 --> 00:20:41,080
Speaker 4: or measure the temperature in the morning based on a

360
00:20:41,119 --> 00:20:44,080
Speaker 4: thermometer that's calibrated in a certain way. We log all

361
00:20:44,119 --> 00:20:46,800
Speaker 4: of those as become data, right, And of course those

362
00:20:46,840 --> 00:20:50,480
Speaker 4: methodologies are generally created by people with an interest in

363
00:20:50,520 --> 00:20:52,840
Speaker 4: answering a certain set of questions, maybe not another set

364
00:20:52,840 --> 00:20:56,320
Speaker 4: of questions. In the case of Google or another company,

365
00:20:56,520 --> 00:21:00,760
Speaker 4: they were interested in measuring consumer behavior in servis of

366
00:21:01,600 --> 00:21:05,119
Speaker 4: giving advertisers access to a market. This is, in effect,

367
00:21:05,200 --> 00:21:08,439
Speaker 4: like a rough proxy created to answer very particular questions

368
00:21:08,480 --> 00:21:12,000
Speaker 4: by a very particular industry at a very particular historical conjuncture,

369
00:21:12,760 --> 00:21:15,800
Speaker 4: and it is then leveraged for all these other things.

370
00:21:15,840 --> 00:21:20,920
Speaker 4: And I think shrouding that data construction process in sort

371
00:21:20,960 --> 00:21:24,679
Speaker 4: of scientific language, you know, assuming that like data is

372
00:21:24,720 --> 00:21:29,400
Speaker 4: simply a synonym for objective fact, that's a very convenient

373
00:21:30,119 --> 00:21:33,359
Speaker 4: way of strouding some of those intentions, right, Shrouding the

374
00:21:33,560 --> 00:21:36,960
Speaker 4: distinction between those who get to measure and thus get

375
00:21:37,000 --> 00:21:40,120
Speaker 4: to decide, and those who are measured and are thus

376
00:21:40,200 --> 00:21:41,800
Speaker 4: subject to those decisions.

377
00:21:42,400 --> 00:21:44,920
Speaker 1: I mean, twenty twenty four was the year of this

378
00:21:45,920 --> 00:21:51,359
Speaker 1: hypothetical of what was raised at Google becoming real in Gaza.

379
00:21:51,560 --> 00:21:54,000
Speaker 1: It was also the year that two Nobel Prizes, one

380
00:21:54,040 --> 00:21:56,119
Speaker 1: in chemistry and one in physics, went to one former

381
00:21:56,160 --> 00:22:00,560
Speaker 1: and one current Google employee, and both in the realm

382
00:22:00,760 --> 00:22:05,480
Speaker 1: of pushing fundamental science forward. So how do you see

383
00:22:05,520 --> 00:22:07,760
Speaker 1: the moment in an AI and both this kind of

384
00:22:07,880 --> 00:22:11,960
Speaker 1: enormous hope and something drug discovery and the obvious peril

385
00:22:12,080 --> 00:22:14,200
Speaker 1: in getting people based on their metadata.

386
00:22:15,320 --> 00:22:18,960
Speaker 4: Well, I think there's a lot of things AI could

387
00:22:19,000 --> 00:22:22,879
Speaker 4: do very well right. Drug discovery is one, it's very interesting,

388
00:22:23,000 --> 00:22:25,200
Speaker 4: and the idea of being able to create drugs for

389
00:22:25,640 --> 00:22:29,320
Speaker 4: things that affect smaller numbers of people thus aren't generally incentivized.

390
00:22:29,359 --> 00:22:32,200
Speaker 3: All of that, like, yes, let's leverage what we.

391
00:22:32,200 --> 00:22:34,399
Speaker 4: Have to make life as beautiful as possible for as

392
00:22:34,400 --> 00:22:38,720
Speaker 4: many people as possible. Amen, The issue comes down to

393
00:22:38,760 --> 00:22:42,720
Speaker 4: the political economy of AI and what we're actually talking about.

394
00:22:43,359 --> 00:22:47,400
Speaker 4: And I think when we look at the AI industry

395
00:22:47,480 --> 00:22:50,919
Speaker 4: right now, and when we look at the technologies that

396
00:22:51,000 --> 00:22:55,000
Speaker 4: the Nobel Prize was awarded for either creating or leveraging,

397
00:22:56,440 --> 00:23:01,200
Speaker 4: we begin to recognize that AI is an incredibly centralized

398
00:23:01,400 --> 00:23:07,720
Speaker 4: set of technologies. It relies on huge amounts of computational resources,

399
00:23:08,280 --> 00:23:11,680
Speaker 4: and then there is data that's required to train these models.

400
00:23:11,720 --> 00:23:14,040
Speaker 4: And so the paradigm we're in right now is the

401
00:23:14,080 --> 00:23:17,400
Speaker 4: bigger is better. The more chips, the more compute, the

402
00:23:17,440 --> 00:23:21,280
Speaker 4: more data, and you combine those in whatever architecture and

403
00:23:21,320 --> 00:23:24,560
Speaker 4: you create a model, and then that model is purportedly

404
00:23:24,760 --> 00:23:28,360
Speaker 4: better performing and therefore better, and therefore we're advancing science.

