1
00:00:03,920 --> 00:00:09,400
Speaker 1: Uhmerica, I don't know why I opened the show this way.

2
00:00:09,760 --> 00:00:12,440
Speaker 1: I really, I really don't know why you did. I

3
00:00:12,480 --> 00:00:17,920
Speaker 1: don't know why I did that. Um, Jamie, Jamie, Jamie, Hi, Hi,

4
00:00:19,800 --> 00:00:23,919
Speaker 1: is this a podcast? Yeah, it's like, after all these years,

5
00:00:24,480 --> 00:00:27,479
Speaker 1: it's a podcast. It's a podcast. It's a podcast. Sometimes

6
00:00:27,560 --> 00:00:30,880
Speaker 1: podcast I get on the phone to podcast and I

7
00:00:30,920 --> 00:00:32,720
Speaker 1: think that I'm just talking to my friends, But then

8
00:00:32,760 --> 00:00:36,320
Speaker 1: I remember that every relationship in my life is dominated

9
00:00:36,600 --> 00:00:39,280
Speaker 1: by podcasting, and I don't even know how to interact

10
00:00:39,360 --> 00:00:42,000
Speaker 1: with people outside of the filter of a zoom call anymore.

11
00:00:42,040 --> 00:00:49,440
Speaker 1: That it is incredibly accurately. I really, that's actually the

12
00:00:49,720 --> 00:00:53,000
Speaker 1: saddest thing I've ever heard of my entire life. Thank you, Jamie.

13
00:00:53,080 --> 00:00:56,160
Speaker 1: How are you doing it to me? Like? I'm gonna agree?

14
00:00:56,440 --> 00:01:00,920
Speaker 1: I was? I was terrible, and then I was great,

15
00:01:00,920 --> 00:01:03,960
Speaker 1: and now I'm fine. Excellent. That sounds like a solid trajectory.

16
00:01:03,960 --> 00:01:06,240
Speaker 1: That's a good little hero's journey. That's like the that's

17
00:01:06,280 --> 00:01:08,640
Speaker 1: like The Green Night. More or less, I didn't see

18
00:01:08,640 --> 00:01:11,120
Speaker 1: that movie. It looked long. It is it is, but

19
00:01:11,160 --> 00:01:14,000
Speaker 1: it's quite a film. I would not see a movie

20
00:01:14,040 --> 00:01:18,400
Speaker 1: over an hour and forty minutes anymore. That's marred out.

21
00:01:18,520 --> 00:01:21,280
Speaker 1: You know what, I don't recommend for short movies as

22
00:01:21,319 --> 00:01:24,440
Speaker 1: Herbert West Reanimator or The Reanimator. Sorry, it was before

23
00:01:24,480 --> 00:01:27,120
Speaker 1: the Herbert West one, which was a Halloween movie. I

24
00:01:27,160 --> 00:01:30,720
Speaker 1: loved and then rewatched this Halloween and it was really

25
00:01:30,760 --> 00:01:34,480
Speaker 1: great up until one one really horrifying scene that I

26
00:01:34,520 --> 00:01:37,959
Speaker 1: had kind of forgotten Reanimator before. It's got some amazing

27
00:01:38,000 --> 00:01:39,840
Speaker 1: stuff in it if you're a horror movie fan, and

28
00:01:39,880 --> 00:01:44,440
Speaker 1: then there is an incredibly uncomfortable sexual assault scene. Um

29
00:01:44,640 --> 00:01:50,160
Speaker 1: that is like really bad, like really bad. Wow, thank you.

30
00:01:50,200 --> 00:01:52,680
Speaker 1: I think we're opening this episode in a really strong,

31
00:01:53,040 --> 00:01:56,280
Speaker 1: powerful than you focused way. I want to recommend a

32
00:01:56,280 --> 00:02:00,400
Speaker 1: horror movie to you. If it's Midsummer, I'm not going

33
00:02:00,440 --> 00:02:04,320
Speaker 1: to listen. It's not. It's not. I only watch Midsummer

34
00:02:04,360 --> 00:02:07,480
Speaker 1: to get all horny for Will Poulter, and then I

35
00:02:07,520 --> 00:02:11,920
Speaker 1: turned it off Shifter he dies. But that's a reasonable

36
00:02:11,960 --> 00:02:14,840
Speaker 1: thing for a person to say. Absolutely, when the person

37
00:02:14,919 --> 00:02:16,799
Speaker 1: that I want to have sex with dies, I turn

38
00:02:16,840 --> 00:02:19,600
Speaker 1: off the movie because it's boring. After that, it's a

39
00:02:19,720 --> 00:02:22,480
Speaker 1: very healthy way to go through life. Thank you so much. Okay,

40
00:02:22,560 --> 00:02:24,960
Speaker 1: So the movie, which I will show you at some

41
00:02:25,000 --> 00:02:29,960
Speaker 1: point is called Pin heard of Pin so Pin Pin

42
00:02:30,360 --> 00:02:35,080
Speaker 1: is a It's about a pediatrician who has like a

43
00:02:35,120 --> 00:02:39,400
Speaker 1: life size medical dummy sitting in his office. Oh, this

44
00:02:39,440 --> 00:02:42,680
Speaker 1: doesn't seem like it's going a good place. The pediatrician

45
00:02:43,000 --> 00:02:46,280
Speaker 1: the only way. It's like a psychological thriller. It's not

46
00:02:46,400 --> 00:02:49,840
Speaker 1: really that gory. But the pediatrician the only way that

47
00:02:49,880 --> 00:02:53,720
Speaker 1: he can communicate with emotional honesty with his children is

48
00:02:53,760 --> 00:02:58,080
Speaker 1: by making a little ventriloquist ventriloquist voice for the dummy.

49
00:02:58,160 --> 00:03:02,720
Speaker 1: And so the ventriloquist dummy gives the kids sex head

50
00:03:02,840 --> 00:03:07,680
Speaker 1: the ventrala because dummy does all this stuff, and then um,

51
00:03:07,760 --> 00:03:10,040
Speaker 1: and then one of the kids thinks that the ventrals

52
00:03:10,160 --> 00:03:13,760
Speaker 1: quest dummy whose name is Pin, is real. And then

53
00:03:13,840 --> 00:03:16,960
Speaker 1: Pin starts to like control his thoughts and actions. And

54
00:03:17,000 --> 00:03:20,560
Speaker 1: then there's a scene where a nurse has sex with Pin,

55
00:03:21,080 --> 00:03:24,480
Speaker 1: but he's just a dummy. Okay, it's the greatest movie

56
00:03:24,480 --> 00:03:27,040
Speaker 1: I've ever seen. I'm not doing it justice that that

57
00:03:27,040 --> 00:03:29,800
Speaker 1: that sounds like quite a film, Jamie. That sounds like

58
00:03:29,840 --> 00:03:33,640
Speaker 1: quite a film. Mars has sex with Pin, Robert and

59
00:03:34,720 --> 00:03:36,840
Speaker 1: I don't. Yeah, I mean that definitely does sound like

60
00:03:36,880 --> 00:03:41,480
Speaker 1: a scene that would make me have very specific physical reaction.

61
00:03:41,520 --> 00:03:43,240
Speaker 1: And the best part about that scene is that the

62
00:03:43,320 --> 00:03:45,720
Speaker 1: nurse never comes back and it's never addressed in the

63
00:03:45,760 --> 00:03:49,320
Speaker 1: movie again. And that now that now you're speaking the language,

64
00:03:49,480 --> 00:03:54,000
Speaker 1: the language of shoddy filmmaking, something horrible happens and then

65
00:03:54,400 --> 00:03:57,160
Speaker 1: you just canonically have to forget in order to watch

66
00:03:57,240 --> 00:03:59,800
Speaker 1: the rest of the movie. No, I'm on board with that.

67
00:04:00,040 --> 00:04:02,440
Speaker 1: And you know, Jamie, now that you mentioned quitting a

68
00:04:02,440 --> 00:04:03,920
Speaker 1: movie as soon as the person you want to have

69
00:04:03,960 --> 00:04:06,480
Speaker 1: sex with dies, that may explain where I've never made

70
00:04:06,520 --> 00:04:09,080
Speaker 1: it past the halfway point in the first Star Wars movie.

71
00:04:09,400 --> 00:04:12,200
Speaker 1: Really wow, Yeah, once Alec Guinness is out of the picture,

72
00:04:12,240 --> 00:04:15,160
Speaker 1: why even keep watching? You know? See? Well, yeah, you're like, well,

73
00:04:15,200 --> 00:04:17,479
Speaker 1: the hottest person is gone. I don't want not a

74
00:04:17,560 --> 00:04:20,960
Speaker 1: single fuckable face. And the rest of that film a

75
00:04:21,040 --> 00:04:24,719
Speaker 1: bunch of ug goes Jamie. So you know who else

76
00:04:24,800 --> 00:04:28,400
Speaker 1: is a bunch of ug? Gosh that we did it.

77
00:04:28,560 --> 00:04:35,600
Speaker 1: It's so perfect. The people who run Facebook, it is

78
00:04:35,680 --> 00:04:38,680
Speaker 1: it is. We're talking today about the Facebook papers, which

79
00:04:38,760 --> 00:04:40,960
Speaker 1: is we'll talk about this a little bit more in detailed,

80
00:04:40,960 --> 00:04:43,839
Speaker 1: but an enormous cash of internal Facebook documents that just

81
00:04:43,880 --> 00:04:46,159
Speaker 1: got leaked revealing a tremendous amount of funked up shit.

82
00:04:46,839 --> 00:04:49,800
Speaker 1: And I think we have to start with the uncomfortable

83
00:04:50,000 --> 00:04:54,240
Speaker 1: situation that is everybody talking shit about how Mark Zuckerberg

84
00:04:54,240 --> 00:04:56,840
Speaker 1: looks like an android. I feel so mixed about it

85
00:04:56,880 --> 00:05:00,960
Speaker 1: because on one hand, yes, the thing that's bad about

86
00:05:01,040 --> 00:05:04,080
Speaker 1: him is not his appearance, but also yes, he does.

87
00:05:04,600 --> 00:05:07,800
Speaker 1: He does hit the uncanny valley. There's something missing in

88
00:05:07,839 --> 00:05:10,599
Speaker 1: his eyes. Look, there's something and and that's ivy league

89
00:05:10,600 --> 00:05:13,800
Speaker 1: boy syndrome, right, Like, that's not just him, that's anyone

90
00:05:13,839 --> 00:05:20,880
Speaker 1: who who graduates screen photos. I truly. I mean, even

91
00:05:20,920 --> 00:05:22,960
Speaker 1: though we did just refer to the entire cast of

92
00:05:23,040 --> 00:05:26,520
Speaker 1: Star Wars one as a bunch of ugos hideous, do

93
00:05:26,760 --> 00:05:31,360
Speaker 1: feel that it's like the worst, the most lazy thing

94
00:05:31,440 --> 00:05:35,000
Speaker 1: you could do is go after how someone looks when

95
00:05:35,000 --> 00:05:37,240
Speaker 1: there are so many other evil facets of him. I

96
00:05:37,279 --> 00:05:39,400
Speaker 1: will agree that there is no light in his eyes.

97
00:05:39,760 --> 00:05:42,599
Speaker 1: There's certainly no light in his eyes. There's nothing like

98
00:05:42,640 --> 00:05:46,400
Speaker 1: the pupils are very There's not that little thing that's

99
00:05:46,400 --> 00:05:48,440
Speaker 1: supposed to be there is not there. Yeah, he could

100
00:05:48,440 --> 00:05:50,279
Speaker 1: watch a kitten get murdered and it would just be

101
00:05:50,320 --> 00:05:53,400
Speaker 1: a dial tone inside his soul. He looks like a

102
00:05:53,520 --> 00:05:57,520
Speaker 1: character from the Polar Express. He looks he looks like

103
00:05:57,600 --> 00:06:00,920
Speaker 1: Herbert West from the movie Rhianna of Matter, but less

104
00:06:00,920 --> 00:06:05,200
Speaker 1: cares to check that he looks like Pin. Look up Pen.

105
00:06:05,680 --> 00:06:08,839
Speaker 1: I'm not going to look up Pen. At this point

106
00:06:08,880 --> 00:06:11,919
Speaker 1: in the history of the show, Jamie, we've recorded a

107
00:06:11,920 --> 00:06:14,520
Speaker 1: lot of podcast episodes about Facebook. You have been there

108
00:06:14,560 --> 00:06:17,960
Speaker 1: for what three of them? Um? Yeah, I forget. I

109
00:06:18,000 --> 00:06:20,800
Speaker 1: forget entirely how many episodes we did on Facebook. We

110
00:06:20,839 --> 00:06:23,359
Speaker 1: did three episodes on like the creation of Facebook, and

111
00:06:23,400 --> 00:06:25,720
Speaker 1: it's kind of a brief list of its crimes. I

112
00:06:25,720 --> 00:06:27,839
Speaker 1: think we did at least one follow up, maybe two

113
00:06:27,839 --> 00:06:32,120
Speaker 1: follow ups. Um. And then we've mentioned Facebook foppery fakery

114
00:06:32,320 --> 00:06:34,400
Speaker 1: in episodes of Like It could happen here in Worst

115
00:06:34,440 --> 00:06:36,599
Speaker 1: Year Ever. Facebook is personal to me for a couple

116
00:06:36,640 --> 00:06:39,800
Speaker 1: of reasons. Number one, number of people who helped raise

117
00:06:39,880 --> 00:06:42,240
Speaker 1: me have slowly lost their grip on reality and the

118
00:06:42,240 --> 00:06:46,160
Speaker 1: face of viral propaganda spread be a via Facebook's engagement algorithm,

119
00:06:46,240 --> 00:06:49,200
Speaker 1: and that's kind of bumped me out. Um yeah. And

120
00:06:49,360 --> 00:06:51,520
Speaker 1: number two, my friends and I all lost our jobs

121
00:06:51,520 --> 00:06:53,599
Speaker 1: in the company we built for a decade due to

122
00:06:53,720 --> 00:06:56,719
Speaker 1: the fact that Facebook told criminal lies about video metrics

123
00:06:57,000 --> 00:06:59,320
Speaker 1: that they have currently been find forty million dollars for,

124
00:07:00,279 --> 00:07:03,000
Speaker 1: which also frustrating. Actually over it, it sounds like you're

125
00:07:03,040 --> 00:07:05,720
Speaker 1: really over it. And you know, I I we've talked

126
00:07:05,760 --> 00:07:07,920
Speaker 1: about this. I don't know if it was on micro net,

127
00:07:07,960 --> 00:07:10,320
Speaker 1: but yeah, I also asked my job to that. I

128
00:07:10,360 --> 00:07:13,760
Speaker 1: mean we worked at the same place. Yeah, worked it

129
00:07:13,880 --> 00:07:17,360
Speaker 1: like yeah, I mean they all the whole industry went

130
00:07:17,440 --> 00:07:20,840
Speaker 1: up in flames. It's so sace. I'm still I'm still

131
00:07:20,880 --> 00:07:23,120
Speaker 1: mad about it. Yeah, I am mad about it, even

132
00:07:23,120 --> 00:07:25,560
Speaker 1: though like things have been going fine, great for me

133
00:07:25,680 --> 00:07:29,280
Speaker 1: career wise. Um, it's just like kind of bullshit. It's

134
00:07:29,360 --> 00:07:32,080
Speaker 1: kind of frustrating. It used to be. Did you catch

135
00:07:32,080 --> 00:07:35,640
Speaker 1: that Robert's career is going great? Yea? In my head,

136
00:07:35,680 --> 00:07:39,520
Speaker 1: I went, You're welcome. I'm a human rocket ship. I'm

137
00:07:39,560 --> 00:07:43,600
Speaker 1: the iggy pop of talking about Facebook Welcome. It's kind

138
00:07:43,640 --> 00:07:46,720
Speaker 1: of nice that we all just had to pivot to, like, Okay,

139
00:07:47,040 --> 00:07:49,480
Speaker 1: you can still talk about what you're passionate about, but

140
00:07:49,600 --> 00:07:51,400
Speaker 1: just no one has to look at you anymore. And

141
00:07:51,400 --> 00:07:53,880
Speaker 1: I'm like, that's actually not the worst thing. That's that ideal.

142
00:07:54,360 --> 00:07:56,520
Speaker 1: And in my case, it was like you don't have

143
00:07:56,600 --> 00:08:01,120
Speaker 1: to write articles. You just have to tell on a microphone,

144
00:08:01,240 --> 00:08:05,720
Speaker 1: which involves writing an article. But I don't know, they're easier.

145
00:08:06,520 --> 00:08:09,520
Speaker 1: It's true. You can, like you can really have a

146
00:08:09,560 --> 00:08:11,720
Speaker 1: series of bullet points and be like, well, you just

147
00:08:11,760 --> 00:08:15,720
Speaker 1: don't have to edit anymore. That's that's what that's what

148
00:08:15,840 --> 00:08:17,840
Speaker 1: we that's what we got rid of in this pivot

149
00:08:17,880 --> 00:08:21,800
Speaker 1: to podcasting. Editors were all, we're all green walding a

150
00:08:21,800 --> 00:08:24,440
Speaker 1: little bit. Yeah, it all worked out, but you know,

151
00:08:24,520 --> 00:08:29,240
Speaker 1: but also Q and on stealing the fact that there

152
00:08:29,240 --> 00:08:31,440
Speaker 1: will soon be death squads roving many of our neighbor

153
00:08:31,480 --> 00:08:34,160
Speaker 1: Like there's there's downsides to it too, you know. Um,

154
00:08:34,640 --> 00:08:37,200
Speaker 1: but then also meta, you know, and then yea meta,

155
00:08:37,240 --> 00:08:39,280
Speaker 1: thank god, we're getting meta. We'll be talking about meta

156
00:08:39,320 --> 00:08:42,880
Speaker 1: at some point. Um. But yeah, like it's it's Facebook's bad.

157
00:08:42,920 --> 00:08:45,440
Speaker 1: I don't like Facebook. But in one of those episodes,

158
00:08:45,480 --> 00:08:47,480
Speaker 1: and I forget which of those episodes, I said something

159
00:08:47,480 --> 00:08:49,760
Speaker 1: along the lines of, at this point, there's no moral

160
00:08:49,840 --> 00:08:54,240
Speaker 1: case for remaining employed by Facebook. Um. And earlier this year,

161
00:08:54,280 --> 00:08:57,200
Speaker 1: a Facebook employee named Francis Hoggan came to the same

162
00:08:57,240 --> 00:08:59,880
Speaker 1: conclusion on her own. Rather than jump out with her

163
00:09:00,040 --> 00:09:02,480
Speaker 1: dock options or whatever perk she decrued, and then get

164
00:09:02,480 --> 00:09:04,600
Speaker 1: another tech industry job, which is what a lot of

165
00:09:04,600 --> 00:09:06,960
Speaker 1: people do. I know people who have done this when

166
00:09:07,000 --> 00:09:09,880
Speaker 1: they were like, Facebook's kind of fucked up. I'm just

167
00:09:09,920 --> 00:09:12,920
Speaker 1: gonna go hop to another country company and make even

168
00:09:12,920 --> 00:09:16,200
Speaker 1: more money. Uh. Instead of doing that, Francis spent weeks

169
00:09:16,200 --> 00:09:21,640
Speaker 1: painstakingly photographing internal Facebook has its own internal communications network

170
00:09:21,720 --> 00:09:24,160
Speaker 1: that is patterned off of Facebook, but it's for like

171
00:09:24,200 --> 00:09:29,000
Speaker 1: the corporate the the employees to use. Um well, I

172
00:09:29,000 --> 00:09:32,160
Speaker 1: mean it's it's like it's like Slack, but probably more

173
00:09:32,200 --> 00:09:36,240
Speaker 1: all consuming and soul destroying. Um yeah, but she is.

174
00:09:36,679 --> 00:09:40,640
Speaker 1: There's like nothing worse. I mean, personality wise, I can't

175
00:09:40,679 --> 00:09:44,640
Speaker 1: stand someone who's like really killing it on Slack. That's

176
00:09:44,640 --> 00:09:47,760
Speaker 1: one of my least favorite traits in a person, as

177
00:09:47,800 --> 00:09:51,160
Speaker 1: if they're they're really giving it a percent on Slack.