405
00:23:28,880 --> 00:23:31,720
Speaker 1: And that's a useful argument, I guess for very very

406
00:23:31,760 --> 00:23:34,760
Speaker 1: large companies whose inherent logic is to grow larger. Right,

407
00:23:34,800 --> 00:23:37,080
Speaker 1: I mean, if you Amazon or Googo, little Microsoft, you

408
00:23:37,960 --> 00:23:41,840
Speaker 1: what is there to do for growth other than popularizing

409
00:23:41,840 --> 00:23:45,000
Speaker 1: this model of how the progress of AI works?

410
00:23:45,440 --> 00:23:48,680
Speaker 4: Yeah, I mean it would be very bad for these companies'

411
00:23:48,680 --> 00:23:52,800
Speaker 4: bottom lines if it turned out that much smaller models

412
00:23:53,359 --> 00:23:57,280
Speaker 4: using much more bespoke data were in fact much better, right,

413
00:23:57,320 --> 00:24:00,480
Speaker 4: Because they've thrown a lot of money behind this. Narratives

414
00:24:00,480 --> 00:24:02,520
Speaker 4: we hear about tech, and the narratives we hear about

415
00:24:02,520 --> 00:24:07,560
Speaker 4: progress are not always born of scientific and I think

416
00:24:07,640 --> 00:24:10,200
Speaker 4: when you look at that, then you have to realize,

417
00:24:10,240 --> 00:24:13,000
Speaker 4: like AI is not a it's not a product of

418
00:24:13,040 --> 00:24:17,199
Speaker 4: scientific innovation that everyone could leverage and these guys just

419
00:24:17,359 --> 00:24:20,600
Speaker 4: figured it out first, right, It's a product of this

420
00:24:20,760 --> 00:24:24,639
Speaker 4: concentration of resources, and in fact, in the early twenty

421
00:24:24,680 --> 00:24:27,960
Speaker 4: tens when the sort of current AI moment took off,

422
00:24:28,760 --> 00:24:34,159
Speaker 4: the algorithms that were animating that boom were created in

423
00:24:34,200 --> 00:24:38,359
Speaker 4: the late nineteen eighties. But what was new were the

424
00:24:38,400 --> 00:24:42,560
Speaker 4: computational resources, the chips and the ability to use them

425
00:24:42,600 --> 00:24:45,600
Speaker 4: in certain ways, and the data and the data is

426
00:24:45,640 --> 00:24:49,800
Speaker 4: that product of this surveillance business model that Google, Meta

427
00:24:49,840 --> 00:24:53,000
Speaker 4: at Alia sort of participated in and had built their

428
00:24:53,040 --> 00:24:55,919
Speaker 4: infrastructures and businesses around. So of course, who has the

429
00:24:55,960 --> 00:25:00,680
Speaker 4: access to those infrastructures, who can meaningfully apply these old

430
00:25:00,720 --> 00:25:04,080
Speaker 4: algorithms and begin to fund research sort of optimizing them,

431
00:25:04,200 --> 00:25:10,119
Speaker 4: building new approaches to this big data, big compute AI paradigm.

432
00:25:10,359 --> 00:25:15,919
Speaker 4: It's the same surveillance platform companies, And so all of

433
00:25:15,920 --> 00:25:17,879
Speaker 4: this comes back to the question of like, well, the

434
00:25:17,920 --> 00:25:21,320
Speaker 4: Nobel is awarded for AI this year, don't we want

435
00:25:21,320 --> 00:25:24,280
Speaker 4: better drug discoveries? And it's great, yes, we absolutely do,

436
00:25:25,080 --> 00:25:28,880
Speaker 4: but if you look at the market conditions, it's unclear

437
00:25:28,920 --> 00:25:32,240
Speaker 4: how that better drug discovery is actually going to lead

438
00:25:32,320 --> 00:25:37,280
Speaker 4: to better health outcomes, and so who will the customers

439
00:25:37,320 --> 00:25:38,159
Speaker 4: of that AI be?

440
00:25:38,680 --> 00:25:38,880
Speaker 2: Right?

441
00:25:39,320 --> 00:25:43,639
Speaker 4: Most likely it will be pharmaceutical companies, not altruistic organization.

442
00:25:43,800 --> 00:25:47,600
Speaker 4: It'll be insurance companies who may actually want to limit care. Right,

443
00:25:47,640 --> 00:25:50,240
Speaker 4: So you have a diagnostic algorithm, but it's not used

444
00:25:50,280 --> 00:25:53,240
Speaker 4: to make sure that I'm given the care I need,

445
00:25:53,760 --> 00:25:56,920
Speaker 4: even if it's expensive early so I live a long life.

446
00:25:56,960 --> 00:25:59,480
Speaker 4: It's used to make sure I'm excluded from the pool

447
00:26:00,240 --> 00:26:00,960
Speaker 4: so that I am.

448
00:26:00,800 --> 00:26:02,359
Speaker 3: Not a cost going forward.

449
00:26:02,359 --> 00:26:04,240
Speaker 4: So I think that's the lens that we need to

450
00:26:04,240 --> 00:26:07,240
Speaker 4: put on some of this, because the excitement is very warranted.