178
00:09:51,280 --> 00:09:53,959
Speaker 1: I'm like, just I'm I'm asleep. I hate it. No,

179
00:09:54,200 --> 00:09:56,640
Speaker 1: I mean, Sophie barely hears from me when we need

180
00:09:56,679 --> 00:09:58,920
Speaker 1: to work, let alone, when we don't need to work

181
00:10:01,760 --> 00:10:09,720
Speaker 1: hardcent not true good stuff. So rather than you know,

182
00:10:10,440 --> 00:10:12,839
Speaker 1: take the money and run, so she gets in this

183
00:10:12,920 --> 00:10:15,920
Speaker 1: like internal communications app that Facebook has and because they

184
00:10:15,920 --> 00:10:19,360
Speaker 1: have protections about like because like they know that this

185
00:10:19,440 --> 00:10:21,720
Speaker 1: is a risk people leaking internal documents as a risk,

186
00:10:21,720 --> 00:10:23,840
Speaker 1: they have like security things set up and to get

187
00:10:23,840 --> 00:10:26,280
Speaker 1: past them, she just photographs a bunch of internal documents

188
00:10:26,280 --> 00:10:29,360
Speaker 1: on her phone. Um a huge Like she gets a

189
00:10:29,360 --> 00:10:31,920
Speaker 1: lot of shit. Uh, And then she didn't leaks those

190
00:10:31,920 --> 00:10:34,840
Speaker 1: files to several news agencies and are good friends at

191
00:10:34,840 --> 00:10:37,880
Speaker 1: the sec. UH. This week, we're going to go through

192
00:10:37,960 --> 00:10:40,120
Speaker 1: some of those documents and all of the damning ship

193
00:10:40,160 --> 00:10:43,600
Speaker 1: they reveal. The first and perhaps most important revelation is

194
00:10:43,640 --> 00:10:47,440
Speaker 1: this many Facebook employees understand that their company is a

195
00:10:47,480 --> 00:10:49,800
Speaker 1: force for evil. Some of them have vowed to fix

196
00:10:49,840 --> 00:10:52,400
Speaker 1: it from the inside. Others are convinced the evil is

197
00:10:52,440 --> 00:10:55,720
Speaker 1: outweighed by some nebulous good. But at the core of it,

198
00:10:56,040 --> 00:10:59,080
Speaker 1: they know that what they were a part of is problematic,

199
00:10:59,160 --> 00:11:00,840
Speaker 1: and a lot of them hate themselves for it. You

200
00:11:00,840 --> 00:11:04,520
Speaker 1: can really see that coming across in some of these conversations,

201
00:11:05,760 --> 00:11:10,000
Speaker 1: evidence of the yeah it's good stuff, don't you love? Yeah,

202
00:11:10,000 --> 00:11:13,559
Speaker 1: when human beings compromise the very nature of their soul

203
00:11:13,920 --> 00:11:16,920
Speaker 1: uh in the in seeking profit Yeah. And then and

204
00:11:16,960 --> 00:11:19,560
Speaker 1: then you know, you watch the you watch the light

205
00:11:19,640 --> 00:11:21,440
Speaker 1: leave their eyes, and then you're supposed to feel bad

206
00:11:21,480 --> 00:11:25,240
Speaker 1: for them. Uh. Evidence of the struggle over the soul

207
00:11:25,280 --> 00:11:27,920
Speaker 1: of Facebook can be found in the reactions of employees

208
00:11:27,960 --> 00:11:30,120
Speaker 1: to the growth of the Black Lives Matter movement. After

209
00:11:30,160 --> 00:11:33,959
Speaker 1: the murder of George Floyd by a cop that June,

210
00:11:33,960 --> 00:11:37,280
Speaker 1: as protests reached their height, a Facebook employee posted a

211
00:11:37,320 --> 00:11:41,560
Speaker 1: message to the company racial justice chatboards stating, get bright

212
00:11:41,640 --> 00:11:44,080
Speaker 1: Bart out of news tap. He was enraged at the

213
00:11:44,080 --> 00:11:46,559
Speaker 1: fact that the far right publisher was pushing this information

214
00:11:46,600 --> 00:11:49,680
Speaker 1: about violence at protests, and included screen grabs of bright

215
00:11:49,720 --> 00:11:54,199
Speaker 1: Bart articles with titles like Minneapolis Mayhem, massive looting, buildings

216
00:11:54,240 --> 00:11:59,760
Speaker 1: and flames, bonfires exclamation point, and BLM protesters pummel police cars.

217
00:12:00,080 --> 00:12:02,680
Speaker 1: Wonder how much more attention they paid the police cars

218
00:12:02,679 --> 00:12:05,400
Speaker 1: than the man who was choked to death by a cop. Anyway,

219
00:12:05,520 --> 00:12:09,960
Speaker 1: good stuff, good journalism, nailing it. This employee claimed that

220
00:12:09,960 --> 00:12:12,680
Speaker 1: these articles were part of a concerted effort by Breitbart

221
00:12:12,720 --> 00:12:15,200
Speaker 1: and other right wing media sites to quote paint Black

222
00:12:15,240 --> 00:12:18,440
Speaker 1: Americans and Black lead movements in a negative way. He

223
00:12:18,559 --> 00:12:21,320
Speaker 1: argued that none of those hyperpartisan sites deserved to be

224
00:12:21,400 --> 00:12:25,959
Speaker 1: highlighted by the Facebook news tab, so Facebook's News tab

225
00:12:26,160 --> 00:12:29,520
Speaker 1: consists of two tiers of content providers. Right, It's just

226
00:12:29,559 --> 00:12:30,960
Speaker 1: like the tab that tells you what's going on in

227
00:12:30,960 --> 00:12:33,400
Speaker 1: the world, and all of the people whose stories get

228
00:12:33,400 --> 00:12:36,119
Speaker 1: in there have been vetted to some extent by Facebook.

229
00:12:36,400 --> 00:12:39,040
Speaker 1: So there's a first tier of big publishers like the

230
00:12:39,040 --> 00:12:42,280
Speaker 1: New York Times, the Wall Street Journal, like the big dogs,

231
00:12:42,280 --> 00:12:44,319
Speaker 1: and they get paid. Facebook gives the money to be

232
00:12:44,360 --> 00:12:46,319
Speaker 1: a part of the News tab. And then there is

233
00:12:46,360 --> 00:12:48,760
Speaker 1: a second tier of news sites who are not paid

234
00:12:48,800 --> 00:12:51,440
Speaker 1: but did have to get vetted as a reliable news

235
00:12:51,480 --> 00:12:53,600
Speaker 1: source for Facebook to put them on their News tab.

236
00:12:53,960 --> 00:12:57,200
Speaker 1: And bright Bart is in that ladder tier, which means

237
00:12:57,200 --> 00:13:00,480
Speaker 1: Facebook isn't giving them money directly, but is institutionally pumping

238
00:13:00,480 --> 00:13:03,880
Speaker 1: a shipload of of like traffic towards their propaganda and

239
00:13:03,920 --> 00:13:06,440
Speaker 1: throwing a lot of their propaganda out into people's news feeds.

240
00:13:06,840 --> 00:13:10,160
Speaker 1: In their public facing statements, Facebook claims to only include

241
00:13:10,160 --> 00:13:13,280
Speaker 1: sites on their newstab who do quality news reporting. Sites

242
00:13:13,320 --> 00:13:17,320
Speaker 1: that repeatedly share disinformation, it claims are banned. This functions

243
00:13:17,320 --> 00:13:21,120
Speaker 1: on a strike system. In July, President Trump tweeted a

244
00:13:21,160 --> 00:13:24,440
Speaker 1: Brightbart video claiming you don't need a mask to protect

245
00:13:24,440 --> 00:13:28,960
Speaker 1: against COVID nineteen. The video also spread misinformation about hydroxy clarquin.

246
00:13:29,280 --> 00:13:32,880
Speaker 1: Despite the fact that this video clearly violated Facebook stated standards,

247
00:13:32,880 --> 00:13:34,800
Speaker 1: it was able to reach millions of people through the

248
00:13:34,800 --> 00:13:38,080
Speaker 1: newstab before Facebook took it down from the Wall Street Journal.

249
00:13:38,960 --> 00:13:41,720
Speaker 1: According to Facebook's fact checking rules, pages can be punished

250
00:13:41,720 --> 00:13:44,320
Speaker 1: if they acquire too many strikes, meaning they published content

251
00:13:44,400 --> 00:13:47,679
Speaker 1: deemed false by third party fact checkers. It requires two

252
00:13:47,720 --> 00:13:50,319
Speaker 1: strikes within ninety days to be deemed a repeat offender,

253
00:13:50,440 --> 00:13:53,240
Speaker 1: which can result in a user being suspended from posting content.

254
00:13:53,520 --> 00:13:56,840
Speaker 1: More strikes can lead to reductions and distribution and advertising revenue.

255
00:13:57,000 --> 00:13:59,840
Speaker 1: In a town hall, Mr Zuckerberg said Brightbart wasn't punished

256
00:13:59,840 --> 00:14:02,200
Speaker 1: for video because that was its only infraction in a

257
00:14:02,280 --> 00:14:05,720
Speaker 1: ninety day period. According to internal chats described to describe

258
00:14:05,760 --> 00:14:09,480
Speaker 1: the Yeah, now that seems wrong, right, knowing Brightbart that

259
00:14:09,520 --> 00:14:12,800
Speaker 1: they would have one strike and that's so I mean,

260
00:14:13,080 --> 00:14:16,160
Speaker 1: and was the reason that that video reached so many

261
00:14:16,160 --> 00:14:17,960
Speaker 1: people before it was taken down. It was that just

262
00:14:18,040 --> 00:14:21,120
Speaker 1: like a delay in fact checking, does that mean that

263
00:14:21,600 --> 00:14:24,320
Speaker 1: a certain number of people need to like? No? I mean?

264
00:14:24,320 --> 00:14:26,800
Speaker 1: Trump tweeted it and it spread and Facebook didn't want

265
00:14:26,800 --> 00:14:29,160
Speaker 1: to take it down until it had it had already

266
00:14:29,360 --> 00:14:31,640
Speaker 1: kind of made them some money, I think. I also

267
00:14:31,720 --> 00:14:33,200
Speaker 1: think it's just like they don't put a lot of

268
00:14:33,240 --> 00:14:35,400
Speaker 1: work into checking on this stuff. They don't. They don't

269
00:14:35,400 --> 00:14:36,840
Speaker 1: want to piss it. We're talking about all this, but

270
00:14:36,880 --> 00:14:39,920
Speaker 1: like they also just don't want to piss any conservatives off.

271
00:14:40,000 --> 00:14:41,880
Speaker 1: Like there's a lot of things going into why this

272
00:14:41,920 --> 00:14:46,280
Speaker 1: stuff is not in fashioned to any degree. Now you

273
00:14:46,480 --> 00:14:49,200
Speaker 1: express surprise at the fact that that Brightbart only had

274
00:14:49,200 --> 00:14:53,160
Speaker 1: one strike in ninety days. Let's talk about why. Yeah, So,

275
00:14:53,320 --> 00:14:56,160
Speaker 1: thanks to Francis Hoggans leaked documents, we now know that

276
00:14:56,200 --> 00:15:00,440
Speaker 1: Brightbart was one of the news sites Facebook considered managed partners.

277
00:15:00,480 --> 00:15:03,000
Speaker 1: These sites are part of a program whereby the social

278
00:15:03,040 --> 00:15:06,840
Speaker 1: network pairs handlers who work at Facebook with the website.

279
00:15:07,120 --> 00:15:09,920
Speaker 1: These handlers give Facebook a back channel to sites that

280
00:15:10,000 --> 00:15:13,360
Speaker 1: spread disinformation, which allows them to have content removed or

281
00:15:13,400 --> 00:15:16,720
Speaker 1: altered without giving the content maker a strike. So, in

282
00:15:16,720 --> 00:15:18,840
Speaker 1: other words, they put out the content it gets few

283
00:15:18,880 --> 00:15:22,720
Speaker 1: millions of times. Facebook one employee like messages and editor

284
00:15:22,760 --> 00:15:24,840
Speaker 1: and says like, hey, you need to change this now.

285
00:15:24,880 --> 00:15:27,240
Speaker 1: It gets changed after it's spread around, and they have

286
00:15:27,280 --> 00:15:29,280
Speaker 1: a way to strike and thus stay on the news tab.

287
00:15:30,200 --> 00:15:33,640
Speaker 1: That's a good way to do a back channel. That's

288
00:15:33,800 --> 00:15:36,800
Speaker 1: so dark, Okay, I mean that. I I guess if

289
00:15:36,800 --> 00:15:39,320
Speaker 1: you're looking for a way to keep misinformation up, that

290
00:15:39,520 --> 00:15:42,760
Speaker 1: is a logical way to go about it. Yeah. Yeah,

291
00:15:43,200 --> 00:15:46,120
Speaker 1: they do a perfect job. So it's actually you're saying

292
00:15:46,160 --> 00:15:50,680
Speaker 1: that bright Bart is accurate. Yes, perfectly accurate. That is

293
00:15:50,720 --> 00:15:53,600
Speaker 1: what we always say about Bridbart and Andrew Brightbart, a

294
00:15:53,640 --> 00:15:56,320
Speaker 1: man who did not do cocaine to until he died.

295
00:15:56,960 --> 00:15:59,840
Speaker 1: Um about someone who's who's got light in his eyes.

296
00:16:00,920 --> 00:16:05,240
Speaker 1: Not anymore, It doesn't so actual strikes were automatically escalated

297
00:16:05,240 --> 00:16:08,200
Speaker 1: for review by senior Facebook executives, who could decide to

298
00:16:08,240 --> 00:16:11,600
Speaker 1: overturn the punishment and remove a strike. Through these methods,

299
00:16:11,640 --> 00:16:15,400
Speaker 1: Facebook strike system for spreading disinformation actually proved to be

300
00:16:15,480 --> 00:16:18,320
Speaker 1: nothing at all, and he sufficiently large right wing website

301
00:16:18,360 --> 00:16:22,040
Speaker 1: was given numerous opportunities to avoid strikes without being delisted.

302
00:16:22,360 --> 00:16:24,800
Speaker 1: This was a problem that went further than Breitbart, as

303
00:16:24,800 --> 00:16:27,680
Speaker 1: The Wall Street General reports. In an internal memo, the

304
00:16:27,720 --> 00:16:29,840
Speaker 1: engineer said that he based his assessment in part on

305
00:16:29,840 --> 00:16:32,640
Speaker 1: a queue of three dozen escalations that he had stumbled onto.

306
00:16:32,720 --> 00:16:35,120
Speaker 1: The vast majority of which were on behalf of conservative

307
00:16:35,160 --> 00:16:38,280
Speaker 1: content producers. A summary of the engineer's findings was posted

308
00:16:38,280 --> 00:16:40,960
Speaker 1: to an internal message board. One case he cited regarding

309
00:16:41,000 --> 00:16:44,120
Speaker 1: pro Trump influencers Diamond and Silk, third party fact checkers

310
00:16:44,200 --> 00:16:47,120
Speaker 1: rated as false post on their page that said was

311
00:16:47,160 --> 00:16:50,200
Speaker 1: like porn stars Wait diamonded Silk? Oh do you not

312
00:16:50,240 --> 00:16:53,600
Speaker 1: know about Diamond and Silk? Oh? They are? Are they banned?

313
00:16:54,120 --> 00:16:56,720
Speaker 1: They're not great? No, no, but they're bad. They they're

314
00:16:56,800 --> 00:17:00,640
Speaker 1: not not nice people, not good people. Um so they

315
00:17:00,680 --> 00:17:04,040
Speaker 1: got yeah, it's a fact Checkers rated false a post

316
00:17:04,119 --> 00:17:06,280
Speaker 1: a post that Diamond and Silk made stating how the

317
00:17:06,280 --> 00:17:08,840
Speaker 1: hell is allocating million dollars in order to give a

318
00:17:08,960 --> 00:17:10,880
Speaker 1: raise to House members that don't give a damn about

319
00:17:10,880 --> 00:17:15,399
Speaker 1: Americans going to help stimulate America's economy. When fact checkers

320
00:17:15,480 --> 00:17:18,119
Speaker 1: rated that post false, a Facebook staff are involved in

321
00:17:18,119 --> 00:17:20,360
Speaker 1: the partner program, argued that there should be no punishment,

322
00:17:20,400 --> 00:17:23,159
Speaker 1: noting the publisher has not hesitated going public about their

323
00:17:23,160 --> 00:17:26,720
Speaker 1: concerns around alleged anti conservative bias on Facebook. So this

324
00:17:26,800 --> 00:17:28,439
Speaker 1: is a pretty minor case, but it shows what's going

325
00:17:28,480 --> 00:17:30,640
Speaker 1: on there. They post something that's not accurate. This raise

326
00:17:30,720 --> 00:17:34,200
Speaker 1: is not something that's like going through UM and fact

327
00:17:34,280 --> 00:17:36,920
Speaker 1: checkers flag it is inaccurate, which should mean it gets removed.

328
00:17:37,119 --> 00:17:39,240
Speaker 1: But then someone at Facebook is like, if we remove it,

329
00:17:39,240 --> 00:17:41,520
Speaker 1: they're gonna yell about us removing their post and it's

330
00:17:41,560 --> 00:17:43,120
Speaker 1: going to be a pain in the asked for us.

331
00:17:43,240 --> 00:17:46,120
Speaker 1: So just like fuck it. Yeah, this is I feel

332
00:17:46,119 --> 00:17:49,440
Speaker 1: like this is always the route that Facebook does. It's

333
00:17:49,520 --> 00:17:56,240
Speaker 1: just like this big, gigantic, bureaucratic style operation that people

334
00:17:56,760 --> 00:18:00,879
Speaker 1: uh do shitty things so that they're not convenience or

335
00:18:00,960 --> 00:18:04,639
Speaker 1: yelled at by someone else. Like it's also insidious and

336
00:18:04,760 --> 00:18:09,200
Speaker 1: also so boring at the same time. Yeah, it's it's

337
00:18:09,280 --> 00:18:13,840
Speaker 1: the consequences that aren't boring. Um. And to some extent,

338
00:18:13,960 --> 00:18:16,320
Speaker 1: this is true of a lot of the worst things

339
00:18:16,400 --> 00:18:21,199
Speaker 1: in history. There were an awful lot of men intotalitarian

340
00:18:21,320 --> 00:18:24,960
Speaker 1: societies who signed effectively or literally the death warrants of

341
00:18:25,000 --> 00:18:28,000
Speaker 1: their fellow man because like, well, otherwise it's going to

342
00:18:28,080 --> 00:18:30,520
Speaker 1: be a real pain in the ass for me, My

343
00:18:31,080 --> 00:18:33,440
Speaker 1: day at the office is going to be terror I

344
00:18:33,480 --> 00:18:35,720
Speaker 1: don't want to take this to the boss. I don't

345
00:18:35,720 --> 00:18:40,240
Speaker 1: want to escalate this yet just kill them. Yeah, yeah,

346
00:18:40,320 --> 00:18:42,800
Speaker 1: I mean it is so I like the most evil

347
00:18:42,840 --> 00:18:45,160
Speaker 1: stuff is done in a very slow and boring way.

348
00:18:45,200 --> 00:18:47,960
Speaker 1: I feel like it's just because if you can get

349
00:18:47,960 --> 00:18:50,280
Speaker 1: people to, you know, fall asleep, you can get away

350
00:18:50,280 --> 00:18:54,800
Speaker 1: with fucking murder. Like literally, Yep, it's good stuff. Loving

351
00:18:54,840 --> 00:18:58,800
Speaker 1: these papers, Robert, Yeah, they're very fun. So Diamond and

352
00:18:58,840 --> 00:19:01,440
Speaker 1: Silk were able to be the third party fact checker

353
00:19:01,520 --> 00:19:04,440
Speaker 1: into changing the rating on their article down from to

354
00:19:04,680 --> 00:19:07,639
Speaker 1: partly false, and with the help of the managed partner

355
00:19:07,760 --> 00:19:11,320
Speaker 1: escalation process, all of their strikes were removed. Um the

356
00:19:11,400 --> 00:19:14,479
Speaker 1: chat conversations that the general review showed that inside the company,

357
00:19:14,480 --> 00:19:19,040
Speaker 1: Facebook employees repeatedly demanded that higher ups explain the allegations. Quote,

358
00:19:19,240 --> 00:19:22,520
Speaker 1: we are apparently providing hate speech policy consulting and consequence

359
00:19:22,600 --> 00:19:26,560
Speaker 1: mitigation services to select partners, wrote one employee. Leadership is

360
00:19:26,600 --> 00:19:29,960
Speaker 1: scared of being accused of bias, wrote another. So that

361
00:19:30,080 --> 00:19:33,560
Speaker 1: seems bad. That seem good. Now, that seems like the

362
00:19:33,600 --> 00:19:35,320
Speaker 1: root of a lot of problems we've been having as

363
00:19:35,359 --> 00:19:38,600
Speaker 1: a society. Then like, well, conservatives are loud when they're angry,

364
00:19:38,640 --> 00:19:41,200
Speaker 1: so let's just let them lie and try to get

365
00:19:41,200 --> 00:19:45,040
Speaker 1: people killed. Diamond and Silk was doing that in that case,

366
00:19:45,040 --> 00:19:47,360
Speaker 1: but that that's a thing in writing media. Now when

367
00:19:47,359 --> 00:19:49,080
Speaker 1: you're saying, I don't know what to picture when you

368
00:19:49,119 --> 00:19:51,480
Speaker 1: say diamond and silk. So at first I was picturing

369
00:19:52,000 --> 00:19:56,960
Speaker 1: porn stars. Then I was picturing hair metal band looked

370
00:19:56,960 --> 00:20:00,440
Speaker 1: more like gospel singers. I was picturing two Springer sp annuals.