451
00:26:07,840 --> 00:26:11,400
Speaker 4: But in the world we have, it's pretty clear without

452
00:26:11,800 --> 00:26:16,960
Speaker 4: other changes to our systems, that AI, given the cost,

453
00:26:17,520 --> 00:26:21,040
Speaker 4: given the concentrated power, given the fact that these entities

454
00:26:21,160 --> 00:26:24,960
Speaker 4: need to recoup massive investment, is not going to be

455
00:26:25,320 --> 00:26:29,760
Speaker 4: an altruistic resource, but will likely be ramifying the power

456
00:26:30,040 --> 00:26:33,800
Speaker 4: of actors who have already proven themselves pretty problematic in

457
00:26:33,920 --> 00:26:37,639
Speaker 4: terms of fomenting public good so to speak.

458
00:26:40,160 --> 00:26:42,880
Speaker 2: Coming up, we'll hear about Meredith Whitaker's speech upon winning

459
00:26:42,920 --> 00:26:46,359
Speaker 2: the Helmet Schmidt Future Prize and why we should rethink

460
00:26:46,359 --> 00:26:48,960
Speaker 2: our preconceptions about data stayed with us.

461
00:26:57,320 --> 00:27:00,359
Speaker 1: Talk about Signal and not for nothing. Was a was

462
00:27:00,400 --> 00:27:03,040
Speaker 1: the first place you went when you learned about Project Maven,

463
00:27:03,880 --> 00:27:05,480
Speaker 1: and is now where you spend a lot more time.

464
00:27:05,680 --> 00:27:07,359
Speaker 1: So for those who don't know what it is and

465
00:27:07,800 --> 00:27:09,639
Speaker 1: how it works and why it's important, I mean, just

466
00:27:09,760 --> 00:27:10,640
Speaker 1: tell us a bit about it.

467
00:27:11,359 --> 00:27:14,719
Speaker 3: Yeah, Well, you know, Signal is incredible.

468
00:27:14,880 --> 00:27:18,399
Speaker 4: I'm really honored to work here and it's been along

469
00:27:18,440 --> 00:27:21,600
Speaker 4: on the journey and various capacities for a very long time.

470
00:27:22,119 --> 00:27:22,840
Speaker 3: It is the.

471
00:27:22,800 --> 00:27:29,320
Speaker 4: World's largest actually private communications network in an environment as

472
00:27:29,320 --> 00:27:33,000
Speaker 4: we just discussed, where almost every one of our activities,

473
00:27:33,400 --> 00:27:37,520
Speaker 4: digital and increasingly analog are surveilled in one way or

474
00:27:37,560 --> 00:27:42,760
Speaker 4: another by a large tech company, government, or some ad mixture.

475
00:27:42,920 --> 00:27:49,000
Speaker 4: So it's incredibly important infrastructure. We hear human rights workers

476
00:27:49,040 --> 00:27:53,000
Speaker 4: rely on signal. Journalists rely on signal to protect their sources.

477
00:27:53,280 --> 00:27:58,200
Speaker 4: We have militaries rely on signal to get communications out

478
00:27:58,240 --> 00:28:01,000
Speaker 4: when they're necessary. We have board rooms, you know, most

479
00:28:01,119 --> 00:28:04,880
Speaker 4: bour room chatter, most CEOs I meet use signal religiously.

480
00:28:05,040 --> 00:28:09,560
Speaker 3: Governments use signals. So anywhere where the confidentiality of communications

481
00:28:10,000 --> 00:28:15,280
Speaker 3: is necessary, certainly anywhere where the stakes of privacy are

482
00:28:15,359 --> 00:28:18,200
Speaker 3: life or death. We know that people rely on Signal

483
00:28:18,480 --> 00:28:22,320
Speaker 3: and our core encryption protocol was released in twenty thirteen

484
00:28:22,720 --> 00:28:24,359
Speaker 3: and it has stood the test of time. It is

485
00:28:24,400 --> 00:28:26,800
Speaker 3: now the gold standard across the industry.

486
00:28:26,400 --> 00:28:27,800
Speaker 1: And it's used by WhatsApp as well.

487
00:28:27,880 --> 00:28:29,720
Speaker 3: Right, yeah, it's used by WhatsApp.

488
00:28:29,720 --> 00:28:32,760
Speaker 4: It's used across the tech industry because one of the

489
00:28:32,800 --> 00:28:36,240
Speaker 4: things about encryption is it's very hard to get right.

490
00:28:36,400 --> 00:28:39,200
Speaker 4: It's very easy to get wrong, and so if someone

491
00:28:39,240 --> 00:28:41,520
Speaker 4: gets it right, you want to reuse that. You want

492
00:28:41,560 --> 00:28:44,520
Speaker 4: to use that formula. You don't want to DIY because

493
00:28:44,520 --> 00:28:48,000
Speaker 4: it's almost certain you'll have some small error in there,

494
00:28:48,480 --> 00:28:52,520
Speaker 4: and when we're talking about life or death consequences, you

495
00:28:52,560 --> 00:28:53,280
Speaker 4: can't afford that.

496
00:28:54,200 --> 00:28:56,520
Speaker 1: But why if what seven Signal used the same open

497
00:28:56,560 --> 00:28:59,520
Speaker 1: source code, am I so much safe for using Signal

498
00:28:59,520 --> 00:28:59,960
Speaker 1: and WhatsApp?