371
00:20:00,520 --> 00:20:03,119
Speaker 1: Most recently, I don't think I'll stay there. Yeah, no,

372
00:20:03,280 --> 00:20:07,359
Speaker 1: I wouldn't. I wouldn't. You should? You should look them up. UM, Yeah,

373
00:20:07,400 --> 00:20:10,640
Speaker 1: they're they're they're. There are two musicians who like opposed

374
00:20:10,640 --> 00:20:13,560
Speaker 1: with Trump and have like a I think they're on TikTok.

375
00:20:13,640 --> 00:20:18,000
Speaker 1: They're just like right wing media influencers, and they're they're

376
00:20:18,160 --> 00:20:21,240
Speaker 1: they're not they're not great people. UM. In a farewell

377
00:20:21,280 --> 00:20:24,840
Speaker 1: memo of colleagues in late one member of Facebook's Integrity

378
00:20:24,880 --> 00:20:27,280
Speaker 1: team UM and the Integrity team, their job is to

379
00:20:27,359 --> 00:20:31,240
Speaker 1: reduce harmful behavior on the platform, UH complained that Facebook's

380
00:20:31,280 --> 00:20:35,040
Speaker 1: tolerance for Britbarts stop them from effectively fighting hate speech. Quote.

381
00:20:35,280 --> 00:20:37,919
Speaker 1: We make special exceptions to our written policies for them,

382
00:20:37,960 --> 00:20:40,720
Speaker 1: and we even explicitly endorsed them by including the mistrusted

383
00:20:40,800 --> 00:20:45,720
Speaker 1: partners in our core products. Yeah, it's it's bad, and

384
00:20:45,880 --> 00:20:48,480
Speaker 1: you can see, like there's this constant with the Facebook

385
00:20:48,480 --> 00:20:52,159
Speaker 1: papers revealed, there's this constant seesaw and aggressive between the

386
00:20:52,160 --> 00:20:56,080
Speaker 1: integrity team, the people whose job is to reduce the

387
00:20:56,160 --> 00:20:59,600
Speaker 1: harm the site does, and everyone else who's only real

388
00:20:59,720 --> 00:21:02,119
Speaker 1: job is to increase engagement on the site. Right. That

389
00:21:02,240 --> 00:21:03,880
Speaker 1: is how you get your bonus, That is how you

390
00:21:04,320 --> 00:21:06,520
Speaker 1: get the kudos from the boss is keeping people on

391
00:21:06,560 --> 00:21:09,560
Speaker 1: the site for longer. So most of Facebook that is

392
00:21:09,600 --> 00:21:11,600
Speaker 1: their job, and a small number of people their job

393
00:21:11,640 --> 00:21:13,560
Speaker 1: is to try and make sure the site doesn't contribute

394
00:21:13,560 --> 00:21:16,560
Speaker 1: to an ethnic cleansing, and the ethnic cleansing people like

395
00:21:16,600 --> 00:21:19,000
Speaker 1: the people trying to stop that. The best way to

396
00:21:19,000 --> 00:21:20,640
Speaker 1: do that is always going to be to do things

397
00:21:20,640 --> 00:21:23,359
Speaker 1: that cut down on engagement with the site, and so

398
00:21:23,400 --> 00:21:27,879
Speaker 1: they nearly always lose the fights they have with everybody else. Um,

399
00:21:28,000 --> 00:21:32,679
Speaker 1: Jesus Christ. Yeah, it's great, Okay, okay, Yeah, that is

400
00:21:32,720 --> 00:21:36,160
Speaker 1: the scariest extension of that logic, yep ye. One thing

401
00:21:36,200 --> 00:21:38,040
Speaker 1: we know thanks to the Facebook papers is that the

402
00:21:38,080 --> 00:21:40,800
Speaker 1: social network launched a study in two thousand nineteen to

403
00:21:40,880 --> 00:21:43,840
Speaker 1: determine what level of trust its users had in different

404
00:21:43,840 --> 00:21:47,280
Speaker 1: media organizations. Out of dozens of websites across the US

405
00:21:47,320 --> 00:21:51,080
Speaker 1: and UK, Bright Bart was dead last. Facebook themselves rated

406
00:21:51,119 --> 00:21:53,720
Speaker 1: it as low quality, which, again based on the company's

407
00:21:53,760 --> 00:21:55,800
Speaker 1: own claims about how they decide who to include in

408
00:21:55,800 --> 00:21:59,200
Speaker 1: the news tab, would disqualify bright Bart. And guess what,

409
00:21:59,480 --> 00:22:02,320
Speaker 1: bright Bart is still a trusted Facebook partner. Oh hey,

410
00:22:02,440 --> 00:22:04,639
Speaker 1: what's this unrelated news clip from a too in November

411
00:22:05,359 --> 00:22:08,639
Speaker 1: one Washington Post article doing in my script? Quote? Right,

412
00:22:08,680 --> 00:22:11,399
Speaker 1: Bart is the most influential producer of climate change denial

413
00:22:11,440 --> 00:22:14,040
Speaker 1: posts on Facebook, according to a report released Tuesday that

414
00:22:14,080 --> 00:22:16,840
Speaker 1: suggests a small number of publishers play an outsize role

415
00:22:16,880 --> 00:22:20,560
Speaker 1: in creating content that undermines climate science. Good ship, Right,

416
00:22:20,920 --> 00:22:24,680
Speaker 1: that's still number one after all these years. Good Isn't

417
00:22:24,680 --> 00:22:26,760
Speaker 1: that a good thing? Isn't that a good thing? So

418
00:22:26,840 --> 00:22:29,760
Speaker 1: they haven't said two inaccurate things in the last ninety days,

419
00:22:29,800 --> 00:22:35,360
Speaker 1: which I find never Facebook's terrorists thought of offending conservatives

420
00:22:35,359 --> 00:22:38,760
Speaker 1: by cracking down on hate speech and rampant disinformation started.

421
00:22:38,840 --> 00:22:40,879
Speaker 1: I don't know if it started, but it really, it

422
00:22:40,960 --> 00:22:44,520
Speaker 1: really hit the ground running in two thousands sixteen, during

423
00:22:44,560 --> 00:22:47,160
Speaker 1: the only election that was even worse than this last election.

424
00:22:47,680 --> 00:22:50,720
Speaker 1: In May of that year, Gizmoto wrote an article reporting

425
00:22:50,720 --> 00:22:55,080
Speaker 1: that Facebook's trending topics lists suppressed conservative views. A handful

426
00:22:55,119 --> 00:22:58,280
Speaker 1: of ex employees made claims that seemed to back these allegations.

427
00:22:58,359 --> 00:23:01,080
Speaker 1: Up now. Reporting later in the year by NPR it

428
00:23:01,119 --> 00:23:04,280
Speaker 1: made it clear that this was bullshit quote. NPR called

429
00:23:04,320 --> 00:23:07,520
Speaker 1: up half a dozen technology experts, including data scientists who

430
00:23:07,560 --> 00:23:11,120
Speaker 1: have special access to Facebook's internal metrics. The consensus there

431
00:23:11,160 --> 00:23:14,080
Speaker 1: is no statistical evidence to support the argument that Facebook

432
00:23:14,080 --> 00:23:16,840
Speaker 1: does not give conservative views a fair shake. But truth

433
00:23:16,960 --> 00:23:20,320
Speaker 1: never matters when you're arguing with conservatives. They needed a

434
00:23:20,320 --> 00:23:23,480
Speaker 1: reason to threaten Facebook with regulation, et cetera. When Trump,

435
00:23:23,560 --> 00:23:26,439
Speaker 1: and when Trump won later that year, the social network

436
00:23:26,480 --> 00:23:28,720
Speaker 1: decides these threats may have teeth, and so we're going

437
00:23:28,800 --> 00:23:31,080
Speaker 1: to spend the next four years allowing them to say

438
00:23:31,119 --> 00:23:33,359
Speaker 1: whatever the funk they want, no matter how racist, no

439
00:23:33,359 --> 00:23:36,240
Speaker 1: matter how conspiratorial, no matter how many shootings it may

440
00:23:36,440 --> 00:23:38,560
Speaker 1: help to inspire, no matter how no matter how many

441
00:23:38,560 --> 00:23:41,240
Speaker 1: shootings may be live streamed on the platform, like the

442
00:23:41,320 --> 00:23:44,240
Speaker 1: christ shoot shooting. Um, we're gonna let it all in

443
00:23:45,040 --> 00:23:51,879
Speaker 1: all because yeah, money, well, because otherwise they yelled at

444
00:23:51,920 --> 00:23:56,320
Speaker 1: and maybe regulated. The conservatives are good angry. I don't

445
00:23:56,320 --> 00:23:59,639
Speaker 1: want the conservatives to get angry. The funny thing is

446
00:23:59,760 --> 00:24:02,320
Speaker 1: they're no stopping them from getting angry. Right, you know

447
00:24:02,560 --> 00:24:05,600
Speaker 1: how this works. I know how this works. They're gonna

448
00:24:05,600 --> 00:24:07,600
Speaker 1: be angry, and they're going to claim bias no matter what,

449
00:24:07,680 --> 00:24:10,240
Speaker 1: which is what they do. And so as Facebook gives

450
00:24:10,280 --> 00:24:13,199
Speaker 1: them a free pass and their content is consistently the

451
00:24:13,200 --> 00:24:16,919
Speaker 1: most influential stuff on the entire site, allegations of anti

452
00:24:17,000 --> 00:24:20,080
Speaker 1: right wing bias continue to spread, even though again like

453
00:24:20,119 --> 00:24:21,920
Speaker 1: eight of to nine out of the tin top shared

454
00:24:21,960 --> 00:24:24,080
Speaker 1: post on any even day are from right wing media.

455
00:24:24,280 --> 00:24:26,800
Speaker 1: But you know what's not from right wing media? Jamie?

456
00:24:27,160 --> 00:24:29,840
Speaker 1: What all these products and services that you're not at

457
00:24:29,840 --> 00:24:33,000
Speaker 1: all sure? You can't at all, not at all. We

458
00:24:33,240 --> 00:24:34,920
Speaker 1: have a different we have a different brand of brain

459
00:24:34,960 --> 00:24:37,760
Speaker 1: pills than the ones Alex Jones sells, and ours have

460
00:24:37,920 --> 00:24:41,520
Speaker 1: less than half brain pills, less than half the lead.

461
00:24:41,760 --> 00:24:44,240
Speaker 1: That is that. That is a promise, Jamie, every much

462
00:24:44,280 --> 00:24:47,200
Speaker 1: lad you think a post should have it's less than that.

463
00:24:47,400 --> 00:24:51,399
Speaker 1: Because we care, I'll take your sick little centrist brain pill.

464
00:24:51,640 --> 00:24:55,119
Speaker 1: See if I care, which I can start watching MSNBC

465
00:24:55,240 --> 00:25:01,560
Speaker 1: at any moment. Okay, take brain cooked, get brain cooked. Uh,

466
00:25:02,840 --> 00:25:11,479
Speaker 1: here's some other products. All right, so we're back. Okay,

467
00:25:11,600 --> 00:25:16,080
Speaker 1: So we're back, So we're back. Are think it's gonna

468
00:25:16,080 --> 00:25:18,320
Speaker 1: get happy. Are think he's gonna get happy? I think

469
00:25:18,320 --> 00:25:22,120
Speaker 1: it's gonna get funny. Not really, okay, just checking, Yeah,

470
00:25:22,200 --> 00:25:25,639
Speaker 1: that's not really. I mean Mark Zuckerberg will like, I

471
00:25:25,680 --> 00:25:29,000
Speaker 1: don't know, fall down a man hole someday. Maybe we're lucky.

472
00:25:29,440 --> 00:25:32,120
Speaker 1: That would be funny. That would be funny. In two

473
00:25:32,160 --> 00:25:35,560
Speaker 1: thousand eighteen, if Facebook engineer claimed on an internal message

474
00:25:35,560 --> 00:25:38,800
Speaker 1: board that the company was intolerant of his beliefs, the

475
00:25:38,880 --> 00:25:41,320
Speaker 1: reality is almost certainly that his co workers found him

476
00:25:41,320 --> 00:25:43,520
Speaker 1: to be an obnoxious bigot. I say this because he

477
00:25:43,600 --> 00:25:46,199
Speaker 1: left the company shortly thereafter and hit the grifting circuit,

478
00:25:46,240 --> 00:25:48,879
Speaker 1: showing up on Tucker Carlson's show. He just he does

479
00:25:48,920 --> 00:25:51,280
Speaker 1: the thing that like you remember two eighteen nineteen, a

480
00:25:51,320 --> 00:25:53,600
Speaker 1: bunch of these guys were like leaving big tech companies

481
00:25:53,600 --> 00:25:56,280
Speaker 1: and like going on the Alex Jones show. There was

482
00:25:56,320 --> 00:25:58,840
Speaker 1: one guy left Google and claimed like brought a bunch

483
00:25:58,880 --> 00:26:00,679
Speaker 1: of leaks, but they weren't in a thing because it

484
00:26:00,720 --> 00:26:02,760
Speaker 1: was never anything that people being like this guy kind

485
00:26:02,800 --> 00:26:07,680
Speaker 1: of seems like it sucks. It's very funny. Those press

486
00:26:07,720 --> 00:26:10,320
Speaker 1: tours were Yeah, that was truly that feels like it

487
00:26:10,359 --> 00:26:13,119
Speaker 1: was ten years ago, but yeah, it was. It was

488
00:26:13,160 --> 00:26:15,480
Speaker 1: funny because like I think the first one of these

489
00:26:15,600 --> 00:26:18,840
Speaker 1: dudes did all right money wise, but after that, like

490
00:26:19,320 --> 00:26:21,080
Speaker 1: this Spiggott dried up and so they were just like

491
00:26:21,200 --> 00:26:25,280
Speaker 1: detonating their careers in the tech industry for nothing, going

492
00:26:25,320 --> 00:26:32,760
Speaker 1: to work for gab afterwards. To watch after the election,

493
00:26:32,800 --> 00:26:34,840
Speaker 1: and I apologize for the rate that we're jumping around

494
00:26:34,880 --> 00:26:37,679
Speaker 1: here on the timeline, but it's unavoidable. Facebook became the

495
00:26:37,720 --> 00:26:40,119
Speaker 1: subject of bad pr from the left as well. The

496
00:26:40,160 --> 00:26:42,520
Speaker 1: Cambridge Analytic a scandal broke, and the outrage in the

497
00:26:42,520 --> 00:26:45,000
Speaker 1: wake of Trump's election meant that Facebook was being pressured

498
00:26:45,040 --> 00:26:48,760
Speaker 1: to do something about bigotry and disinformation spreading on their platform.

499
00:26:48,960 --> 00:26:51,280
Speaker 1: At the same time, the Republicans are in charge now,

500
00:26:51,320 --> 00:26:54,760
Speaker 1: so they can't actually do anything otherwise they'll be attacked

501
00:26:54,800 --> 00:26:58,000
Speaker 1: for being biased and maybe regulated. So they tested a

502
00:26:58,000 --> 00:27:02,040
Speaker 1: couple of different changes. One was a tool called sparing sharing,

503
00:27:02,320 --> 00:27:05,399
Speaker 1: which sought to reduce the reach of hyper posters. Hyper

504
00:27:05,440 --> 00:27:07,760
Speaker 1: posters are exactly what they sound like. These are users

505
00:27:07,760 --> 00:27:11,040
Speaker 1: that have been shown to mostly share false and hateful information,

506
00:27:11,400 --> 00:27:14,560
Speaker 1: and so reducing their virility was seen as potentially helpful.

507
00:27:14,920 --> 00:27:18,119
Speaker 1: This seems like a sensible change, right. Oh, these people

508
00:27:18,160 --> 00:27:21,640
Speaker 1: are are sharing at an incredible rate and it's all

509
00:27:21,840 --> 00:27:27,360
Speaker 1: violent trash. Let's reduce the number of stuff. I guess

510
00:27:27,359 --> 00:27:30,080
Speaker 1: that's like, that's a that's a real band aid just

511
00:27:30,200 --> 00:27:32,240
Speaker 1: be like, Okay, we're gonna have them. They can still

512
00:27:32,280 --> 00:27:36,480
Speaker 1: share stuff, but just less hateful stuff. Yeah, and it's

513
00:27:36,480 --> 00:27:39,160
Speaker 1: not less garbage. It's not even a shadow band because

514
00:27:39,160 --> 00:27:41,840
Speaker 1: the shadow ban would imply that, like you are actually reducing,

515
00:27:42,480 --> 00:27:46,680
Speaker 1: like artificially the spread, You're no longer artificially inflating their

516
00:27:46,680 --> 00:27:50,159
Speaker 1: reach because their stuff gets great engagement, right, because it

517
00:27:50,200 --> 00:27:53,919
Speaker 1: pisses people off even though it's untrue, and the algorithms

518
00:27:54,160 --> 00:27:57,800
Speaker 1: the default is, oh, this pisses people off. Let's let

519
00:27:57,800 --> 00:28:00,600
Speaker 1: everybody see what this asshole saying. And they're just being like, well,

520
00:28:00,640 --> 00:28:03,120
Speaker 1: let's not do that for these specific assholes, right, that's

521
00:28:03,119 --> 00:28:05,840
Speaker 1: all they're doing. Um, it's not a band, it's a

522
00:28:06,160 --> 00:28:08,920
Speaker 1: we're going to stop inflating these people's reached to the

523
00:28:09,000 --> 00:28:11,960
Speaker 1: same extent that we were. Seems like a sensible change.

524
00:28:12,480 --> 00:28:16,200
Speaker 1: You know who disagreed with that, Jamie Loftus, who Robert Evans,

525
00:28:16,320 --> 00:28:19,639
Speaker 1: Joel Kaplan, former deputy chief of staff to George W.

526
00:28:19,800 --> 00:28:24,240
Speaker 1: Bush and Facebook head of Public Policy UM famous right

527
00:28:24,320 --> 00:28:27,560
Speaker 1: wing ship had Joel Kaplan, who is huge at Facebook

528
00:28:27,760 --> 00:28:31,600
Speaker 1: UM and is a major driving force behind don't piss

529
00:28:31,600 --> 00:28:35,120
Speaker 1: off Conservatives. That's that's the guy that he is. That's

530
00:28:35,119 --> 00:28:37,360
Speaker 1: his whole job. How are we supposed to work together

531
00:28:37,440 --> 00:28:43,600
Speaker 1: if we're pissing on the conservatives? It actually at rising tide. Yeah,

532
00:28:43,720 --> 00:28:47,120
Speaker 1: so Kaplan's like, most of these hyper posters are conservatives.

533
00:28:47,160 --> 00:28:50,360
Speaker 1: This is this is you know, unfair, and he convinces

534
00:28:50,400 --> 00:28:53,480
Speaker 1: Zuckerberg to weaken have his engineers weakened the tools so

535
00:28:53,520 --> 00:28:55,959
Speaker 1: that they do kind of reduce the influence that these

536
00:28:56,040 --> 00:28:58,640
Speaker 1: hyper posters have, but not by as much as they

537
00:28:58,680 --> 00:29:01,160
Speaker 1: wanted to, and it doesn't really seem to have a

538
00:29:01,280 --> 00:29:03,280
Speaker 1: much of an impact. As we will talk about later,

539
00:29:03,760 --> 00:29:06,760
Speaker 1: this is still the way Facebook works. So however, to

540
00:29:06,760 --> 00:29:09,640
Speaker 1: whatever extent they did reduce the harm, it was not

541
00:29:09,720 --> 00:29:13,440
Speaker 1: by much. Another attempt is also like way too cool

542
00:29:13,760 --> 00:29:17,360
Speaker 1: of a word to describe what that is, which is

543
00:29:17,600 --> 00:29:20,200
Speaker 1: spreaders of hate speech. Why give them a cool name

544
00:29:20,240 --> 00:29:22,280
Speaker 1: like that? Yeah, why give him a cool name like that?

545
00:29:23,160 --> 00:29:25,480
Speaker 1: Although I don't know that might have that sounds like

546
00:29:25,520 --> 00:29:28,360
Speaker 1: something we might have said as like an insult to

547
00:29:29,360 --> 00:29:32,080
Speaker 1: people when I was young and on the internet. You're

548
00:29:32,080 --> 00:29:34,600
Speaker 1: a hyper poster. I don't know, dude, you're like hyper

549
00:29:34,680 --> 00:29:36,600
Speaker 1: posting right now. You need to chill the funk out.