499
00:29:00,680 --> 00:29:05,920
Speaker 4: Because the Signal protocol is one slice of a very

500
00:29:06,000 --> 00:29:10,640
Speaker 4: large stack, and WhatsApp uses that slice to protect what

501
00:29:10,720 --> 00:29:14,800
Speaker 4: you say. But Signal protects way more than what you say.

502
00:29:15,040 --> 00:29:17,760
Speaker 4: We protect who you talk to. We don't have any

503
00:29:18,040 --> 00:29:21,680
Speaker 4: idea who's talking to whom. We don't know your profile name,

504
00:29:21,840 --> 00:29:24,320
Speaker 4: we don't know your profile photo. We don't know your

505
00:29:24,360 --> 00:29:27,120
Speaker 4: contact list, and it goes on and on and on.

506
00:29:27,520 --> 00:29:32,040
Speaker 4: So everything we do we sweat the details in order

507
00:29:32,360 --> 00:29:36,160
Speaker 4: to get as close to collecting as little data as possible.

508
00:29:36,440 --> 00:29:39,000
Speaker 1: Those things that you're describing that Signal doesn't collector are

509
00:29:39,000 --> 00:29:42,480
Speaker 1: the inputs to the signature strikes we were discussing earlier.

510
00:29:43,120 --> 00:29:44,120
Speaker 3: Could be? Could be.

511
00:29:44,320 --> 00:29:48,160
Speaker 4: I mean one of the issues with proprietary technology or

512
00:29:48,440 --> 00:29:54,000
Speaker 4: classified information is we don't totally know, but it's that

513
00:29:54,160 --> 00:29:58,200
Speaker 4: type of information that has been reported and mentioned as

514
00:29:58,400 --> 00:29:59,760
Speaker 4: inputs to those strikes.

515
00:29:59,840 --> 00:30:02,479
Speaker 1: Yeah, and you're not the CEO of Signal, you're the

516
00:30:02,520 --> 00:30:06,480
Speaker 1: president of the Signal Foundation. Can you kind of explain

517
00:30:07,280 --> 00:30:09,360
Speaker 1: why that is and what it means?

518
00:30:09,920 --> 00:30:14,400
Speaker 4: Well, I think in this moment in the tech industry,

519
00:30:14,520 --> 00:30:20,320
Speaker 4: that would be dangerous. Although it's not profits that we

520
00:30:20,560 --> 00:30:26,360
Speaker 4: are opposed to, it is the particular incentive structure of

521
00:30:26,480 --> 00:30:34,360
Speaker 4: the tech industry which ultimately you either collect data and

522
00:30:34,480 --> 00:30:38,040
Speaker 4: monetize that you're creating models to sell ads as we discussed,

523
00:30:38,120 --> 00:30:40,880
Speaker 4: or training AI models or selling its data brokers or

524
00:30:40,920 --> 00:30:44,640
Speaker 4: what have you, so surveillance, or you provide sort of

525
00:30:44,680 --> 00:30:46,760
Speaker 4: goods and services to those who do the picks and

526
00:30:46,760 --> 00:30:51,280
Speaker 4: shovels of the surveillance industry. So you can imagine a

527
00:30:51,320 --> 00:30:54,840
Speaker 4: board with a fiduciary duty governing a for profit signal.

528
00:30:55,840 --> 00:30:58,160
Speaker 4: At some point that fiduciary duty is going to take

529
00:30:58,200 --> 00:31:01,280
Speaker 4: precedence to the obsessive focus on privacy.

530
00:31:01,840 --> 00:31:04,200
Speaker 3: So that's really the crux of it.

531
00:31:04,760 --> 00:31:07,400
Speaker 4: And we're in an industry where consumer apps are quote

532
00:31:07,440 --> 00:31:11,120
Speaker 4: unquote free, right there is not a norm of paying

533
00:31:11,160 --> 00:31:14,360
Speaker 4: for these things upfront by the consumer. Is not for

534
00:31:14,480 --> 00:31:17,440
Speaker 4: lack of innovation that we don't have many many more

535
00:31:17,640 --> 00:31:21,440
Speaker 4: signals or signal like products that are focused on democracy,

536
00:31:21,520 --> 00:31:25,400
Speaker 4: focused on privacy, focused on ensuring fundamental rights, focused on

537
00:31:25,440 --> 00:31:29,080
Speaker 4: the public good. It's really because many of those things

538
00:31:29,200 --> 00:31:33,000
Speaker 4: cannot be subject to the surveillance business model and thus

539
00:31:33,240 --> 00:31:37,600
Speaker 4: don't really exist because capitalizing them is such an issue.