550
00:29:36,880 --> 00:29:40,640
Speaker 1: I'm picturing someone sitting at their a filthy keyboard in

551
00:29:40,680 --> 00:29:44,760
Speaker 1: a Power Ranger suit. I am. I am imagining a

552
00:29:44,840 --> 00:29:47,760
Speaker 1: filthy and the filthy Power Rangers suit, Jamie. Oh, it's

553
00:29:47,920 --> 00:29:51,960
Speaker 1: really dirty and it doesn't fit either way too big

554
00:29:52,040 --> 00:29:55,520
Speaker 1: or way too small. Yeah. They they have soiled themselves

555
00:29:55,520 --> 00:29:58,120
Speaker 1: in it on more than one occasion because they can't

556
00:29:58,120 --> 00:30:01,520
Speaker 1: stop posting, Robert, because they're post too much. It was

557
00:30:01,600 --> 00:30:04,000
Speaker 1: It was not an accident. It was a choice they

558
00:30:04,040 --> 00:30:06,760
Speaker 1: made in order to finish an argument. I'm going to

559
00:30:06,920 --> 00:30:13,200
Speaker 1: make an oil painting of that exact image, Jamie. I

560
00:30:13,240 --> 00:30:15,080
Speaker 1: swear to you, I will put that up in my

561
00:30:15,160 --> 00:30:17,240
Speaker 1: living room. I will put it on my roof like

562
00:30:17,280 --> 00:30:23,680
Speaker 1: the Suesteem chapel. Really absolutely, don't underestimate how much free

563
00:30:23,680 --> 00:30:26,680
Speaker 1: time I have, Robert, I would never do that. You

564
00:30:26,720 --> 00:30:30,200
Speaker 1: work for the Internet. So another attempted tool to make

565
00:30:30,240 --> 00:30:34,000
Speaker 1: Facebook better was called informed engagement. This was supposed to

566
00:30:34,000 --> 00:30:37,280
Speaker 1: reduce the reach of posts that Facebook determined were more

567
00:30:37,320 --> 00:30:40,320
Speaker 1: likely to be shared by people who had not read them.

568
00:30:40,360 --> 00:30:43,680
Speaker 1: This rule was instituted, and over time, Facebook noticed a

569
00:30:43,720 --> 00:30:47,840
Speaker 1: significant decline in disinformation and toxic content. They commissioned to study,

570
00:30:47,880 --> 00:30:50,760
Speaker 1: which is where the problems started from. The Wall Street Journal.

571
00:30:51,640 --> 00:30:55,920
Speaker 1: The study, dubbed a political ideology analysis, suggested the company

572
00:30:55,960 --> 00:30:58,680
Speaker 1: had been suppressing the traffic of major far right publishers,

573
00:30:58,800 --> 00:31:01,520
Speaker 1: even though that wasn't it's an tint. According to the documents,

574
00:31:01,960 --> 00:31:04,719
Speaker 1: very conservative sites it found would benefit the most if

575
00:31:04,760 --> 00:31:07,600
Speaker 1: the tools were removed. With bright parts traffic increasing and

576
00:31:07,680 --> 00:31:13,080
Speaker 1: estimated Washington Times is eighteen percent, Western Journal sixteen percent,

577
00:31:13,200 --> 00:31:16,320
Speaker 1: and Epic Times by eleven percent. According to the documents,

578
00:31:16,960 --> 00:31:21,760
Speaker 1: the study was designed that's why you never conduct a study, Robert, Yeah. Yeah.

579
00:31:22,040 --> 00:31:25,040
Speaker 1: They find basically, hey, if you don't let people, if

580
00:31:25,080 --> 00:31:27,920
Speaker 1: you reduce the amount by which people share, like, by

581
00:31:28,040 --> 00:31:30,280
Speaker 1: by which posts shared by people who have not read

582
00:31:30,320 --> 00:31:32,840
Speaker 1: the article, if you if you make those spread less

583
00:31:32,960 --> 00:31:37,600
Speaker 1: bright parts, traffic drops. I still I still think that

584
00:31:37,600 --> 00:31:40,920
Speaker 1: that those like um, the tools that have developed over

585
00:31:41,000 --> 00:31:44,400
Speaker 1: time to be like, are you sure to read the article?

586
00:31:44,600 --> 00:31:48,360
Speaker 1: Our soul? Goofy? I I do like when Twitter, um,

587
00:31:48,640 --> 00:31:50,960
Speaker 1: I feel like they're like I just picturely a little

588
00:31:50,960 --> 00:31:53,840
Speaker 1: shaking person. Are you Are you sure you don't want

589
00:31:53,880 --> 00:31:57,600
Speaker 1: to read the article before you retweet it? What do

590
00:31:57,600 --> 00:32:01,360
Speaker 1: you think? I was like No, I felt so bad

591
00:32:01,400 --> 00:32:04,360
Speaker 1: because there's been times when I've like retweeted, like shared

592
00:32:04,400 --> 00:32:07,400
Speaker 1: my own articles that I've written, and because like I

593
00:32:07,920 --> 00:32:10,959
Speaker 1: wrote them, I didn't necessarily click the link before sharing.

594
00:32:11,000 --> 00:32:12,680
Speaker 1: I just like woke up and I like shared it,

595
00:32:12,680 --> 00:32:14,200
Speaker 1: and it's like, are you sure you don't want to

596
00:32:14,200 --> 00:32:16,960
Speaker 1: read this? And I just click to share because it's

597
00:32:16,960 --> 00:32:18,960
Speaker 1: like nine or ten in the morning and I haven't

598
00:32:18,960 --> 00:32:21,120
Speaker 1: had my coffee yet. But I feel bad even though

599
00:32:21,120 --> 00:32:23,520
Speaker 1: it's like, well, I wrote the motherfucker, like I know

600
00:32:23,680 --> 00:32:26,400
Speaker 1: what's in there. Usually by the time I share something,

601
00:32:26,440 --> 00:32:29,040
Speaker 1: I have already read it. But I do I think

602
00:32:29,040 --> 00:32:32,040
Speaker 1: that that function. I think it has a good purpose

603
00:32:32,120 --> 00:32:33,960
Speaker 1: and it like pains something in your brain that is

604
00:32:34,040 --> 00:32:35,800
Speaker 1: you know, yeah, I think it is a good thing.

605
00:32:36,240 --> 00:32:38,840
Speaker 1: It's just funny. It's this little Oliver twist that appears

606
00:32:38,840 --> 00:32:40,120
Speaker 1: in front of you and it's like, are you sure

607
00:32:40,160 --> 00:32:42,160
Speaker 1: you don't want to read the article for your share?

608
00:32:42,240 --> 00:32:46,040
Speaker 1: We're following like like I'm good I can read it's

609
00:32:46,040 --> 00:32:48,120
Speaker 1: all good. Would you like to maybe read the article

610
00:32:48,200 --> 00:32:52,760
Speaker 1: before suggesting that an ethnic group be slotted for their crimes? Right,

611
00:32:52,840 --> 00:32:56,560
Speaker 1: And that's where he really comes in. Handy, isn't those Yeah, yeah,

612
00:32:56,920 --> 00:33:05,280
Speaker 1: that guy shouldn't be British. Um. So, so the study, like,

613
00:33:05,360 --> 00:33:07,920
Speaker 1: the reason they conduct this study is in the hopes

614
00:33:08,000 --> 00:33:09,880
Speaker 1: that it will like allow them to say that it's

615
00:33:09,920 --> 00:33:12,520
Speaker 1: not biased. But then it turns out that, like, I

616
00:33:12,520 --> 00:33:16,360
Speaker 1: wouldn't call it biased. But this change, which is unequivocally

617
00:33:16,400 --> 00:33:20,640
Speaker 1: a good thing, impacts conservative sites, which are lower quality

618
00:33:20,680 --> 00:33:22,640
Speaker 1: and more often shared by people who haven't read the

619
00:33:22,720 --> 00:33:25,880
Speaker 1: articles but are incensed by a shitty, aggressive headline like

620
00:33:25,880 --> 00:33:28,240
Speaker 1: the brightbart ones. We just read those get shared a

621
00:33:28,280 --> 00:33:30,560
Speaker 1: lot and they don't read the article, and that's great

622
00:33:30,600 --> 00:33:34,560
Speaker 1: for bright part um. But they decided, like, oh shit, actually,

623
00:33:35,080 --> 00:33:37,640
Speaker 1: this study, the results of this study were absolutely going

624
00:33:37,680 --> 00:33:40,040
Speaker 1: to be called out for bias. One of the researchers

625
00:33:40,080 --> 00:33:42,640
Speaker 1: wrote in an internal memo, we could face significant backlash.

626
00:33:42,640 --> 00:33:46,760
Speaker 1: We're having experimented with distribution at the expensive conservative publishers.

627
00:33:46,800 --> 00:33:49,600
Speaker 1: So the company dumps this plan, they kill it. It

628
00:33:49,760 --> 00:33:51,680
Speaker 1: is bad for Brightbart, good for the world if it's

629
00:33:51,720 --> 00:33:54,000
Speaker 1: bad for bright part. If it's bad for the Bart,

630
00:33:54,800 --> 00:33:56,880
Speaker 1: we gotta we gotta. Can the plan bad for the Bart?

631
00:33:57,160 --> 00:34:00,200
Speaker 1: Can you have always said that Facebook? Yeah, I say

632
00:34:01,080 --> 00:34:03,360
Speaker 1: you are Cheryl Sandberg. Actually not a lot of people

633
00:34:03,400 --> 00:34:09,200
Speaker 1: know that. Um yeah, I you know, listen off of

634
00:34:09,320 --> 00:34:14,520
Speaker 1: making fun of Cheryl's very share eff So the good

635
00:34:14,520 --> 00:34:17,480
Speaker 1: news is that Facebook didn't just make craven decisions when

636
00:34:17,520 --> 00:34:19,920
Speaker 1: threatened with the possibility of being called out for bias.

637
00:34:20,120 --> 00:34:22,960
Speaker 1: They were also craven whenever a feature promised to improve

638
00:34:23,000 --> 00:34:25,759
Speaker 1: the safety of their network at the cost of absolutely

639
00:34:25,800 --> 00:34:29,799
Speaker 1: any profitability. In two thousand nineteen, Facebook researchers open to

640
00:34:29,840 --> 00:34:32,319
Speaker 1: study into the influence of the like button, which is

641
00:34:32,560 --> 00:34:34,480
Speaker 1: one of the most basic and central features of the

642
00:34:34,600 --> 00:34:39,240
Speaker 1: entire platform. Unfortunately, as we're discussing more detail later, likes

643
00:34:39,280 --> 00:34:42,480
Speaker 1: are one of the most toxic things about Facebook. Researchers

644
00:34:42,480 --> 00:34:45,719
Speaker 1: found that removing the like button, along with removing emoji

645
00:34:45,760 --> 00:34:49,920
Speaker 1: reactions from posts on Instagram, reduced quote stress and anxiety

646
00:34:50,040 --> 00:34:52,640
Speaker 1: felt by the youngest users on the platform, who all

647
00:34:52,680 --> 00:34:56,120
Speaker 1: reported significant anxiety due to the feature. But Facebook also

648
00:34:56,120 --> 00:34:58,799
Speaker 1: found that hiding the like button reduced the rate at

649
00:34:58,800 --> 00:35:02,799
Speaker 1: which users interacted with posts and clicked on ads. So

650
00:35:02,840 --> 00:35:05,880
Speaker 1: now this is more of yeah, yeah, no, this is

651
00:35:05,920 --> 00:35:07,200
Speaker 1: I will say, more of a mixed bag than the

652
00:35:07,239 --> 00:35:11,479
Speaker 1: last thing because removing the like didn't like, it made

653
00:35:11,600 --> 00:35:14,319
Speaker 1: one group of young people feel better, um, but not

654
00:35:14,360 --> 00:35:16,560
Speaker 1: other groups of young people like. It didn't reduce it

655
00:35:16,600 --> 00:35:19,400
Speaker 1: reduced like kids social anxiety, but it didn't have as

656
00:35:19,440 --> 00:35:21,799
Speaker 1: much of an impact on teens really, so it's not

657
00:35:21,960 --> 00:35:24,799
Speaker 1: as clear cut as the last one, but still a

658
00:35:24,880 --> 00:35:28,040
Speaker 1: protective effect had been found among the most vulnerable people

659
00:35:28,040 --> 00:35:31,960
Speaker 1: on Instagram in particular. Um, but they don't they don't

660
00:35:32,000 --> 00:35:35,920
Speaker 1: do anything about it because you know, that's so frustrated.

661
00:35:36,360 --> 00:35:41,880
Speaker 1: That is genuinely, very very valuable interesting information where I

662
00:35:41,880 --> 00:35:43,920
Speaker 1: don't know, I mean, I I feel like it probably

663
00:35:44,000 --> 00:35:47,719
Speaker 1: didn't affect teenagers because by that point it's like, I mean,

664
00:35:47,800 --> 00:35:50,759
Speaker 1: you don't want to say like too late, but by

665
00:35:50,760 --> 00:35:53,200
Speaker 1: that point you're so hardwired to be like, well, I

666
00:35:53,239 --> 00:35:56,280
Speaker 1: can tell what is important or like what is worth

667
00:35:56,360 --> 00:35:59,920
Speaker 1: discussing based on likes, and once that that's just such

668
00:35:59,920 --> 00:36:03,600
Speaker 1: a sticky I mean, I still feel that way, even

669
00:36:03,600 --> 00:36:05,600
Speaker 1: though it's like you can objectively know it's not true,

670
00:36:05,680 --> 00:36:08,399
Speaker 1: but once you've been kind of pilled in that way.

671
00:36:08,480 --> 00:36:13,200
Speaker 1: It's it's very hard to undo. Yeah, it's Um, I

672
00:36:13,239 --> 00:36:16,239
Speaker 1: don't know what it is, Jamie, it's not great. Um

673
00:36:16,400 --> 00:36:22,520
Speaker 1: upsetting that. Yeah, it's not great. Uh So, as time

674
00:36:22,520 --> 00:36:25,240
Speaker 1: went on, research made it increasingly clear that core features

675
00:36:25,239 --> 00:36:29,439
Speaker 1: of Facebook products were fundamentally harmful. From BuzzFeed quote, time

676
00:36:29,480 --> 00:36:32,239
Speaker 1: and again, they determine that people misused key features or

677
00:36:32,280 --> 00:36:35,840
Speaker 1: that those features amplified toxic content, among other effects. In

678
00:36:35,880 --> 00:36:39,440
Speaker 1: August and in August two, nineteen internal memos, several researchers

679
00:36:39,480 --> 00:36:42,400
Speaker 1: said it was Facebook's core product mechanics, meaning the basics

680
00:36:42,400 --> 00:36:44,920
Speaker 1: of how the product functioned, that had let misinformation and

681
00:36:44,920 --> 00:36:47,440
Speaker 1: hate speech flourish on the site. The mechanics of our

682
00:36:47,440 --> 00:36:52,319
Speaker 1: platform are not neutral, they concluded. So there's Facebook employees

683
00:36:52,600 --> 00:36:56,120
Speaker 1: recognize internally, we are making decisions that are allowing hatred

684
00:36:56,160 --> 00:36:59,439
Speaker 1: and other and just unhealthy toxic content to spread, and

685
00:37:00,080 --> 00:37:02,640
Speaker 1: we're not. The bias is not in us fighting it.

686
00:37:02,800 --> 00:37:05,200
Speaker 1: Our biases in refusing to fight it, like we are

687
00:37:05,280 --> 00:37:08,320
Speaker 1: not being neutral because we're letting this spread. The people

688
00:37:08,480 --> 00:37:11,279
Speaker 1: making the site work recognize this. They talk about it

689
00:37:11,280 --> 00:37:14,600
Speaker 1: to each other. Um, it makes it. They feel guilt

690
00:37:14,640 --> 00:37:18,439
Speaker 1: over it. They talk about it. You know, we know this. Yeah, yeah,

691
00:37:18,440 --> 00:37:21,440
Speaker 1: I mean we've discussed that before. Of just like the

692
00:37:21,480 --> 00:37:24,960
Speaker 1: existential stress of working at Facebook. Not the most empathetic

693
00:37:25,080 --> 00:37:27,719
Speaker 1: problem in the world, but not at all. A lot

694
00:37:27,760 --> 00:37:31,120
Speaker 1: of them are making bank but clear paper trail though

695
00:37:31,280 --> 00:37:38,480
Speaker 1: of deep existential guilt and distress. Now, uh, it's it's

696
00:37:38,520 --> 00:37:41,719
Speaker 1: pretty cool. Yeah, it's pretty cool. Pretty cool. So, rather

697
00:37:41,760 --> 00:37:45,080
Speaker 1: than expanding their tests on the impact of removing the

698
00:37:45,080 --> 00:37:47,640
Speaker 1: like button on a broader scale, Mark Zuckerberg and other

699
00:37:47,640 --> 00:37:50,480
Speaker 1: executives agreed to allow testing only on a much more

700
00:37:50,520 --> 00:37:53,600
Speaker 1: limited scale, not to reduce harm, but to quote build

701
00:37:53,640 --> 00:37:57,200
Speaker 1: a positive press narrative around Instagram. So not not to

702
00:37:57,280 --> 00:38:00,640
Speaker 1: actually help human beings, but to have us something to

703
00:38:00,680 --> 00:38:04,320
Speaker 1: brag about. Right, for him to be like, we're so nice,

704
00:38:04,680 --> 00:38:07,480
Speaker 1: We're so cool, We're gonna have fucking rad we are

705
00:38:07,840 --> 00:38:12,000
Speaker 1: I'm the guy site to stop the dated slang. He's

706
00:38:12,040 --> 00:38:14,319
Speaker 1: going to be like, yeah, this is gonna be uh,

707
00:38:14,360 --> 00:38:17,680
Speaker 1: we gotta get some lit press around Instagram, know what

708
00:38:17,719 --> 00:38:20,480
Speaker 1: I mean. Yeah, in three years, someone's going to tell

709
00:38:20,520 --> 00:38:22,640
Speaker 1: him the word poggers and then he's gonna say it

710
00:38:22,719 --> 00:38:26,640
Speaker 1: like thirty times in a week to all of his

711
00:38:26,719 --> 00:38:33,960
Speaker 1: imaginary alien friends on meta'd like that's bro sucker, I

712
00:38:34,000 --> 00:38:38,239
Speaker 1: hate that, screaming into avoid. In September of Facebook rolled

713
00:38:38,239 --> 00:38:40,600
Speaker 1: out a study of the share button. This came in

714
00:38:40,640 --> 00:38:43,719
Speaker 1: the wake of a summer of unprecedented political violence, much

715
00:38:43,719 --> 00:38:47,520
Speaker 1: of its stoked via viral Facebook posts and Facebook groups. Uh,

716
00:38:47,520 --> 00:38:49,880
Speaker 1: the ship at Kenosha started on Facebook in a lot

717
00:38:49,920 --> 00:38:52,560
Speaker 1: of ways. Uh, we're tracking it that night. A hell

718
00:38:52,600 --> 00:38:54,520
Speaker 1: of a lot of that ship got started on Facebook.

719
00:38:54,520 --> 00:38:56,400
Speaker 1: A lot of the like let's get a militia together

720
00:38:56,400 --> 00:38:59,880
Speaker 1: and protect businesses. You know, it's good stuff. Company research

721
00:39:00,160 --> 00:39:04,480
Speaker 1: identified reshare aggregation units, which were automatically generated groups of

722
00:39:04,520 --> 00:39:08,040
Speaker 1: posts that you're so they identify one of the problems

723
00:39:08,120 --> 00:39:10,280
Speaker 1: leading to all of this is what they called reshare

724
00:39:10,280 --> 00:39:13,880
Speaker 1: aggregation units. And these were automatically generated groups of posts

725
00:39:13,920 --> 00:39:17,279
Speaker 1: that were sent to you that we're posts your friends liked, right,

726
00:39:17,560 --> 00:39:19,040
Speaker 1: and they recognize this is how a lot of this

727
00:39:19,080 --> 00:39:22,680
Speaker 1: bad shit is spreading so right, right, that's creating a

728
00:39:22,760 --> 00:39:26,440
Speaker 1: feedback loop on purpose. Yes, in part because users are

729
00:39:26,520 --> 00:39:29,200
Speaker 1: much more likely to share posts that have already been

730
00:39:29,239 --> 00:39:31,600
Speaker 1: liked by other people they know, even if those posts

731
00:39:31,640 --> 00:39:34,880
Speaker 1: are hateful, bigoted, bullying, or contain an accurate information. So

732
00:39:34,920 --> 00:39:37,040
Speaker 1: if somebody gets the same post in two different ways,

733
00:39:37,080 --> 00:39:40,359
Speaker 1: if they just like see a bigoted article pop up

734
00:39:40,360 --> 00:39:43,440
Speaker 1: on their on their their Facebook feed, but they're not

735
00:39:43,480 --> 00:39:45,480
Speaker 1: being informed that other people they know have liked it

736
00:39:45,600 --> 00:39:47,799
Speaker 1: or shared it, they're less likely to share it than

737
00:39:47,840 --> 00:39:49,520
Speaker 1: if like, well, my buddy shared it. So maybe now

738
00:39:49,600 --> 00:39:52,040
Speaker 1: I have permission, right, and you can think about how

739
00:39:52,080 --> 00:39:54,399
Speaker 1: this happens on like a societal's level, how this has

740
00:39:54,400 --> 00:39:57,120
Speaker 1: contributed to everything we're dealing with right now. Um, so

741
00:39:57,160 --> 00:40:00,520
Speaker 1: I feel like everyone knows someone who has probably was

742
00:40:00,760 --> 00:40:03,879
Speaker 1: very very influenced by that exact function. That's yeah, God,

743
00:40:03,920 --> 00:40:08,200
Speaker 1: that's awful. So company researchers in September like these research

744
00:40:08,280 --> 00:40:10,160
Speaker 1: aggregations the fact that we don't have to do it

745
00:40:10,200 --> 00:40:12,600
Speaker 1: this way, right, we can only show people that the

746
00:40:12,800 --> 00:40:15,080
Speaker 1: articles that their friends comment on at the very least

747
00:40:15,120 --> 00:40:17,960
Speaker 1: as supposed to just like or just share without much commentary. Like,

748
00:40:18,160 --> 00:40:20,160
Speaker 1: they have a number of options here that could at

749
00:40:20,160 --> 00:40:24,799
Speaker 1: the very least reduce um um harmful content because when

750
00:40:24,800 --> 00:40:27,480
Speaker 1: every like I think the numbers like three times as

751
00:40:27,560 --> 00:40:30,240
Speaker 1: likely to share content that's presented to them in this way.