540
00:31:38,640 --> 00:31:41,560
Speaker 1: You mentioned that you know there's a consumer expectation that

541
00:31:41,880 --> 00:31:44,960
Speaker 1: tech is free, or at least free at the point

542
00:31:45,000 --> 00:31:47,600
Speaker 1: of delivery. You also have heard a piece of Wired

543
00:31:47,680 --> 00:31:50,120
Speaker 1: recently kind of arguing twenty twenty five might be but

544
00:31:50,200 --> 00:31:52,560
Speaker 1: the beginning of the end of big tech. But I

545
00:31:52,640 --> 00:31:56,680
Speaker 1: guess my question is big tech of so successfully kind

546
00:31:56,680 --> 00:32:01,040
Speaker 1: of created this expectation on the consumer side that tech

547
00:32:01,040 --> 00:32:04,040
Speaker 1: products are free and that essentially I pay for the

548
00:32:04,080 --> 00:32:08,720
Speaker 1: product by allowing myself to be surveiled. Can there be

549
00:32:08,760 --> 00:32:13,640
Speaker 1: a fundamental shift away from the big tech hegemony un

550
00:32:13,680 --> 00:32:18,200
Speaker 1: lessen until consumers are willing to pay with money rather

551
00:32:18,280 --> 00:32:21,280
Speaker 1: than data for infrastructure services.

552
00:32:22,920 --> 00:32:26,400
Speaker 3: I mean, I absolutely believe there can be. This is

553
00:32:26,440 --> 00:32:26,880
Speaker 3: what we have.

554
00:32:27,160 --> 00:32:29,680
Speaker 4: Thirty years of this, twenty years of this. This is

555
00:32:30,480 --> 00:32:35,000
Speaker 4: hyper novel in terms of human history. This is in

556
00:32:35,040 --> 00:32:38,400
Speaker 4: no way natural. There's you know, there's a longer history

557
00:32:38,440 --> 00:32:42,200
Speaker 4: of the sort of particular regulatory and political environment this

558
00:32:42,320 --> 00:32:47,560
Speaker 4: came out of. We now have the FBI advising people

559
00:32:47,640 --> 00:32:52,560
Speaker 4: not to send SMS messages following a massive hack that

560
00:32:52,840 --> 00:32:56,400
Speaker 4: enabled China and I don't know who else to access

561
00:32:56,680 --> 00:33:02,040
Speaker 4: telephony networks and surveil US citizens, including many high ranking citizens.

562
00:33:02,080 --> 00:33:07,720
Speaker 4: And so we're in a geopolitically volatile moment where the

563
00:33:07,760 --> 00:33:11,560
Speaker 4: lines between nations and jurisdictions and interests are getting a

564
00:33:11,560 --> 00:33:15,560
Speaker 4: bit more brittle, a bit more crisp, and I think

565
00:33:15,600 --> 00:33:21,480
Speaker 4: the imperative of moving away from this centralized surveillance tech

566
00:33:22,000 --> 00:33:25,800
Speaker 4: business model is really becoming clear to almost everyone. And

567
00:33:26,040 --> 00:33:28,000
Speaker 4: that was what animated the piece and wired you know,

568
00:33:28,040 --> 00:33:31,960
Speaker 4: I would say it's more of a wish casting into

569
00:33:31,960 --> 00:33:35,719
Speaker 4: the future than necessarily a prediction of what will happen.

570
00:33:35,840 --> 00:33:37,960
Speaker 3: But insofar as like hope.

571
00:33:37,680 --> 00:33:40,400
Speaker 4: Is an invitation for action, I really think there's a

572
00:33:40,480 --> 00:33:43,640
Speaker 4: lot of creative work that could be done to undo

573
00:33:43,760 --> 00:33:46,400
Speaker 4: that business model. Right, the question is really where do

574
00:33:46,440 --> 00:33:49,000
Speaker 4: the resources come from? And that's a question we can

575
00:33:49,040 --> 00:33:51,880
Speaker 4: begin to answer. Okay, are there endowments, Are there certain

576
00:33:51,920 --> 00:33:55,480
Speaker 4: technologies like signal, like some of the cloud and core

577
00:33:55,520 --> 00:33:58,840
Speaker 4: infrastructure that should be more openly governed that really, you know,

578
00:33:58,880 --> 00:34:02,320
Speaker 4: the more dependent we are on these infrastructures, the more

579
00:34:02,360 --> 00:34:05,840
Speaker 4: these infrastructures are controlled by you know, a small number

580
00:34:06,000 --> 00:34:09,720
Speaker 4: of actors, the more perilous those single points of failure become.

581
00:34:09,880 --> 00:34:13,960
Speaker 4: And I think the more attentive people are becoming to

582
00:34:14,040 --> 00:34:17,239
Speaker 4: that peril, and the more appetite there is for solutions

583
00:34:17,280 --> 00:34:19,640
Speaker 4: that may not have been on the table even a

584
00:34:19,640 --> 00:34:20,480
Speaker 4: couple of years ago.

585
00:34:21,200 --> 00:34:22,960
Speaker 1: Yeah, it is interesting. I mean I was a talk

586
00:34:23,000 --> 00:34:25,719
Speaker 1: the other day and somebody's talking about cloud computing and

587
00:34:25,760 --> 00:34:28,520
Speaker 1: how the metaphor kind of suggests that this kind of

588
00:34:28,520 --> 00:34:31,319
Speaker 1: decentralized model and it's kind of around all of us.

589
00:34:31,360 --> 00:34:33,719
Speaker 1: But like what cloud computing has actually meant it's this

590
00:34:33,800 --> 00:34:38,280
Speaker 1: incredible kind of centralization of computational power in the hands

591
00:34:38,320 --> 00:34:41,600
Speaker 1: of basically two nations and ten companies.

592
00:34:42,120 --> 00:34:47,080
Speaker 4: Yep, yeah, exactly, and you know, two nations, but the

593
00:34:47,200 --> 00:34:51,279
Speaker 4: US being dominant there it's Amazon, Microsoft, and Google have

594
00:34:51,440 --> 00:34:54,560
Speaker 4: seventy percent of the global market seven zero insane. So

595
00:34:54,960 --> 00:34:58,560
Speaker 4: that means other nation states are running on Amazon, running

596
00:34:58,560 --> 00:35:00,840
Speaker 4: on Google, right, And there is no way to simply

597
00:35:02,080 --> 00:35:05,000
Speaker 4: create a competitor because you are talking about a form

598
00:35:05,040 --> 00:35:08,600
Speaker 4: that was understood in the utilities context, particularly in telephony,

599
00:35:08,640 --> 00:35:11,040
Speaker 4: as a natural monopoly for a very long time, and

600
00:35:11,080 --> 00:35:16,479
Speaker 4: that has you know, ultimately been kind of entered into

601
00:35:16,480 --> 00:35:20,240
Speaker 4: the gravitational pull of these handful of companies. So it's a.

602
00:35:20,120 --> 00:35:20,960
Speaker 3: Very big problem.

603
00:35:21,000 --> 00:35:25,400
Speaker 4: But again, it doesn't mean that things like signal don't exist.

604
00:35:25,480 --> 00:35:31,360
Speaker 3: Signal exists. Swimming upstream signal is proving that one extremely

605
00:35:31,480 --> 00:35:36,720
Speaker 3: innovative technologies can and actually do exist, that even swimming

606
00:35:36,800 --> 00:35:41,040
Speaker 3: up against stream, we can create something that has set

607
00:35:41,080 --> 00:35:43,879
Speaker 3: the gold standard and proven that yes, there's a ton

608
00:35:43,920 --> 00:35:47,920
Speaker 3: of innovation left to do in tech, in privacy, in

609
00:35:48,040 --> 00:35:50,200
Speaker 3: rights preservation, in creating.

610
00:35:49,840 --> 00:35:52,719
Speaker 4: Tech that actually serves a social need. You know, it

611
00:35:52,840 --> 00:35:56,200
Speaker 4: is popular, and Okay, now let's solve the issue of

612
00:35:56,239 --> 00:36:01,600
Speaker 4: this toxic business model to actually foment innovation, to actually

613
00:36:01,600 --> 00:36:05,239
Speaker 4: foment progress, whatever the claims of the marketing on the

614
00:36:05,280 --> 00:36:06,440
Speaker 4: tan of big tech may be.

615
00:36:07,880 --> 00:36:10,719
Speaker 1: Yeah, and just to close, I've really was taken by

616
00:36:10,719 --> 00:36:13,680
Speaker 1: your Helmut Schmidt speech and you had a line, make

617
00:36:13,760 --> 00:36:17,400
Speaker 1: no mistake, I'm optimistic, but my optimism is an invitation

618
00:36:17,480 --> 00:36:21,839
Speaker 1: to analysis in action, not a ticket to complacency. Yes,

619
00:36:22,200 --> 00:36:22,960
Speaker 1: talk about.

620
00:36:22,680 --> 00:36:26,680
Speaker 4: That, Well, that is true, and I think that was

621
00:36:26,719 --> 00:36:29,080
Speaker 4: me also pushing back a little bit on the idea

622
00:36:29,200 --> 00:36:34,000
Speaker 4: that a grim diagnosis is somehow pessimism, right, uh huh, Like,

623
00:36:34,200 --> 00:36:36,719
Speaker 4: we know it's not great, we know it's not great.

624
00:36:36,719 --> 00:36:39,000
Speaker 4: Along a lot of axis tech is not the only

625
00:36:39,040 --> 00:36:43,360
Speaker 4: one we know things aren't working. But the most dangerous,

626
00:36:43,360 --> 00:36:46,960
Speaker 4: and frankly, the most pessimistic place to be is pushing

627
00:36:47,000 --> 00:36:52,279
Speaker 4: that aside, disavowing that, diluting that analysis in service of

628
00:36:52,320 --> 00:36:53,440
Speaker 4: immediate comfort.

629
00:36:53,239 --> 00:36:57,000
Speaker 1: Stoping to bother to critique because you think there's no point. Essentially,

630
00:36:57,080 --> 00:36:57,800
Speaker 1: that's the ultimes.

631
00:36:57,960 --> 00:37:00,560
Speaker 4: Well, I would say that's almost nihilism. I would say

632
00:37:00,600 --> 00:37:03,400
Speaker 4: the pessimism is where we paint a rosy picture so

633
00:37:03,440 --> 00:37:06,200
Speaker 4: we don't have to feel uncomfortable and then base our

634
00:37:06,239 --> 00:37:09,319
Speaker 4: strategy on that rosy picture in a way that is

635
00:37:09,960 --> 00:37:13,440
Speaker 4: wholly insufficient to actually tackle the problem, because we never

636
00:37:13,480 --> 00:37:17,760
Speaker 4: scope the problem, because we were unwilling to carry the

637
00:37:17,800 --> 00:37:24,000
Speaker 4: intellectual emotional responsibility of recognizing what we're actually up against.