752
00:40:30,520 --> 00:40:33,400
Speaker 1: So in May of that year, um, while you know,

753
00:40:34,400 --> 00:40:37,759
Speaker 1: so this is actually months before Facebook researchers find this out.

754
00:40:38,080 --> 00:40:41,440
Speaker 1: Myself and a journalist named Jason Wilson published what I

755
00:40:41,480 --> 00:40:44,080
Speaker 1: still think is the first proper forensic investigation into the

756
00:40:44,120 --> 00:40:46,840
Speaker 1: Boogaloo movement. Um. It noted how the spread of the

757
00:40:46,880 --> 00:40:49,680
Speaker 1: movement and its crucial early days was enabled entirely by

758
00:40:49,680 --> 00:40:53,240
Speaker 1: Facebook's group recommendation algorithm, which was often spread to people

759
00:40:53,280 --> 00:40:56,240
Speaker 1: by these reshare aggregation units. You see, oh my buddy

760
00:40:56,320 --> 00:40:59,040
Speaker 1: joined this group where everybody's sharing these like Hawaiian shirt

761
00:40:59,040 --> 00:41:01,719
Speaker 1: photos and pictures of maybe I'll hop in there, and

762
00:41:01,840 --> 00:41:04,520
Speaker 1: you know, the cycle goes on from there. When you've

763
00:41:04,560 --> 00:41:06,799
Speaker 1: joined one group, it sends you advice like, hey, check

764
00:41:06,840 --> 00:41:08,520
Speaker 1: out this other group, check out this other group, and

765
00:41:08,560 --> 00:41:12,000
Speaker 1: it starts with like we're sharing memes about like Hawaiian

766
00:41:12,120 --> 00:41:15,560
Speaker 1: shirts and and you know, the boogaloo. And then five

767
00:41:15,680 --> 00:41:18,520
Speaker 1: or six groups down the line, you're making serious plans

768
00:41:18,560 --> 00:41:23,239
Speaker 1: to assassinate federal agents or kidnap a governor. You know, yeah,

769
00:41:23,320 --> 00:41:27,640
Speaker 1: I mean that he's uh, you know, I remember where

770
00:41:27,680 --> 00:41:30,040
Speaker 1: I was when I read it, because the how steep

771
00:41:30,120 --> 00:41:34,399
Speaker 1: the escalation is and how quickly it like, it's not

772
00:41:34,520 --> 00:41:36,600
Speaker 1: I guess not completely surprising, but at the time I

773
00:41:36,640 --> 00:41:39,320
Speaker 1: was like, oh, that that is a very short timeline

774
00:41:39,480 --> 00:41:44,160
Speaker 1: from yeah, so I don't like it. It's not good.

775
00:41:44,280 --> 00:41:48,080
Speaker 1: It's not good, and in fairness, um, there are actually

776
00:41:48,120 --> 00:41:51,959
Speaker 1: some face like it kind of becomes accepted the stuff

777
00:41:52,000 --> 00:41:54,400
Speaker 1: that Jason and I were writing about in May by

778
00:41:54,440 --> 00:41:57,000
Speaker 1: a lot of Facebook researchers around September of that year.

779
00:41:57,000 --> 00:41:59,560
Speaker 1: But there were people within Facebook who actually tried to

780
00:41:59,600 --> 00:42:02,720
Speaker 1: blow the it on this earlier um in fact, significantly earlier.

781
00:42:03,120 --> 00:42:06,200
Speaker 1: In mid two thousand nineteen, an internal researcher created an

782
00:42:06,280 --> 00:42:09,880
Speaker 1: experimental Facebook account, which is something that like certain researchers

783
00:42:09,880 --> 00:42:11,319
Speaker 1: would do from time to time to see what the

784
00:42:11,320 --> 00:42:15,120
Speaker 1: algorithm is feeding people. This experimental account account was a

785
00:42:15,160 --> 00:42:18,319
Speaker 1: fake conservative mom, and they made this acount because they

786
00:42:18,320 --> 00:42:20,800
Speaker 1: wanted to see what the recommendation algorithm would feed this account.

787
00:42:20,960 --> 00:42:23,560
Speaker 1: And I'm gonna read from BuzzFeed again here. The internal research,

788
00:42:23,640 --> 00:42:26,600
Speaker 1: titled Carol's Journey to Q and On, detailed how the

789
00:42:26,600 --> 00:42:29,400
Speaker 1: Facebook account for an imaginary woman named Carol Smith had

790
00:42:29,440 --> 00:42:33,160
Speaker 1: followed pages for Fox News and Sinclair Broadcasting. Within days,

791
00:42:33,200 --> 00:42:36,200
Speaker 1: Facebook had recommended pages and groups related to Q and On,

792
00:42:36,360 --> 00:42:39,440
Speaker 1: the conspiracy theory that falsely claimed Mr Trump was facing

793
00:42:39,440 --> 00:42:42,560
Speaker 1: down a shadowy cabal of democratic pedophiles. By the end

794
00:42:42,600 --> 00:42:45,320
Speaker 1: of three weeks, Carol Smith's Facebook feed had devolved further.

795
00:42:45,560 --> 00:42:48,279
Speaker 1: It became a constant flow of misleading, polarizing, and low

796
00:42:48,360 --> 00:42:53,120
Speaker 1: quality content. The researcher wrote, Yeah, how did so so

797
00:42:53,239 --> 00:42:56,000
Speaker 1: some some some jerk was like, let's call it Carol,

798
00:42:56,440 --> 00:43:02,160
Speaker 1: like they Carol stereotype. Statistically not unlikely that it could

799
00:43:02,360 --> 00:43:04,439
Speaker 1: It could have been a Carol that is. I mean

800
00:43:04,840 --> 00:43:08,240
Speaker 1: that that's interesting that we also all know a Carol.

801
00:43:09,719 --> 00:43:14,040
Speaker 1: A Carol. Yeah, unfortunately. So they're in They're in the

802
00:43:14,040 --> 00:43:17,479
Speaker 1: they're the Duncan Donuts drive through. I I walk among them. Yeah,

803
00:43:18,480 --> 00:43:22,200
Speaker 1: they live among us. They are dating you walk among them.

804
00:43:22,320 --> 00:43:28,080
Speaker 1: That's so funny. You eat hot dogs next to all people. Yeah,

805
00:43:28,120 --> 00:43:31,359
Speaker 1: we're in mine. I think that hot dog eaters are

806
00:43:31,400 --> 00:43:36,560
Speaker 1: maybe more politically astute bunch. But the Dunkan Donuts line

807
00:43:36,680 --> 00:43:41,320
Speaker 1: is just absolutely unmitigated chaos. There could be the politics

808
00:43:41,320 --> 00:43:43,480
Speaker 1: of the Dunkey Donuts line are all over the place.

809
00:43:43,800 --> 00:43:46,319
Speaker 1: They are they are anarchists. In the line. You have

810
00:43:46,719 --> 00:43:48,600
Speaker 1: you have the scariest people you've ever met in the line.

811
00:43:49,000 --> 00:43:51,200
Speaker 1: You have been Affleck in the line looking like his

812
00:43:51,400 --> 00:43:54,640
Speaker 1: entire family just died in a bus crash, but Afflex

813
00:43:54,960 --> 00:43:58,240
Speaker 1: in the line and you can see his little dragon

814
00:43:58,280 --> 00:44:01,960
Speaker 1: back piece. Oh my god. So Phoenix, Jamie, come on,

815
00:44:02,280 --> 00:44:05,239
Speaker 1: come on, sorry, that was disrespectful. That was disrespectful. And

816
00:44:05,280 --> 00:44:09,000
Speaker 1: you're right and you're right, thank you, thank you. So

817
00:44:10,160 --> 00:44:15,000
Speaker 1: this this study with this fake account that immediately gets radicalized.

818
00:44:15,400 --> 00:44:18,800
Speaker 1: This study, uh we it comes out in the Facebook

819
00:44:18,800 --> 00:44:20,719
Speaker 1: papers right this year, but it was done in two

820
00:44:20,760 --> 00:44:23,960
Speaker 1: thousand nineteen. And when this year, like information of this

821
00:44:24,040 --> 00:44:27,359
Speaker 1: dropped and journalists start writing about it, um, they do

822
00:44:27,400 --> 00:44:29,680
Speaker 1: what journalists do, which is you you, you put together

823
00:44:29,719 --> 00:44:31,840
Speaker 1: your article and then you go for comment. Right. And

824
00:44:31,920 --> 00:44:35,520
Speaker 1: so the comment that Facebook made about this experiment that

825
00:44:35,560 --> 00:44:38,320
Speaker 1: this researcher didn't two thousand nineteen was well, this was

826
00:44:38,360 --> 00:44:40,600
Speaker 1: a study of one hypothetical user. It is a perfect

827
00:44:40,640 --> 00:44:43,040
Speaker 1: example of research the company does to improve improve our

828
00:44:43,080 --> 00:44:45,520
Speaker 1: systems and helped inform our decision to remove Q and

829
00:44:45,600 --> 00:44:48,840
Speaker 1: on from the platform. That did not happen until January

830
00:44:48,880 --> 00:44:53,359
Speaker 1: of this year. They didn't ship for two years after this.

831
00:44:53,880 --> 00:44:56,280
Speaker 1: They only did ship because people stormed the fucking capital

832
00:44:56,280 --> 00:45:00,480
Speaker 1: waving Q and on banners. You motherfucker's sounds like them,

833
00:45:01,160 --> 00:45:03,160
Speaker 1: sound like that. They're like, Oh, let's wait until things

834
00:45:03,239 --> 00:45:06,680
Speaker 1: get so desperately bad that the company will be, you know,

835
00:45:06,760 --> 00:45:10,600
Speaker 1: severely impacted if we don't do something. The huge amount

836
00:45:10,680 --> 00:45:13,680
Speaker 1: of the of the radicalization Q and on gets supercharged

837
00:45:13,719 --> 00:45:16,040
Speaker 1: by the lockdowns, right because suddenly all these people, like

838
00:45:16,120 --> 00:45:18,239
Speaker 1: a lot of them, are in financial distress, they're locked

839
00:45:18,239 --> 00:45:20,680
Speaker 1: in their fucking houses, they're online all the goddamn time.

840
00:45:21,120 --> 00:45:23,880
Speaker 1: And they knew they could have dealt with this problem

841
00:45:24,000 --> 00:45:27,360
Speaker 1: and reduced massively the number of people who got fed

842
00:45:27,400 --> 00:45:30,120
Speaker 1: this poison during the lockdown. If they've done a goddamn

843
00:45:30,160 --> 00:45:31,920
Speaker 1: thing and do that's a nineteen. They had the option,

844
00:45:32,280 --> 00:45:38,000
Speaker 1: they did not. Yeah, okay, so no surprises here that

845
00:45:38,160 --> 00:45:42,640
Speaker 1: researcher side. Nothing bad happened, right, It did not name

846
00:45:42,680 --> 00:45:47,200
Speaker 1: one did nothing happened? Well, all happened actually and was

847
00:45:47,200 --> 00:45:50,920
Speaker 1: pretty heavily tied to this. Uh. In August of that

848
00:45:51,000 --> 00:45:54,040
Speaker 1: researcher left the company. She wrote an exit note where

849
00:45:54,040 --> 00:45:58,360
Speaker 1: she accused the company Facebook of quote knowingly exposing users

850
00:45:58,360 --> 00:46:01,080
Speaker 1: to harm. We've known for over a year now that

851
00:46:01,120 --> 00:46:04,120
Speaker 1: our recommendation systems can very quickly lead users down a

852
00:46:04,160 --> 00:46:06,920
Speaker 1: path to conspiracy theories and groups. In the meantime, the

853
00:46:06,960 --> 00:46:09,719
Speaker 1: fringe groups slash set of beliefs has has grown to

854
00:46:09,840 --> 00:46:12,960
Speaker 1: national prominence, with q and on congressional candidates and kean

855
00:46:13,000 --> 00:46:16,760
Speaker 1: on hashtags and groups trending on the mainstream. Out of fears,

856
00:46:17,200 --> 00:46:20,839
Speaker 1: out of fears over potential public and policy stakeholder responses,

857
00:46:21,080 --> 00:46:24,600
Speaker 1: we are knowingly exposing users to risks of integrity harms.

858
00:46:24,960 --> 00:46:27,359
Speaker 1: During the time that we've hesitated, I've seen folks from

859
00:46:27,360 --> 00:46:29,759
Speaker 1: my hometown go further and further down the rabbit hole.

860
00:46:29,960 --> 00:46:35,040
Speaker 1: It has been painful to observe. Okay, okay, and I

861
00:46:35,040 --> 00:46:39,920
Speaker 1: mean it is I mean, no arguments there. It is

862
00:46:40,040 --> 00:46:43,399
Speaker 1: very painful to observe that happen to people who are

863
00:46:43,440 --> 00:46:46,960
Speaker 1: not Yeah, that's a Facebook employee who's not going to

864
00:46:47,040 --> 00:46:50,360
Speaker 1: get any ship from me. Um. She identified the problem,

865
00:46:50,400 --> 00:46:53,839
Speaker 1: she did good research to try to make clear what

866
00:46:53,880 --> 00:46:58,160
Speaker 1: the problem was, and when she realized that her company

867
00:46:58,239 --> 00:46:59,920
Speaker 1: was never going to take action on this because it

868
00:47:00,040 --> 00:47:02,880
Speaker 1: to produce their profits, she fucking leaves. And she she

869
00:47:03,520 --> 00:47:06,320
Speaker 1: does everything she can to let people know how unacceptable

870
00:47:06,360 --> 00:47:10,160
Speaker 1: the situation is within the company. You know, I mean

871
00:47:10,160 --> 00:47:12,960
Speaker 1: that is good that's what that is the minimum. That

872
00:47:13,120 --> 00:47:15,520
Speaker 1: is the minimum, right, Yeah, I mean it is a

873
00:47:15,520 --> 00:47:18,320
Speaker 1: little silly to be like, and I just recently realized

874
00:47:18,360 --> 00:47:21,600
Speaker 1: that I don't think Facebook is ethical. You're like, like,

875
00:47:21,840 --> 00:47:23,719
Speaker 1: shut up, no way, I don't. I don't know when

876
00:47:23,719 --> 00:47:26,959
Speaker 1: this person started, but like, yeah, there they she got

877
00:47:27,040 --> 00:47:31,880
Speaker 1: there and she's clearly horrified by what she like realized

878
00:47:31,920 --> 00:47:34,600
Speaker 1: the company was doing. Like again, we've all been where

879
00:47:34,600 --> 00:47:36,319
Speaker 1: she is. Where you just see these people you grew

880
00:47:36,400 --> 00:47:41,359
Speaker 1: up with lose their goddamn minds and it's bad. It's bad.

881
00:47:41,400 --> 00:47:43,680
Speaker 1: And you're like, oh and I'm I'm and I work

882
00:47:43,719 --> 00:47:46,960
Speaker 1: in nucleus at the problem of the problem. Interesting. Yeah,

883
00:47:47,160 --> 00:47:48,960
Speaker 1: I mean that's why I had to quit working at

884
00:47:48,960 --> 00:47:53,239
Speaker 1: Preduce Pharmaceuticals. Yeah. I do miss the freedom, la. I know,

885
00:47:53,440 --> 00:47:55,360
Speaker 1: I know, I know, I know all those I was

886
00:47:55,400 --> 00:48:01,600
Speaker 1: a great salesman. You're so good plays a produced pharmaceuticals salesman.

887
00:48:02,360 --> 00:48:05,880
Speaker 1: Or no, maybe it's that Will Poulter. Oh yeah, okay.

888
00:48:06,200 --> 00:48:08,160
Speaker 1: You know who didn't is Alec Ginnis. Because he never

889
00:48:08,200 --> 00:48:10,799
Speaker 1: loved He never lived to see opiate's become what they

890
00:48:10,840 --> 00:48:15,440
Speaker 1: are today. Tragic. He never lived to taste the sweet

891
00:48:15,480 --> 00:48:18,319
Speaker 1: taste of tramadol or delatted and we don't talk about

892
00:48:18,320 --> 00:48:20,720
Speaker 1: that enough. We don't talk about that enough. What a shame?

893
00:48:21,320 --> 00:48:25,440
Speaker 1: What heartbreaks? Say I will what if Alec Ginnis had

894
00:48:25,440 --> 00:48:30,920
Speaker 1: access to high quality, pharmaceutical grade painkillers an essay? I

895
00:48:30,960 --> 00:48:33,560
Speaker 1: think it would have been sweet. I'm pitching it. I'm

896
00:48:33,600 --> 00:48:39,120
Speaker 1: pitching it. Okay, Um, someone who's better at movies than

897
00:48:39,160 --> 00:48:41,600
Speaker 1: I could have made a train spotting joke there because

898
00:48:41,640 --> 00:48:45,160
Speaker 1: he and McGregor in the Heroin movie than he played on.

899
00:48:45,440 --> 00:48:47,239
Speaker 1: I don't know, there's some joke there, but I didn't

900
00:48:47,239 --> 00:48:49,239
Speaker 1: come up with it. Okay. Someone Yeah, someone figure that

901
00:48:49,239 --> 00:48:51,239
Speaker 1: out in the reddit, and then don't tell us about it.

902
00:48:51,360 --> 00:48:53,440
Speaker 1: Do not tell us because I've never seen train spotting.

903
00:48:53,440 --> 00:48:57,600
Speaker 1: I'm just aware that woggin. Yeah, I haven't seen it,

904
00:48:57,680 --> 00:49:00,480
Speaker 1: but like, you know what it's about. I'm so really yeah,

905
00:49:00,480 --> 00:49:03,560
Speaker 1: I know what it's about. I also get so stressed

906
00:49:03,560 --> 00:49:06,799
Speaker 1: out when I anytime it's not often, but anytime I've

907
00:49:06,800 --> 00:49:09,680
Speaker 1: had to say even McGregor's name, I say, it's so weird.

908
00:49:09,920 --> 00:49:12,480
Speaker 1: It's the worst thing in the world. Saying when McGregor's

909
00:49:12,560 --> 00:49:14,759
Speaker 1: name is the most frightening experience a human can have.

910
00:49:16,000 --> 00:49:17,960
Speaker 1: I can't. I can't make my mouth make that shape.

911
00:49:17,960 --> 00:49:20,399
Speaker 1: It's embarrassing. I think what he has to live with.

912
00:49:20,560 --> 00:49:23,279
Speaker 1: Thankfully he's gorgeous. That must make it easier. Yeah, I

913
00:49:23,280 --> 00:49:29,400
Speaker 1: mean being sexy has to help. It has to help. Yeah.

914
00:49:29,600 --> 00:49:32,400
Speaker 1: You know who does know how to pronounce Ewan McGregor's

915
00:49:32,480 --> 00:49:35,080
Speaker 1: name and never feels any anxiety over it because their

916
00:49:35,120 --> 00:49:37,279
Speaker 1: friends they hang out on the weekend? Oh nice? Who

917
00:49:37,400 --> 00:49:40,480
Speaker 1: who's at the products and services that support this podcast

918
00:49:40,760 --> 00:49:44,319
Speaker 1: are all good buddies with Ewan McGregor hangs out with

919
00:49:44,320 --> 00:50:01,720
Speaker 1: the highway patrol. Ah, we are back. As the election

920
00:50:01,760 --> 00:50:05,000
Speaker 1: grew nearer, disinformation continued to proliferate on Facebook, and the

921
00:50:05,040 --> 00:50:08,600
Speaker 1: political temperature in the United States rose ever higher. Facebook

922
00:50:08,640 --> 00:50:11,839
Speaker 1: employees grew concerned about the wide variety of worst case

923
00:50:11,840 --> 00:50:14,840
Speaker 1: scenarios that might result if something went wrong with the election.