638
00:37:24,200 --> 00:37:26,600
Speaker 4: So I think the most optimistic thing we can be

639
00:37:26,680 --> 00:37:30,480
Speaker 4: doing right now is really understanding where we stand, really

640
00:37:30,600 --> 00:37:34,960
Speaker 4: understanding what is the tech industry, how does it function,

641
00:37:35,200 --> 00:37:38,239
Speaker 4: how does money move? And okay, what are the alternatives

642
00:37:38,239 --> 00:37:41,560
Speaker 4: that exist, how do we support them, and how do

643
00:37:41,600 --> 00:37:46,920
Speaker 4: we recognize the power that we have to shift things right?

644
00:37:46,960 --> 00:37:49,520
Speaker 4: But that has to be based on a realistic picture.

645
00:37:49,840 --> 00:37:51,240
Speaker 4: It can't be based on delusion.

646
00:37:51,800 --> 00:37:54,279
Speaker 1: And part of that, as you know firsthand, coming back

647
00:37:54,280 --> 00:37:57,200
Speaker 1: to the beginning of the conversation, is acting as a collective.

648
00:37:58,000 --> 00:37:58,320
Speaker 3: Yeah.

649
00:37:58,360 --> 00:38:01,200
Speaker 4: Absolutely, I think no one person is going to do

650
00:38:01,239 --> 00:38:04,640
Speaker 4: this alone ever ever has ever you know, one person

651
00:38:04,680 --> 00:38:11,880
Speaker 4: may have taken credit at some point or another, but ultimately,

652
00:38:11,920 --> 00:38:14,760
Speaker 4: like the Internet was a collective effort, the open source

653
00:38:14,800 --> 00:38:17,759
Speaker 4: Library is you know, every single big tech company that

654
00:38:17,840 --> 00:38:21,560
Speaker 4: has one CEO, it has millions and millions of contributors.

655
00:38:21,640 --> 00:38:27,120
Speaker 4: And so we change this together. The issue is will

656
00:38:27,440 --> 00:38:33,440
Speaker 4: and the issue is resources. The issue is not ideas right.

657
00:38:33,440 --> 00:38:35,759
Speaker 4: We're not waiting for one genius to figure it out.

658
00:38:36,120 --> 00:38:40,160
Speaker 4: We're waiting for a clear map and some space to

659
00:38:40,239 --> 00:38:43,319
Speaker 4: examine it together and share insights and then figure out

660
00:38:43,360 --> 00:38:45,719
Speaker 4: how to push forward to a world where it's you know,

661
00:38:45,880 --> 00:38:51,000
Speaker 4: thousands of interesting projects that are all thinking together about

662
00:38:51,080 --> 00:38:55,799
Speaker 4: creating much better tech and reshaping the industry and its

663
00:38:55,840 --> 00:38:57,879
Speaker 4: incentives in order to nurture that.

664
00:39:01,719 --> 00:39:01,959
Speaker 1: Us.

665
00:39:02,080 --> 00:39:04,360
Speaker 2: First of all, Meredith said about fifteen things that I

666
00:39:04,360 --> 00:39:06,719
Speaker 2: would put on a small office couch pillow if I

667
00:39:06,760 --> 00:39:07,840
Speaker 2: knew how to needle point.

668
00:39:07,680 --> 00:39:10,239
Speaker 1: Well, start with the thing you would love most to

669
00:39:10,239 --> 00:39:11,040
Speaker 1: put on a needle point.

670
00:39:11,239 --> 00:39:14,400
Speaker 2: A grim diagnosis is not an invitation to pessimism.

671
00:39:14,560 --> 00:39:15,279
Speaker 1: I love that too.

672
00:39:15,760 --> 00:39:21,000
Speaker 2: I just think that, especially right now, it's easy and

673
00:39:21,160 --> 00:39:24,600
Speaker 2: right to say that there's a lot to be wary of. Yeah,

674
00:39:24,640 --> 00:39:27,280
Speaker 2: but it also is not an invitation to be weary,

675
00:39:27,360 --> 00:39:28,200
Speaker 2: ad infinitum.

676
00:39:28,400 --> 00:39:30,960
Speaker 1: No. No, I totally agree with you. I mean, I

677
00:39:31,000 --> 00:39:34,520
Speaker 1: think often if you're critical, you'll called out for being pessimistic.

678
00:39:34,840 --> 00:39:36,759
Speaker 1: But I thought the way that Meredith reframed that and

679
00:39:36,840 --> 00:39:40,000
Speaker 1: was basically no, no hold on a second it's optimistic

680
00:39:40,040 --> 00:39:42,239
Speaker 1: to be critical was really fascinating.

681
00:39:42,480 --> 00:39:45,240
Speaker 2: Yeah. I think my favorite line in the whole interview,

682
00:39:45,280 --> 00:39:48,000
Speaker 2: and maybe one of my just favorite lines I've heard

683
00:39:48,120 --> 00:39:51,120
Speaker 2: in reporting on technology ever, is that data is a

684
00:39:51,200 --> 00:39:53,320
Speaker 2: rough proxy for complex reality.

685
00:39:53,440 --> 00:39:56,120
Speaker 1: Sounds like you're going to need two pillows or just a.