924
00:50:15,239 --> 00:50:18,760
Speaker 1: They put together a series of emergency break glass measures.

925
00:50:19,040 --> 00:50:21,640
Speaker 1: These would allow them to automatically slow or stop the

926
00:50:21,680 --> 00:50:24,759
Speaker 1: formation of new Facebook groups if the election was contested.

927
00:50:25,200 --> 00:50:27,560
Speaker 1: This was never stated, but you get the feeling. They

928
00:50:27,600 --> 00:50:30,040
Speaker 1: were looking at Kenosha and how Facebook groups had led

929
00:50:30,080 --> 00:50:33,000
Speaker 1: to spontaneous and deadly militia groups to form up due

930
00:50:33,040 --> 00:50:36,080
Speaker 1: to viral news stories. My interpretation is that they were

931
00:50:36,160 --> 00:50:38,160
Speaker 1: terrified of the same sort of phenomenon that it would

932
00:50:38,200 --> 00:50:40,799
Speaker 1: lead to Facebook like fueling a civil war. Like I

933
00:50:40,800 --> 00:50:43,040
Speaker 1: think that we're literally worried that like something will go wrong,

934
00:50:43,320 --> 00:50:45,680
Speaker 1: people will form militias on Facebook and there will be

935
00:50:45,680 --> 00:50:47,520
Speaker 1: a gun fight that a shipload of people die, and

936
00:50:47,560 --> 00:50:49,400
Speaker 1: that escalates to something worse, and everyone will say it

937
00:50:49,440 --> 00:50:52,120
Speaker 1: started on Facebook because that happened in Kenosha, Like, this

938
00:50:52,160 --> 00:50:54,640
Speaker 1: is not a thing. It happened with the Boololoo stuff

939
00:50:54,640 --> 00:50:57,640
Speaker 1: after Yeah, it happened several times last year at a

940
00:50:57,760 --> 00:51:01,040
Speaker 1: fair anxiety. We've seen it happen. Yeah, it was never

941
00:51:01,120 --> 00:51:05,000
Speaker 1: a massive exchange of gunfire, thank fucking god. Um, but

942
00:51:05,040 --> 00:51:07,239
Speaker 1: they were they thought they saw that possibility. This is

943
00:51:07,280 --> 00:51:11,360
Speaker 1: the thing. I was worried about the entirety of um um,

944
00:51:11,400 --> 00:51:13,680
Speaker 1: and we got really close to it several times. I

945
00:51:13,719 --> 00:51:16,640
Speaker 1: was there for a few of them. It sucked. UM.

946
00:51:16,680 --> 00:51:18,719
Speaker 1: So they're worried about this and they start coming. They

947
00:51:18,760 --> 00:51:21,400
Speaker 1: try to figure out like break like emergency measures they

948
00:51:21,400 --> 00:51:24,279
Speaker 1: can take to basically like shut all that ship down,

949
00:51:24,400 --> 00:51:27,000
Speaker 1: like stop people from joining and making new Facebook groups.

950
00:51:27,000 --> 00:51:30,040
Speaker 1: If they have to write if like it becomes obvious

951
00:51:30,080 --> 00:51:33,640
Speaker 1: that something needs to be done. Um and yeah, so

952
00:51:33,920 --> 00:51:37,800
Speaker 1: they they uh yeah. In September, Mark Zuckerberg wrote on

953
00:51:37,840 --> 00:51:40,840
Speaker 1: an internal company post that his company had quote and

954
00:51:41,040 --> 00:51:45,960
Speaker 1: obligator had quote, a responsibility to protect our democracy. He

955
00:51:46,000 --> 00:51:49,120
Speaker 1: bragged about a voter registration campaign the social network and funded,

956
00:51:49,160 --> 00:51:52,440
Speaker 1: and claimed they'd taken vigorous steps to eliminate voter misinformation

957
00:51:52,480 --> 00:51:55,319
Speaker 1: and block political ads, all with the stated goal of

958
00:51:55,360 --> 00:51:59,399
Speaker 1: reducing the chances of violence and unrest. The election came,

959
00:51:59,600 --> 00:52:02,200
Speaker 1: and it it all right. From Facebook's perspective, the whole

960
00:52:02,239 --> 00:52:05,080
Speaker 1: situation was too fluid and confusing in those early days

961
00:52:05,080 --> 00:52:07,000
Speaker 1: after the election. You know where we're getting the counts in.

962
00:52:07,400 --> 00:52:09,680
Speaker 1: Everything's down to the fucking wire. It was all too

963
00:52:09,719 --> 00:52:11,319
Speaker 1: messy for there to be much in the way of

964
00:52:11,400 --> 00:52:14,399
Speaker 1: violent on the ground action to occur. Wow, like that

965
00:52:14,520 --> 00:52:17,000
Speaker 1: was getting sorted out because people just didn't know where's

966
00:52:17,000 --> 00:52:19,759
Speaker 1: it going to land. Um, so they think, oh, we

967
00:52:19,880 --> 00:52:25,200
Speaker 1: dodged a bullet. Everything was fine because they're dumb. Oh baby. Yeah.

968
00:52:25,719 --> 00:52:28,800
Speaker 1: The reality, of course, is that misinformation about election integrity

969
00:52:28,840 --> 00:52:33,280
Speaker 1: spread immediately like wildfire. On November five, one Facebook employee

970
00:52:33,280 --> 00:52:36,319
Speaker 1: warn't colleagues that disinformation was filling the comments section of

971
00:52:36,360 --> 00:52:39,760
Speaker 1: news posts about the election. The most incendiary and dangerous

972
00:52:39,760 --> 00:52:43,520
Speaker 1: comments were being amplified by Facebook's algorithm to appear as

973
00:52:43,520 --> 00:52:47,040
Speaker 1: the top comment on popular threads. On November nine, a

974
00:52:47,120 --> 00:52:50,399
Speaker 1: Facebook data scientists reported in an internal study that one

975
00:52:50,400 --> 00:52:53,200
Speaker 1: out of every fifty views on Facebook in the United States,

976
00:52:53,360 --> 00:52:56,040
Speaker 1: and fully ten percent of all views of political information

977
00:52:56,080 --> 00:52:59,200
Speaker 1: on the site, was content claiming the election had been stolen.

978
00:52:59,480 --> 00:53:03,680
Speaker 1: Temper into Facebook's political posts are the election was stolen? Yeah,

979
00:53:03,920 --> 00:53:06,319
Speaker 1: one out of fifty views on Facebook is someone saying

980
00:53:06,360 --> 00:53:11,080
Speaker 1: the election stolen. This ship is out of controlsutably, everyone

981
00:53:11,360 --> 00:53:15,399
Speaker 1: engaging to agree. Wow, Okay, I honestly, honestly, I would

982
00:53:15,440 --> 00:53:17,839
Speaker 1: have guessed that it would have been higher. But but

983
00:53:17,960 --> 00:53:21,359
Speaker 1: one intent is still there's a lot going on on Facebook. Yeah.

984
00:53:21,480 --> 00:53:25,360
Speaker 1: The researcher noted there was also a fringe incitement to violence.

985
00:53:25,440 --> 00:53:27,719
Speaker 1: And I would quibble over the word fringe because I

986
00:53:27,760 --> 00:53:31,000
Speaker 1: don't think it was very fringe. Um, But otherwise the

987
00:53:31,040 --> 00:53:33,960
Speaker 1: data is interesting, you know, like, there's a lot of

988
00:53:34,000 --> 00:53:37,040
Speaker 1: ten people. It's not the fringe. He's saying. The incitement

989
00:53:37,080 --> 00:53:39,560
Speaker 1: to violence was a fringe of the posts claiming the

990
00:53:39,560 --> 00:53:44,000
Speaker 1: election had been stolen. I disagree with his interpretation of

991
00:53:44,040 --> 00:53:46,480
Speaker 1: that based on my own amount of time that I

992
00:53:46,520 --> 00:53:49,160
Speaker 1: spent looking through these same posts. But whatever, maybe we're

993
00:53:49,160 --> 00:53:51,359
Speaker 1: looking at different posts. You know, there's a lot going

994
00:53:51,400 --> 00:53:55,080
Speaker 1: on on Facebook. In those days. Facebook did not Yeah,

995
00:53:55,280 --> 00:53:57,760
Speaker 1: Facebook did not blow the whistle or sound the alarm

996
00:53:57,880 --> 00:54:00,880
Speaker 1: or do anything but start canceling. It's some agency procedures.

997
00:54:00,880 --> 00:54:03,560
Speaker 1: They were like, we get it. The critical periods over.

998
00:54:03,600 --> 00:54:08,319
Speaker 1: Everything's gonna be fine and dandy, baby. Um. They thought

999
00:54:08,360 --> 00:54:10,600
Speaker 1: the danger of post election violence was over. And most

1000
00:54:10,680 --> 00:54:12,279
Speaker 1: of all, they thought that if they took action to

1001
00:54:12,320 --> 00:54:15,960
Speaker 1: stop the reach of far right propaganda users, then conservatives

1002
00:54:16,000 --> 00:54:19,520
Speaker 1: would complain. As we now know, the most consequential species

1003
00:54:19,520 --> 00:54:22,440
Speaker 1: of disinformation after November two would be the Stop the

1004
00:54:22,480 --> 00:54:25,680
Speaker 1: Steel movement. The idea behind the campaign had its origins

1005
00:54:25,680 --> 00:54:30,080
Speaker 1: in the election, as essentially a fundraising grift from Roger Stone.

1006
00:54:30,160 --> 00:54:34,239
Speaker 1: Ali Alexander, who is a ship head, adapted it in

1007
00:54:34,280 --> 00:54:36,680
Speaker 1: the wake of the election, and it wound up being

1008
00:54:36,719 --> 00:54:40,200
Speaker 1: a major inspiration for the January sixth Capital riot. As

1009
00:54:40,239 --> 00:54:42,920
Speaker 1: we now know from a secret internal report, Facebook was

1010
00:54:42,960 --> 00:54:45,280
Speaker 1: aware of the Stop the Steel movement from the beginning.

1011
00:54:45,440 --> 00:54:48,920
Speaker 1: Quote from Facebook. The first stop the Steel group emerged

1012
00:54:48,920 --> 00:54:51,239
Speaker 1: on election night. It was flagged for escalation because it

1013
00:54:51,280 --> 00:54:54,000
Speaker 1: contained high levels of hate and violence and incitement V

1014
00:54:54,160 --> 00:54:56,800
Speaker 1: and I and the comments. The group was disabled and

1015
00:54:56,840 --> 00:54:59,440
Speaker 1: an investigation was kicked off looking for early signs of

1016
00:54:59,440 --> 00:55:02,240
Speaker 1: coordination and harm across the news stop the Steel groups

1017
00:55:02,360 --> 00:55:04,799
Speaker 1: that were quickly sprouting up to replace it. With our

1018
00:55:04,840 --> 00:55:07,800
Speaker 1: early signals, it was unclear that coordination was taking place

1019
00:55:07,920 --> 00:55:10,760
Speaker 1: or that there was enough harm to constitute designating the term.

1020
00:55:10,800 --> 00:55:13,120
Speaker 1: It wasn't until later that it became clear just how

1021
00:55:13,200 --> 00:55:15,480
Speaker 1: much of a focal point the catchphrase would be, and

1022
00:55:15,520 --> 00:55:17,560
Speaker 1: that they would serve as a rallying point around which

1023
00:55:17,600 --> 00:55:22,440
Speaker 1: a movement of violent election delegitimization could coalesce. From the

1024
00:55:22,440 --> 00:55:24,719
Speaker 1: earliest groups, we saw high levels of hate, v and

1025
00:55:24,800 --> 00:55:28,480
Speaker 1: I and de legitimization combined with meteoric growth rates. Almost

1026
00:55:28,520 --> 00:55:31,080
Speaker 1: all of the fastest growing Facebook groups were stopped the

1027
00:55:31,080 --> 00:55:33,719
Speaker 1: Steel during their peak growth. Because we were looking at

1028
00:55:33,719 --> 00:55:36,879
Speaker 1: each entity individually rather than as a cohesive movement, we

1029
00:55:36,880 --> 00:55:39,440
Speaker 1: were only able to take down individual groups and pages

1030
00:55:39,480 --> 00:55:42,600
Speaker 1: once they exceeded a violation threshold. We were not able

1031
00:55:42,640 --> 00:55:45,399
Speaker 1: to act on simple objects like posts and comments because

1032
00:55:45,440 --> 00:55:48,279
Speaker 1: they individually tended not to violate even if they were

1033
00:55:48,320 --> 00:55:52,880
Speaker 1: surrounded by hate, violence and misinformation after the Capital insurrection. Yeah,

1034
00:55:53,120 --> 00:55:55,920
Speaker 1: that is such garbage. I mean, it's like, and I

1035
00:55:56,000 --> 00:55:58,439
Speaker 1: know that you have examined far more of these posts

1036
00:55:58,440 --> 00:56:01,240
Speaker 1: than I have in depth, but it's just the fast

1037
00:56:01,280 --> 00:56:04,799
Speaker 1: and looseness that people interpret like just it's like free

1038
00:56:04,840 --> 00:56:07,960
Speaker 1: form jazz of the way people interpret Facebook community rules.

1039
00:56:08,120 --> 00:56:12,680
Speaker 1: Because I feel like in groups like that and in

1040
00:56:12,719 --> 00:56:17,239
Speaker 1: groups like less inflammatory than that, there is just constant

1041
00:56:17,280 --> 00:56:21,920
Speaker 1: breaking of the Facebook community rules. It's just uh, it's yeah, yeah, yeah,

1042
00:56:21,960 --> 00:56:26,759
Speaker 1: they're not really moderated at all. Um. Yeah, So this

1043
00:56:26,840 --> 00:56:28,440
Speaker 1: is interesting to me for a few reasons. For one,

1044
00:56:28,480 --> 00:56:30,560
Speaker 1: it lays out what I suspect is the case these

1045
00:56:30,600 --> 00:56:33,439
Speaker 1: researchers and Facebook employees needed to believe and be able

1046
00:56:33,440 --> 00:56:36,279
Speaker 1: to argue in order to not hate themselves. Um, the

1047
00:56:36,360 --> 00:56:39,279
Speaker 1: idea that like, we just didn't recognize it was coordinated.

1048
00:56:39,320 --> 00:56:40,719
Speaker 1: We thought it was all it was. It was all

1049
00:56:40,800 --> 00:56:44,640
Speaker 1: kind of grassroots and happening like organically, and so it

1050
00:56:44,680 --> 00:56:46,719
Speaker 1: was much more complicated for us to try to deal with.

1051
00:56:47,080 --> 00:56:49,719
Speaker 1: I think they need to believe this. I'm going to

1052
00:56:49,760 --> 00:56:52,879
Speaker 1: explain why it's not a good excuse. Starting in December two,

1053
00:56:53,120 --> 00:56:56,880
Speaker 1: nineteen and going until May, the boogleoo movement expanded rapidly

1054
00:56:56,960 --> 00:57:00,920
Speaker 1: in Facebook groups. Incitements to violence semi regularly got groups nooped,

1055
00:57:00,960 --> 00:57:03,680
Speaker 1: and members adapted new terms in order to avoid getting

1056
00:57:03,719 --> 00:57:06,640
Speaker 1: deep platformed. It became gradually obvious that a number of

1057
00:57:06,640 --> 00:57:09,440
Speaker 1: these groups were cohesive and connected, and this was revealed

1058
00:57:09,440 --> 00:57:11,719
Speaker 1: throughout the year in a string of terrorist attacks by

1059
00:57:11,719 --> 00:57:15,160
Speaker 1: boogleoo types in multiple states. When these attacks began, Facebook

1060
00:57:15,160 --> 00:57:17,760
Speaker 1: engaged in a much more cohesive and effective campaign to

1061
00:57:17,800 --> 00:57:21,240
Speaker 1: ban boogaloo groups. The Boogaloo movement and Stop the Steel

1062
00:57:21,280 --> 00:57:24,080
Speaker 1: are of course not one to one analogs, but the

1063
00:57:24,080 --> 00:57:26,720
Speaker 1: fact that this occurred earlier in the same year, resulting

1064
00:57:26,720 --> 00:57:29,840
Speaker 1: in deaths and widespread violence, shows that Fate's book fucking

1065
00:57:29,920 --> 00:57:32,760
Speaker 1: knew the stakes. They could have recognized what was going

1066
00:57:32,800 --> 00:57:35,080
Speaker 1: on with the Stop the Steel movement earlier, and they

1067
00:57:35,120 --> 00:57:37,680
Speaker 1: could have recognized that it was much likely more cohesive

1068
00:57:37,720 --> 00:57:40,360
Speaker 1: than it may have seemed. A decision was made not

1069
00:57:40,480 --> 00:57:42,240
Speaker 1: to do this, not to act on what they had

1070
00:57:42,280 --> 00:57:45,200
Speaker 1: learned earlier that year, and I would argue, based on

1071
00:57:45,240 --> 00:57:47,640
Speaker 1: everything else, we know that the reason why this decision

1072
00:57:47,680 --> 00:57:50,720
Speaker 1: was made was primarily political, like they didn't want to

1073
00:57:50,720 --> 00:57:54,360
Speaker 1: piss off conservatives, you know. Yeah, I mean, and that's

1074
00:57:54,400 --> 00:57:58,800
Speaker 1: that is like a criminal level of negligence. I would argue,

1075
00:57:59,280 --> 00:58:01,480
Speaker 1: that's leaving a did gun with a six year old,

1076
00:58:01,520 --> 00:58:04,520
Speaker 1: you know yeah, and being like, well, I was pretty

1077
00:58:04,520 --> 00:58:07,520
Speaker 1: sure it had a safety. Yeah. I just I'm like

1078
00:58:07,680 --> 00:58:11,440
Speaker 1: they like, God, there's just I miss when they were

1079
00:58:11,520 --> 00:58:18,600
Speaker 1: radicalized by farm bill. Yeah, white supremacy. Yeah. So my

1080
00:58:18,720 --> 00:58:21,480
Speaker 1: critiques aside, this internal report does provide us with a

1081
00:58:21,480 --> 00:58:23,960
Speaker 1: lot of really useful information info that would have been

1082
00:58:24,040 --> 00:58:26,480
Speaker 1: very helpful to experts seeking to stop the spread of

1083
00:58:26,520 --> 00:58:29,479
Speaker 1: violent extremism online if they had had it back when

1084
00:58:29,520 --> 00:58:32,840
Speaker 1: Facebook found it out. So it's rad that Facebook absolutely

1085
00:58:32,880 --> 00:58:35,760
Speaker 1: never intended to release any of this. Isn't that cool?

1086
00:58:36,480 --> 00:58:39,320
Speaker 1: And that they were never going to put any of

1087
00:58:39,320 --> 00:58:41,520
Speaker 1: this out? There's like really useful. Dad. I have a

1088
00:58:41,600 --> 00:58:43,120
Speaker 1: quote in here. I don't think I'll read it because

1089
00:58:43,120 --> 00:58:45,120
Speaker 1: it's a bunch of numbers and it's only really interesting

1090
00:58:45,120 --> 00:58:47,720
Speaker 1: to nerds about this, but about like how many of

1091
00:58:47,760 --> 00:58:49,680
Speaker 1: the people who get in to stop the steel groups

1092
00:58:49,680 --> 00:58:52,000
Speaker 1: come in through like invites, and like how many people

1093
00:58:52,000 --> 00:58:54,320
Speaker 1: are actually responsible for the invites. What the average number

1094
00:58:54,320 --> 00:58:57,880
Speaker 1: of invites for a person is? Like, it's really interesting stuff.

1095
00:58:58,200 --> 00:58:59,480
Speaker 1: I'll have the links for it in there. You can

1096
00:58:59,520 --> 00:59:02,440
Speaker 1: read these into and on Facebook post. But like you

1097
00:59:02,440 --> 00:59:04,640
Speaker 1: know what, I'll read the quote. Stop the Steel was

1098
00:59:04,680 --> 00:59:08,440
Speaker 1: able to grow rapidly through coordinated group invites. Sixtent of

1099
00:59:08,480 --> 00:59:11,840
Speaker 1: stop the Steel joints came through invites. Moreover, these invites

1100
00:59:11,880 --> 00:59:15,560
Speaker 1: were dominated by a handful of superinviterstent of invites came

1101
00:59:15,600 --> 00:59:18,640
Speaker 1: from just three point three percent of inviters. Invits also

1102
00:59:18,720 --> 00:59:21,600
Speaker 1: tended to be connected to one another through interactions. They

1103
00:59:21,600 --> 00:59:24,360
Speaker 1: comment on tag and share one another's content. Out of

1104
00:59:24,400 --> 00:59:27,520
Speaker 1: six thousand, four hundred and fifty high engagers, four thousand

1105
00:59:27,600 --> 00:59:31,200
Speaker 1: and twenty five of them were directly connected to one another,

1106
00:59:31,440 --> 00:59:34,320
Speaker 1: meaning they interacted with one another's content or messaged one

1107
00:59:34,320 --> 00:59:38,440
Speaker 1: another when using the full information cord or seventy connected

1108
00:59:38,480 --> 00:59:40,520
Speaker 1: to one another. This suggests that the bulk of the

1109
00:59:40,560 --> 00:59:42,600
Speaker 1: stop to steel application was happening as part of a

1110
00:59:42,640 --> 00:59:45,080
Speaker 1: cohesive movement. This would have been great data to have

1111
00:59:45,240 --> 00:59:49,600
Speaker 1: in January, right, That would have been really good to know. YEA.

1112
00:59:49,960 --> 00:59:53,720
Speaker 1: That is speaking as a guy who does this professionally.

1113
00:59:53,760 --> 00:59:56,360
Speaker 1: That would have been great to have. But this is

1114
00:59:56,400 --> 01:00:00,280
Speaker 1: all just internal Like, okay, so we know, so we

1115
01:00:00,400 --> 01:00:02,919
Speaker 1: know this, let us never to speak of it again.

1116
01:00:02,960 --> 01:00:06,800
Speaker 1: Problem is, yeah, now we'll deny it to anyone who

1117
01:00:06,840 --> 01:00:09,600
Speaker 1: alleges this while we have this excellent data that we

1118
01:00:09,640 --> 01:00:13,680
Speaker 1: will not hand out because we're pieces of ship. Yeah yeah,

1119
01:00:14,160 --> 01:00:17,400
Speaker 1: called January six. Facebook employees were as horrified as anyone

1120
01:00:17,400 --> 01:00:20,160
Speaker 1: else by what happened in the capital. This horror was

1121
01:00:20,160 --> 01:00:22,960
Speaker 1: tweaked up several degrees by the undeniable fact that their

1122
01:00:23,000 --> 01:00:25,160
Speaker 1: work for the social network had helped to enable it.

1123
01:00:25,240 --> 01:00:29,360
Speaker 1: Was that it was their fault. Yeah, in that Yeah,

1124
01:00:29,440 --> 01:00:31,640
Speaker 1: in the same way that like when I finish having

1125
01:00:31,680 --> 01:00:34,520
Speaker 1: a gasoline and match fight in my neighbor's house and

1126
01:00:34,560 --> 01:00:39,200
Speaker 1: then inexplicable tragedy ensues, I can't help but feel somewhat responsible,

1127
01:00:39,280 --> 01:00:42,200
Speaker 1: you know. Wait, hold on, I'm feeling this vague melan

1128
01:00:42,360 --> 01:00:44,080
Speaker 1: and I know it's not my fault. Don't worry. I

1129
01:00:44,120 --> 01:00:47,960
Speaker 1: know it's not my fault. Um, but I feel that way. No,

1130
01:00:48,120 --> 01:00:49,680
Speaker 1: you just lit a match. I mean, I think we

1131
01:00:49,720 --> 01:00:53,440
Speaker 1: can hold the match accountable. The match was responsible, the match,

1132
01:00:53,560 --> 01:00:55,760
Speaker 1: the neighbor for having a house A lot of people

1133
01:00:55,760 --> 01:00:59,880
Speaker 1: are to blame here them, That's on them exactly. So

1134
01:01:00,280 --> 01:01:03,560
Speaker 1: they're all horrified. Everybody's horrified. Much of the riot itself

1135
01:01:03,600 --> 01:01:06,680
Speaker 1: was broadcast in a series of Facebook live streams as

1136
01:01:06,720 --> 01:01:09,800
Speaker 1: Mike Pince's secret service details scrambled to extricate him from

1137
01:01:09,800 --> 01:01:12,840
Speaker 1: the Capitol grounds. Facebook employees tried to enact their break

1138
01:01:12,880 --> 01:01:16,600
Speaker 1: the Glass emergency measures, and originally conceived for the immediate

1139
01:01:16,640 --> 01:01:20,240
Speaker 1: post election period, this was too little, too late. That evening,

1140
01:01:20,320 --> 01:01:23,680
Speaker 1: Mark Zuckerberg posted a message on Facebook's internal messaging system

1141
01:01:23,720 --> 01:01:26,880
Speaker 1: with the title employee f y I. He claimed to

1142
01:01:26,880 --> 01:01:29,440
Speaker 1: be horrified about what had happened and reiterated his company's

1143
01:01:29,440 --> 01:01:33,120
Speaker 1: commitment to democracy. Chief technology officer Mike Schrepfer, one of

1144
01:01:33,160 --> 01:01:36,800
Speaker 1: the most internally respected members of Facebook's leadership, also made

1145
01:01:36,800 --> 01:01:39,400
Speaker 1: a post asking employees to hang in there while the

1146
01:01:39,440 --> 01:01:43,520
Speaker 1: company decided on its next steps. The theme from the

1147
01:01:43,560 --> 01:01:49,120
Speaker 1: Trolls song like what Jeffrey Tatsenberg fired the entire hang

1148
01:01:49,200 --> 01:01:52,320
Speaker 1: in their folks, hang in there, you're hanging in there.

1149
01:01:52,400 --> 01:01:57,120
Speaker 1: Here's an amazing song by misss Anna Kendrick. That's right,

1150
01:01:57,160 --> 01:02:00,200
Speaker 1: that's what he did. Uh. So he tells them us

1151
01:02:00,440 --> 01:02:02,640
Speaker 1: uh and an employee responds, we have been hanging in

1152
01:02:02,680 --> 01:02:04,920
Speaker 1: there for years. We must demand more action from our

1153
01:02:05,000 --> 01:02:08,400
Speaker 1: leaders at this point, faith alone is not sufficient. Another

1154
01:02:08,440 --> 01:02:11,440
Speaker 1: message was more pointed, all due respect, but haven't we

1155
01:02:11,480 --> 01:02:13,920
Speaker 1: had enough time to figure out how to manage discourse

1156
01:02:13,960 --> 01:02:16,680
Speaker 1: without enabling violence. We've been fueling this fire for a

1157
01:02:16,720 --> 01:02:18,920
Speaker 1: long time, and we shouldn't be surprised it's now out

1158
01:02:18,960 --> 01:02:24,280
Speaker 1: of control. The Atlantic. Yeah, fair, the Atlantic the poster

1159
01:02:24,440 --> 01:02:26,480
Speaker 1: with a little kitten hanging from the branch that says

1160
01:02:26,560 --> 01:02:29,800
Speaker 1: we have been hanging in there for years? Yea? Or

1161
01:02:29,800 --> 01:02:31,800
Speaker 1: how about we have been fueling this fire for a

1162
01:02:31,840 --> 01:02:34,200
Speaker 1: long time and shouldn't be surprised now it's now out

1163
01:02:34,200 --> 01:02:36,080
Speaker 1: of control. We could do that with the this is

1164
01:02:36,120 --> 01:02:40,520
Speaker 1: fine meme. Guys sitting in the fire. This is on us.

1165
01:02:40,640 --> 01:02:44,760
Speaker 1: You know, we shouldn't be This isn't surprising. We had

1166
01:02:44,800 --> 01:02:47,880
Speaker 1: ample warning of the fire. But hang in there, Hang

1167
01:02:47,920 --> 01:02:50,880
Speaker 1: in there, kiddos. So I think The Atlantic has done

1168
01:02:50,880 --> 01:02:53,560
Speaker 1: some of the best reporting I found on this particular

1169
01:02:53,640 --> 01:02:56,320
Speaker 1: moment when Mark and shreppor get up and like say,

1170
01:02:56,560 --> 01:02:59,200
Speaker 1: don't we like hanging there? We love democracy and like

1171
01:02:59,400 --> 01:03:03,440
Speaker 1: people go off on them January six, like there are

1172
01:03:03,440 --> 01:03:06,320
Speaker 1: people were like a little more open about the frustration

1173
01:03:06,360 --> 01:03:08,760
Speaker 1: they felt about all this stuff then and then they

1174
01:03:08,760 --> 01:03:12,680
Speaker 1: stopped being that open um. It's frustrating, and everybody's treating

1175
01:03:12,720 --> 01:03:15,439
Speaker 1: these people with kid loves whatever. The Atlantic has done

1176
01:03:15,440 --> 01:03:18,480
Speaker 1: really good reporting on this, on this exact momentum, which

1177
01:03:18,480 --> 01:03:20,560
Speaker 1: seems to have been kind of a damn breaking situation

1178
01:03:20,640 --> 01:03:23,760
Speaker 1: for unrest within the company. One employee wrote, what are

1179
01:03:23,800 --> 01:03:26,240
Speaker 1: we going to do differently to make the future better?

1180
01:03:26,320 --> 01:03:28,880
Speaker 1: If the answer is nothing, then we're just that apocryphal

1181
01:03:28,880 --> 01:03:32,520
Speaker 1: Einstein quote on the definition of insanity. Another added to

1182
01:03:32,600 --> 01:03:35,280
Speaker 1: Mike shrep for please understand, I think you are a

1183
01:03:35,320 --> 01:03:37,520
Speaker 1: great person and genuinely one of the good people in

1184
01:03:37,600 --> 01:03:40,040
Speaker 1: leadership that keeps me here. But we cannot deal with

1185
01:03:40,080 --> 01:03:43,720
Speaker 1: fundamentally bad faith actors by using good faith solutions. Um,

1186
01:03:44,600 --> 01:03:46,439
Speaker 1: that's a good way to but yeah, yeah, I would

1187
01:03:46,480 --> 01:03:50,080
Speaker 1: also like democratic leadership to know that. But well, let's

1188
01:03:50,080 --> 01:03:51,840
Speaker 1: not set the bar too high. They're they're not going

1189
01:03:51,880 --> 01:03:54,480
Speaker 1: to figure that out now. In the wake of January six,

1190
01:03:54,600 --> 01:03:58,280
Speaker 1: an awful lot of people me included exhaustively documented Facebook's

1191
01:03:58,280 --> 01:04:01,400
Speaker 1: contribution to the attack and criticized company for enabling political

1192
01:04:01,440 --> 01:04:04,400
Speaker 1: violence on a grand scale. The company responded the way

1193
01:04:04,440 --> 01:04:08,040
Speaker 1: they always respond with lies. Mark Zuckerberg told Congress in

1194
01:04:08,080 --> 01:04:10,520
Speaker 1: March that Facebook quote did our part to secure the

1195
01:04:10,560 --> 01:04:13,840
Speaker 1: integrity of our election. Cheryl Sandberg, boss girl and a

1196
01:04:13,960 --> 01:04:17,120
Speaker 1: chief operating officer for the company, claimed in mid January

1197
01:04:17,160 --> 01:04:20,640
Speaker 1: that the Capitol Yeah claimed in mid January that the

1198
01:04:20,680 --> 01:04:23,919
Speaker 1: Capitol riot was quote largely organized on platforms that don't

1199
01:04:23,960 --> 01:04:27,919
Speaker 1: have our abilities to stop hate. I mean, Robert, first

1200
01:04:27,920 --> 01:04:31,000
Speaker 1: of all, as you know, as a big Sandberg advocate,

1201
01:04:31,400 --> 01:04:33,600
Speaker 1: you can't talk about her that way because she told

1202
01:04:33,920 --> 01:04:38,480
Speaker 1: women that they should negotiate their own salary. You fucking loser.

1203
01:04:39,000 --> 01:04:42,640
Speaker 1: Did you ever think about negotiating your own salary? Fucking

1204
01:04:42,720 --> 01:04:48,080
Speaker 1: dweeb dollars? And I love that. I do love that too.

1205
01:04:48,160 --> 01:04:51,320
Speaker 1: That's my favorite thing that she did. So Sandberg is

1206
01:04:51,320 --> 01:04:53,920
Speaker 1: a lot smarter than Mark Zuckerberg, and her statement was

1207
01:04:53,960 --> 01:04:56,400
Speaker 1: the very clever sort of not technically a lie that

1208
01:04:56,440 --> 01:04:59,919
Speaker 1: spreads a lot more disinformation than just a normal lie. Whatever, man,

1209
01:05:00,400 --> 01:05:03,240
Speaker 1: Because it's technically true that more actual organizing for the

1210
01:05:03,360 --> 01:05:05,600
Speaker 1: riot was done in places like Parlor as well as

1211
01:05:05,600 --> 01:05:08,479
Speaker 1: more private messaging apps. But it's also a lie because

1212
01:05:08,520 --> 01:05:11,160
Speaker 1: the organizing of the actual movement that spawned the riot

1213
01:05:11,240 --> 01:05:14,480
Speaker 1: to stop the steel ship that was almost entirely Facebook.

1214
01:05:14,520 --> 01:05:17,600
Speaker 1: So like, yeah, people didn't like go and open Facebook

1215
01:05:17,600 --> 01:05:19,680
Speaker 1: groups and do like most of the people didn't like

1216
01:05:19,720 --> 01:05:21,400
Speaker 1: go in there and be like, Okay, we're doing a caravan,

1217
01:05:21,440 --> 01:05:25,000
Speaker 1: although some people did and we have quotes of that. Um,

1218
01:05:25,040 --> 01:05:27,000
Speaker 1: a lot of that happened in other apps, but like

1219
01:05:27,280 --> 01:05:30,720
Speaker 1: the overall those other apps if they had not first

1220
01:05:30,760 --> 01:05:34,000
Speaker 1: been on Facebook. No, yeah, that's what I'm saying, Like,

1221
01:05:34,040 --> 01:05:36,960
Speaker 1: that's why it's smarter than her, because Mark Zuckerberg is

1222
01:05:37,000 --> 01:05:41,520
Speaker 1: just lying to Congress, um, because they didn't Cheryl Sandberg

1223
01:05:41,600 --> 01:05:46,040
Speaker 1: is being very intelligent um, and also kind of backhandedly

1224
01:05:46,120 --> 01:05:51,520
Speaker 1: complimenting Facebook in its hour of most blatant failure within

1225
01:05:51,560 --> 01:05:54,960
Speaker 1: the United States. At least not most blatant failure. That

1226
01:05:54,960 --> 01:05:57,640
Speaker 1: would be far. I mean, check the date that year, listen,

1227
01:05:57,640 --> 01:05:59,520
Speaker 1: you know, check the date record at this we may

1228
01:05:59,560 --> 01:06:01,880
Speaker 1: have had an ethnic cleansing enabled by Facebook by the

1229
01:06:01,920 --> 01:06:06,920
Speaker 1: time this episode drops. Yeah, as of this recording of

1230
01:06:06,960 --> 01:06:09,840
Speaker 1: the non ethnic cleansings and also the things that have

1231
01:06:09,880 --> 01:06:11,720
Speaker 1: been done in the United States that were worse than

1232
01:06:11,760 --> 01:06:14,760
Speaker 1: Facebook did, this tops the list, does not top the

1233
01:06:14,760 --> 01:06:16,920
Speaker 1: list of their overall crimes, which include the deaths of

1234
01:06:16,920 --> 01:06:21,960
Speaker 1: tens of thousands us. True, ye, good stuff, good stuff.

1235
01:06:22,080 --> 01:06:24,840
Speaker 1: I just I felt like the need to celebrate I do.

1236
01:06:25,040 --> 01:06:29,000
Speaker 1: Oh those were nice. Those are nice. My crush used

1237
01:06:29,000 --> 01:06:30,880
Speaker 1: to send me those in high school. And now that

1238
01:06:30,920 --> 01:06:33,600
Speaker 1: website is responsible for the deaths of tens of thousands

1239
01:06:33,600 --> 01:06:37,520
Speaker 1: of people. Yeah, it's it's good stuff. So what I

1240
01:06:37,560 --> 01:06:40,320
Speaker 1: find so utterly fascinating about the leaks we have of

1241
01:06:40,400 --> 01:06:43,040
Speaker 1: Facebook employees responding to their bosses on the evening of

1242
01:06:43,120 --> 01:06:46,520
Speaker 1: January six is that it makes it irrevocably, undeniably clear

1243
01:06:46,560 --> 01:06:49,760
Speaker 1: that Zuckerberg and Sandberg and every other Facebook mouthpiece lide

1244
01:06:49,920 --> 01:06:52,200
Speaker 1: when they claimed the company had no responsibility for the

1245
01:06:52,240 --> 01:06:55,800
Speaker 1: violence on January six, The people who worked for Facebook

1246
01:06:56,000 --> 01:06:59,640
Speaker 1: on the day it happened immediately blamed their own company

1247
01:06:59,640 --> 01:07:03,720
Speaker 1: for the arnage quote. Really do appreciate this response, and

1248
01:07:03,800 --> 01:07:05,960
Speaker 1: I can imagine the stress leadership is under, But this

1249
01:07:06,000 --> 01:07:08,400
Speaker 1: feels like the only avenue where we opt to compare

1250
01:07:08,400 --> 01:07:10,919
Speaker 1: ourselves to other companies, rather than taking our own lead.

1251
01:07:11,160 --> 01:07:14,240
Speaker 1: If our headsheets shocked, if our headsets shocked someone, would

1252
01:07:14,240 --> 01:07:16,720
Speaker 1: we say, well, it's still much better than PlayStation VR

1253
01:07:16,880 --> 01:07:20,200
Speaker 1: and it's unprecedented technology, which I felt otherwise, but it's

1254
01:07:20,200 --> 01:07:22,520
Speaker 1: simply not enough to say we're adapting, because we should

1255
01:07:22,520 --> 01:07:25,760
Speaker 1: have adapted already long ago. The atrophy occurs when people

1256
01:07:25,800 --> 01:07:28,280
Speaker 1: know how to circumvent our policies and we're too reactive

1257
01:07:28,320 --> 01:07:30,920
Speaker 1: to stay ahead. There were dozens of Stopped the Steel

1258
01:07:30,920 --> 01:07:33,760
Speaker 1: groups active until recently, and I doubt they minced words

1259
01:07:33,760 --> 01:07:38,160
Speaker 1: about their intense intentions. Again, hugely appreciate your response and

1260
01:07:38,200 --> 01:07:40,680
Speaker 1: the dialogue, but I'm simply exhausted by the weight here.

1261
01:07:40,920 --> 01:07:44,640
Speaker 1: We're at Facebook, not some naive startup. With the unprecedented

1262
01:07:44,640 --> 01:07:47,640
Speaker 1: resources we have, we should do better. Yeah, but I've

1263
01:07:47,920 --> 01:07:50,640
Speaker 1: that's like part of the Zuckerberg eathos is to continue

1264
01:07:50,680 --> 01:07:56,840
Speaker 1: to behave like he's fast and yeah or well or

1265
01:07:57,000 --> 01:08:01,040
Speaker 1: more iconically, Um what if what is the quote that's

1266
01:08:01,040 --> 01:08:09,080
Speaker 1: on my shirt? Um? Oh and that's how I live

1267
01:08:09,160 --> 01:08:16,639
Speaker 1: my life? Ha ha the key And he's still like, yeah,

1268
01:08:16,760 --> 01:08:19,400
Speaker 1: I mean, and I feel like that is whatever. I'm

1269
01:08:19,439 --> 01:08:21,840
Speaker 1: sure that does say so much about him, because like

1270
01:08:22,000 --> 01:08:24,400
Speaker 1: a good person can say, you can break the law

1271
01:08:24,439 --> 01:08:26,559
Speaker 1: and not be unethical, and that's how I live my life.

1272
01:08:26,640 --> 01:08:30,680
Speaker 1: And that's fine because the laws generally trash. Zuckerberg is

1273
01:08:30,720 --> 01:08:32,960
Speaker 1: specifically saying, I get to be a piece of ship

1274
01:08:33,000 --> 01:08:34,960
Speaker 1: as long as I don't technically break the law. And

1275
01:08:34,960 --> 01:08:37,599
Speaker 1: because I have money, I'm never technically breaking the law.

1276
01:08:37,960 --> 01:08:41,600
Speaker 1: Is that not sweet? And then don't forget ha ha

1277
01:08:40,840 --> 01:08:44,960
Speaker 1: ha ha. You can be an ethical and still be illegal.

1278
01:08:45,080 --> 01:08:46,720
Speaker 1: That's the way I live my life. Ha ha. I'll

1279
01:08:46,720 --> 01:08:49,479
Speaker 1: never forgaw, I'll never, never, never forget. I'm getting that

1280
01:08:49,520 --> 01:08:53,960
Speaker 1: tattooed right above my comeback with a warrant tramp stamp.