686
00:39:56,160 --> 00:39:59,000
Speaker 2: Very big one. I actually want to talk about something

687
00:39:59,040 --> 00:40:01,160
Speaker 2: that we've talked about in our production meetings for this

688
00:40:01,200 --> 00:40:03,600
Speaker 2: show is that we don't want this show to be

689
00:40:03,680 --> 00:40:06,239
Speaker 2: all doom and gloom. And you know, when I found

690
00:40:06,239 --> 00:40:08,680
Speaker 2: out that you were interviewing Meredith, I was interested to

691
00:40:08,680 --> 00:40:10,320
Speaker 2: see what she would say outside of a sort of

692
00:40:10,400 --> 00:40:15,080
Speaker 2: technopessimist viewpoint, given that she has been quite critical of

693
00:40:15,160 --> 00:40:19,719
Speaker 2: Google and also of surveillance capitalism. And yet when you

694
00:40:19,960 --> 00:40:24,640
Speaker 2: hear her talk about Signal, I was sort of optimistic

695
00:40:24,760 --> 00:40:28,680
Speaker 2: in terms of, oh, here's a tool that we actually

696
00:40:28,680 --> 00:40:32,359
Speaker 2: have been encouraged by the FBI to use, yes, yet

697
00:40:32,440 --> 00:40:36,200
Speaker 2: still don't use. But here's a tool that is actually

698
00:40:36,440 --> 00:40:40,400
Speaker 2: answering a real fear that a lot of people have

699
00:40:40,480 --> 00:40:47,400
Speaker 2: that Meredith has, of privacy and doing some actionable stuff

700
00:40:47,840 --> 00:40:49,680
Speaker 2: to make us feel more secure.

701
00:40:50,280 --> 00:40:53,920
Speaker 1: Well, Meredith, ultimately the buck stops with her at signal,

702
00:40:53,960 --> 00:40:56,919
Speaker 1: and like, if you're a product lead or if you're

703
00:40:57,160 --> 00:41:00,160
Speaker 1: running a company whose main thing is a product, you

704
00:41:00,239 --> 00:41:02,279
Speaker 1: sure as hell better be optimistic because you're not going

705
00:41:02,280 --> 00:41:02,960
Speaker 1: to get through your day.

706
00:41:03,000 --> 00:41:04,000
Speaker 3: That's right, that's right.

707
00:41:04,719 --> 00:41:06,520
Speaker 1: But but I agree with you. The thing that's really

708
00:41:06,560 --> 00:41:10,000
Speaker 1: stuck with me was this concept of technology being dual use.

709
00:41:10,800 --> 00:41:15,359
Speaker 1: And Meredith, you know, used this phrase signature strike, which

710
00:41:15,360 --> 00:41:19,360
Speaker 1: it turns out can apply both to serving you with

711
00:41:19,400 --> 00:41:22,320
Speaker 1: an AD to make you want to buy something and

712
00:41:22,719 --> 00:41:26,480
Speaker 1: using your metadata to make a calculation that you're most

713
00:41:26,560 --> 00:41:30,040
Speaker 1: probably an enemy combatant or a terrorist and kill you.

714
00:41:30,120 --> 00:41:32,120
Speaker 1: I mean, this is the idea that it's the same

715
00:41:32,520 --> 00:41:36,680
Speaker 1: basically set of analysis that can apply to shopping or

716
00:41:36,719 --> 00:41:38,920
Speaker 1: life and death. I don't know, I just I have

717
00:41:39,040 --> 00:41:40,080
Speaker 1: no sort thinking about that.

718
00:41:40,400 --> 00:41:43,280
Speaker 2: What certainly makes me want to protect that data set

719
00:41:48,080 --> 00:41:50,719
Speaker 2: that's it protects uff today. This episode was produced by

720
00:41:50,800 --> 00:41:54,640
Speaker 2: Victoria Dominguez, Lizzie Jacobs, and Eliza Dennis. It was executive

721
00:41:54,640 --> 00:41:58,560
Speaker 2: produced by me care Price, Oswalashan and Kate Osborne for

722
00:41:58,640 --> 00:42:02,840
Speaker 2: Kaleidoscope and Katrina Norvel for iHeart Podcasts. Our Engineer is

723
00:42:02,880 --> 00:42:05,399
Speaker 2: Bihit Frasier. Al Murdoch wrote our theme song.

724
00:42:05,800 --> 00:42:08,319
Speaker 1: Join us on Friday for tech Stuff's the Weekend tech

725
00:42:08,719 --> 00:42:11,040
Speaker 1: We'll run through our favorite headlines, talk with our friends

726
00:42:11,040 --> 00:42:13,880
Speaker 1: at four or four media, and try to tackle a question,

727
00:42:14,440 --> 00:42:18,440
Speaker 1: when did this become a thing. Please rate and review

728
00:42:18,440 --> 00:42:21,960
Speaker 1: on Apple Podcasts, Spotify, or wherever you get your podcasts,

729
00:42:22,400 --> 00:42:25,160
Speaker 1: and reach out to us at tech stuff podcast at

730
00:42:25,200 --> 00:42:28,160
Speaker 1: gmail dot com with your thoughts and feedback. Thank you,