1281
01:08:54,479 --> 01:08:57,639
Speaker 1: Yeah and yeah and then and I would also recommend

1282
01:08:57,680 --> 01:08:59,720
Speaker 1: getting it on on the other side, right next to

1283
01:08:59,800 --> 01:09:02,320
Speaker 1: your phoenix that I know you're you're planning out. Oh

1284
01:09:02,400 --> 01:09:07,080
Speaker 1: full back. It's actually gonna be a perfect replica of

1285
01:09:07,240 --> 01:09:09,559
Speaker 1: the tattoo that Ben Affleck has. And then over my

1286
01:09:09,640 --> 01:09:13,639
Speaker 1: chest a perfect photo realistic tattoo have Ben Affleck picking

1287
01:09:13,720 --> 01:09:16,200
Speaker 1: up Duncan Donuts and looking like he's just watched his

1288
01:09:16,240 --> 01:09:21,080
Speaker 1: dog shoot itself. No, I don't I like appreciate his

1289
01:09:21,120 --> 01:09:24,200
Speaker 1: devotion to Duncan Donuts. I don't know how he I mean,

1290
01:09:24,240 --> 01:09:26,160
Speaker 1: I guess he's just tired, because I'm like, I don't

1291
01:09:26,160 --> 01:09:28,120
Speaker 1: look that way at Duncan dunt picture I've seen of

1292
01:09:28,160 --> 01:09:30,679
Speaker 1: myself at Duncan Donuts. I look so happy to be there.

1293
01:09:30,760 --> 01:09:32,840
Speaker 1: How could you not be thrilled? I don't know. I

1294
01:09:32,880 --> 01:09:34,800
Speaker 1: don't know. I mean, i'd say it's Boston, but you're

1295
01:09:34,840 --> 01:09:37,280
Speaker 1: from Boston, right, I'm from Boston and I'm so happy

1296
01:09:37,320 --> 01:09:39,519
Speaker 1: to be there. Yeah. People keep saying, No, he doesn't

1297
01:09:39,520 --> 01:09:42,160
Speaker 1: look miserable. He just looks like he's from Boston, and

1298
01:09:42,320 --> 01:09:46,640
Speaker 1: I think he just as to that. I don't go

1299
01:09:46,680 --> 01:09:49,280
Speaker 1: to other girls, Robert. I don't know if you know that.

1300
01:09:49,360 --> 01:09:51,800
Speaker 1: I'm not like other girls. So I'm I'm happy at

1301
01:09:51,840 --> 01:09:56,000
Speaker 1: the Duncan Donuts. Okay, okay, fair enough? Um so yeah,

1302
01:09:56,200 --> 01:09:58,200
Speaker 1: I don't know. At the beginning, I talked about the

1303
01:09:58,200 --> 01:10:00,639
Speaker 1: fact that I have said I think working for Facebook

1304
01:10:00,680 --> 01:10:04,160
Speaker 1: is an a moral activity today given what's known. That said,

1305
01:10:04,280 --> 01:10:06,799
Speaker 1: there were some points made during this employee bitch session

1306
01:10:06,880 --> 01:10:09,400
Speaker 1: that do make me kind of hesitant to suggest employees

1307
01:10:09,479 --> 01:10:13,040
Speaker 1: just bounce from the company. In mass quote, please continue

1308
01:10:13,080 --> 01:10:15,679
Speaker 1: to fight for us, shrep. This is the person talking

1309
01:10:15,680 --> 01:10:22,320
Speaker 1: to the I'm sorry you care you're hearing. I know,

1310
01:10:22,479 --> 01:10:25,920
Speaker 1: I know, I know. Facebook Engineering needs you representing our

1311
01:10:25,960 --> 01:10:28,559
Speaker 1: ethical standards in the highest levels of leadership unless uck,

1312
01:10:28,600 --> 01:10:31,519
Speaker 1: what's is products built by a cast of mercenaries and ghoules.

1313
01:10:31,560 --> 01:10:34,479
Speaker 1: We need to employ thousands of thoughtful, carrying engineers, and

1314
01:10:34,479 --> 01:10:36,639
Speaker 1: it will be difficult to continue to hire and retain

1315
01:10:36,680 --> 01:10:40,920
Speaker 1: them on our present course. That's not a terrible point, Okay,

1316
01:10:42,080 --> 01:10:45,800
Speaker 1: that feels like a half step. Yeah. Yeah, It's one

1317
01:10:45,840 --> 01:10:48,040
Speaker 1: of those things where like, on one hand, it is

1318
01:10:48,160 --> 01:10:50,760
Speaker 1: bad to work for a company that's entire job is

1319
01:10:50,800 --> 01:10:53,120
Speaker 1: to do harm at scale, which is what Facebook does.

1320
01:10:53,160 --> 01:10:55,280
Speaker 1: On the other hand, if they are replaced by people

1321
01:10:55,320 --> 01:10:58,679
Speaker 1: who don't have any ethical standards at all, that also

1322
01:10:59,000 --> 01:11:03,800
Speaker 1: probably isn't great. Now I don't agree with that, Yeah,

1323
01:11:04,040 --> 01:11:07,599
Speaker 1: it's it's complicated because, like I guess you could argue that, like, well,

1324
01:11:07,640 --> 01:11:09,680
Speaker 1: if all of the good engineers leave and they have

1325
01:11:09,760 --> 01:11:13,200
Speaker 1: to hire the ghouls, like, it'll fall apart eventually. And

1326
01:11:13,200 --> 01:11:14,800
Speaker 1: I guess it's the question of like when does the

1327
01:11:15,400 --> 01:11:19,919
Speaker 1: damage done by Facebook, like fading in popularity hopefully eclipse

1328
01:11:19,960 --> 01:11:22,360
Speaker 1: the damage done by the fact that everyone working there

1329
01:11:22,520 --> 01:11:25,920
Speaker 1: is the blackwater equivalent of a guy coding a fucking algorithm.

1330
01:11:25,960 --> 01:11:29,840
Speaker 1: Like I don't know, it's yeah, it's whatever, it's a

1331
01:11:30,240 --> 01:11:34,400
Speaker 1: it's just something to think about. I guess, um, yeah, yeah,

1332
01:11:34,760 --> 01:11:37,240
Speaker 1: Facebook is having a hiring crisis right now. I think

1333
01:11:37,280 --> 01:11:39,400
Speaker 1: it's gotten a little better recently. Um, but they've had

1334
01:11:39,479 --> 01:11:42,240
Speaker 1: massive shortages of engineers and failed to meet their hiring

1335
01:11:42,280 --> 01:11:44,760
Speaker 1: goals in two thousand, nineteen twenty. I don't know if

1336
01:11:44,760 --> 01:11:46,680
Speaker 1: they're gonna if they have, if it's gotten better this

1337
01:11:46,760 --> 01:11:49,599
Speaker 1: year or not. Um, I don't know how much any

1338
01:11:49,640 --> 01:11:52,559
Speaker 1: of that's gonna like help uh matters. I don't know.

1339
01:11:52,680 --> 01:11:55,559
Speaker 1: It seems unlikely that anything will get better anytime. I mean, yeah,

1340
01:11:55,600 --> 01:11:57,519
Speaker 1: I think the real solution is to make the company

1341
01:11:58,080 --> 01:12:02,640
Speaker 1: run by even worse who are less qualified that I

1342
01:12:02,640 --> 01:12:05,640
Speaker 1: don't know. That doesn't that doesn't sound great either. I

1343
01:12:05,680 --> 01:12:08,719
Speaker 1: don't know. I think there are volcanoes and that probably

1344
01:12:08,760 --> 01:12:11,320
Speaker 1: has part of the solution to the Facebook problem in it.

1345
01:12:11,360 --> 01:12:15,240
Speaker 1: There you go cast their servers into the fires. So

1346
01:12:15,479 --> 01:12:18,280
Speaker 1: Mark Zuckerberg and his fellow Nerd horsemen of the Apocalypse

1347
01:12:18,320 --> 01:12:21,000
Speaker 1: have basically built a gun that fires automatically into a

1348
01:12:21,000 --> 01:12:24,240
Speaker 1: crowd called society every couple of seconds. The engineers are

1349
01:12:24,240 --> 01:12:29,640
Speaker 1: the people who keep the gun loaded. It's good, yes,

1350
01:12:29,680 --> 01:12:32,519
Speaker 1: So the engineers they keep the gun loaded, but also

1351
01:12:32,640 --> 01:12:34,880
Speaker 1: sometimes they jerk it away from shooting a child in

1352
01:12:34,920 --> 01:12:37,160
Speaker 1: the face, and if they leave the bull, the gun

1353
01:12:37,240 --> 01:12:38,800
Speaker 1: might run out of bullets. But it's just going to

1354
01:12:38,920 --> 01:12:41,640
Speaker 1: keep shooting straight until the crowd until that point. So

1355
01:12:41,680 --> 01:12:46,120
Speaker 1: maybe it's better to have engineers. Yeah, I had to

1356
01:12:46,200 --> 01:12:49,520
Speaker 1: end with a metaphor um. I don't know, it's complicated

1357
01:12:49,560 --> 01:12:52,320
Speaker 1: whatever I want to end this episode. Yeah, it's it's

1358
01:12:52,320 --> 01:12:54,080
Speaker 1: just a mess. It's a messy thing to think about.

1359
01:12:54,200 --> 01:12:56,479
Speaker 1: We should never have let it get this far. We've

1360
01:12:56,479 --> 01:12:58,200
Speaker 1: been putting fuel on the fire for a while. We

1361
01:12:58,200 --> 01:13:01,600
Speaker 1: shouldn't be surprised that it's burning every thing. Truly, it

1362
01:13:01,640 --> 01:13:04,240
Speaker 1: has become It feels like it is slowly becoming, just

1363
01:13:04,280 --> 01:13:08,400
Speaker 1: like an annual document drop of like, yeah, things have

1364
01:13:09,320 --> 01:13:14,280
Speaker 1: steadily gotten worse. Yeah, the Hell Company is pretty shitty here. Yeah,

1365
01:13:14,400 --> 01:13:21,960
Speaker 1: Hell Companies bad stuff at Nightmare Corp. Bad. It is

1366
01:13:22,040 --> 01:13:25,080
Speaker 1: really bad. It's funny how bad it is. I want

1367
01:13:25,080 --> 01:13:28,040
Speaker 1: to end this episode with my very favorite response from

1368
01:13:28,080 --> 01:13:30,759
Speaker 1: a Facebook employee to that message from ct O Shrepp

1369
01:13:31,400 --> 01:13:34,320
Speaker 1: Facebook employee, if you happen to be listening to this episode,

1370
01:13:34,360 --> 01:13:36,680
Speaker 1: please hit me up because I love you. Here it

1371
01:13:36,760 --> 01:13:39,799
Speaker 1: is never forget the day Trump wrote down the escalator

1372
01:13:39,800 --> 01:13:42,160
Speaker 1: in two thousand fifteen called for a ban on Muslims

1373
01:13:42,240 --> 01:13:45,200
Speaker 1: entering the US, and we determined that it violated our policies,

1374
01:13:45,240 --> 01:13:48,280
Speaker 1: and yet we explicitly overrode the policy and didn't take

1375
01:13:48,320 --> 01:13:50,680
Speaker 1: the video down. There was a straight line that can

1376
01:13:50,720 --> 01:13:52,880
Speaker 1: be drawn from that day to today, one of the

1377
01:13:52,960 --> 01:13:55,720
Speaker 1: darkest days in the history of democracy and self governance.

1378
01:13:55,920 --> 01:13:57,760
Speaker 1: Would it have made a difference in the end, We

1379
01:13:57,800 --> 01:14:02,320
Speaker 1: can never know, but history will not judge us kindly. Wow. Yeah, yeah,

1380
01:14:02,400 --> 01:14:06,280
Speaker 1: they know what they're doing. They know exactly what they're doing,

1381
01:14:06,360 --> 01:14:10,240
Speaker 1: and we get an annual reminder, we get an annual

1382
01:14:10,360 --> 01:14:15,439
Speaker 1: little Pelican summit drop of documents saying that no, they

1383
01:14:15,560 --> 01:14:17,800
Speaker 1: still know what they're doing. They have not forgotten what

1384
01:14:17,840 --> 01:14:21,439
Speaker 1: they're doing. I might suggest that overtake. Are we the

1385
01:14:21,479 --> 01:14:24,240
Speaker 1: batties as like the moment you know things need to change?

1386
01:14:24,240 --> 01:14:26,960
Speaker 1: When when you're like, boy, I think history might judge

1387
01:14:26,960 --> 01:14:30,120
Speaker 1: me from my employment decisions. I think I may be

1388
01:14:30,400 --> 01:14:33,599
Speaker 1: damned by like the historians of the future when they

1389
01:14:33,640 --> 01:14:36,760
Speaker 1: analyze my role in society. Right, And it's like, if

1390
01:14:36,800 --> 01:14:42,599
Speaker 1: you're at that place, Uh, that's not good. That we've

1391
01:14:42,640 --> 01:14:45,000
Speaker 1: passed the point of no return. You know, like five

1392
01:14:45,080 --> 01:14:48,560
Speaker 1: years ago with with Facebook, it's just good Laurd. I

1393
01:14:48,600 --> 01:14:50,560
Speaker 1: mean yeah, and I and I do like applaud the

1394
01:14:50,560 --> 01:14:53,879
Speaker 1: whistleblowers and the people who are continuing to drop documents.

1395
01:14:53,920 --> 01:14:56,080
Speaker 1: And at this point it also does just feel like,

1396
01:14:57,280 --> 01:15:00,000
Speaker 1: you know, getting punched in the face repeatedly because it's like, well,

1397
01:15:00,560 --> 01:15:02,920
Speaker 1: I'm glad that there is the paper trail, I'm glad

1398
01:15:02,960 --> 01:15:05,120
Speaker 1: that there's the evidence, but it's who at this point

1399
01:15:05,160 --> 01:15:10,160
Speaker 1: is surprised, like there's like when people it's whatever. I mean, technically,

1400
01:15:10,160 --> 01:15:12,720
Speaker 1: I think by the definition of the word, these are revelations,

1401
01:15:12,720 --> 01:15:15,800
Speaker 1: but they're also very much not They're just confirmations of

1402
01:15:15,840 --> 01:15:18,800
Speaker 1: things that appeared very obvious based on the conduct of

1403
01:15:18,840 --> 01:15:24,519
Speaker 1: this company already. Yep, yeah, Jamie, do you have any

1404
01:15:25,080 --> 01:15:27,920
Speaker 1: plugables for us? Do you do you perhaps have a

1405
01:15:28,080 --> 01:15:33,400
Speaker 1: Facebook owned Instagram meta thing? It don't We don't need

1406
01:15:33,439 --> 01:15:35,920
Speaker 1: to call it meta. I I think no one needs

1407
01:15:35,960 --> 01:15:37,960
Speaker 1: to call it what. There's only one meta and it's

1408
01:15:37,960 --> 01:15:45,880
Speaker 1: a metal world piece. Wow. Basketball player basketball player Ronald

1409
01:15:45,880 --> 01:15:48,000
Speaker 1: has changed his name a matter world Piece, like many

1410
01:15:48,080 --> 01:15:51,599
Speaker 1: years ago, so he did the meta first. Okay, alright,

1411
01:15:51,600 --> 01:15:55,559
Speaker 1: the metaverse was already taken. I you can you can

1412
01:15:55,960 --> 01:15:58,519
Speaker 1: listen to my podcast. I got a bunch you listen

1413
01:15:58,560 --> 01:16:01,240
Speaker 1: to Bechtel Cast, My Year and Men to Act Castle

1414
01:16:01,360 --> 01:16:05,439
Speaker 1: Lead to podcast or none. I won't know. I'll know

1415
01:16:05,720 --> 01:16:08,840
Speaker 1: in so actually Sophie will know. So maybe you'd better

1416
01:16:08,880 --> 01:16:12,040
Speaker 1: listen to them, or I'll lose my livelihood, which would

1417
01:16:12,080 --> 01:16:15,920
Speaker 1: be which would be interesting? Uh? You could follow me

1418
01:16:15,960 --> 01:16:21,799
Speaker 1: on Twitter Jamie Lofts Help, or Instagram at Jamie christ Superstar,

1419
01:16:21,960 --> 01:16:27,960
Speaker 1: where I bravely continued to use uh Zuckerberg's tools of

1420
01:16:28,000 --> 01:16:30,640
Speaker 1: having Yeah, isn't it funny Jamie that we call it

1421
01:16:30,680 --> 01:16:33,679
Speaker 1: our livelihood, which is just a nicer way of saying

1422
01:16:33,680 --> 01:16:38,320
Speaker 1: our ability to not starve to death in the street. Yeah? Yeah, whoever,

1423
01:16:38,640 --> 01:16:43,799
Speaker 1: whoever back in the day rebranded survival to livelihood. Really

1424
01:16:43,840 --> 01:16:48,000
Speaker 1: real genius, sleight of hand. Incredible. This is why I

1425
01:16:48,040 --> 01:16:51,080
Speaker 1: think we should have a program in schools where we

1426
01:16:51,200 --> 01:16:53,719
Speaker 1: determine which kids are going to be good at marketing,

1427
01:16:54,160 --> 01:16:55,960
Speaker 1: and then we shipped them to an island, and we

1428
01:16:56,040 --> 01:16:58,960
Speaker 1: don't we don't think about it after that point into

1429
01:16:59,000 --> 01:17:01,760
Speaker 1: the island. If they go, it's a nice island, like

1430
01:17:01,800 --> 01:17:04,879
Speaker 1: a good one, like a solid island. Marketing the island

1431
01:17:05,000 --> 01:17:07,680
Speaker 1: you're marketing to the island. That's good. I think we

1432
01:17:08,080 --> 01:17:09,840
Speaker 1: put them on an island and we divert all of

1433
01:17:09,880 --> 01:17:12,639
Speaker 1: the world's military forces to making sure that nothing gets

1434
01:17:12,640 --> 01:17:14,920
Speaker 1: to that island that isn't a marketer or leaves it,

1435
01:17:15,120 --> 01:17:17,280
Speaker 1: and then make sure that and and make fucking sure

1436
01:17:17,320 --> 01:17:19,920
Speaker 1: there's no WiFi signal on the Oh good god. No,

1437
01:17:20,240 --> 01:17:25,040
Speaker 1: would absolutely not absolutely fun. Once a day they can

1438
01:17:25,080 --> 01:17:28,200
Speaker 1: watch Shark Tank together, but that's all they're getting. Oh,

1439
01:17:28,320 --> 01:17:30,559
Speaker 1: that's sounds that actually kind of sounds nice living on

1440
01:17:30,600 --> 01:17:33,320
Speaker 1: an island and watching one episode of Shark Tank day.

1441
01:17:33,800 --> 01:17:38,080
Speaker 1: That's like, that's my like dream lobotomy. That's kind of nice. Yeah,

1442
01:17:38,160 --> 01:17:43,040
Speaker 1: well that's the episode, Paul Robert Jamie. How are you doing?

1443
01:17:43,360 --> 01:17:44,960
Speaker 1: How are you feeling? I feel like, you know, I

1444
01:17:45,439 --> 01:17:49,040
Speaker 1: am feeling just kind of a low thrumming of despair,

1445
01:17:49,560 --> 01:17:51,920
Speaker 1: but but I have good Yeah, that's right, that's how

1446
01:17:51,960 --> 01:17:54,439
Speaker 1: you're supposed to feel. I have felt worse at the

1447
01:17:54,520 --> 01:17:56,360
Speaker 1: end of this show, which I don't know if that

1448
01:17:56,600 --> 01:17:59,880
Speaker 1: says more about like my threshold for despair or you know,

1449
01:18:00,040 --> 01:18:03,679
Speaker 1: happens to be a coincidence, but you know, I'll say

1450
01:18:03,960 --> 01:18:06,360
Speaker 1: I'm hanging in there. But also I've been hanging in

1451
01:18:06,400 --> 01:18:09,040
Speaker 1: there for years. Yeah, that's all you can do is

1452
01:18:09,080 --> 01:18:11,599
Speaker 1: hang in there. Yeah, keep hanging in Are you hanging

1453
01:18:11,600 --> 01:18:15,320
Speaker 1: in there? Robert allegedly Yeah. I mean, by all accounts,

1454
01:18:15,320 --> 01:18:18,040
Speaker 1: you're hanging in there. But like internally, there's no who

1455
01:18:18,040 --> 01:18:21,240
Speaker 1: could say no. No, I'm I'm, I'm. I'm as unmoored

1456
01:18:21,240 --> 01:18:23,200
Speaker 1: and adrift as the rest of us are in these

1457
01:18:23,360 --> 01:18:25,680
Speaker 1: I'm gonna when I visit, I'm going to show you

1458
01:18:25,720 --> 01:18:29,040
Speaker 1: pin and I really think you're gonna like it. Oh good, Okay, Well,

1459
01:18:29,360 --> 01:18:29,920
Speaker 1: how's