1
00:00:03,520 --> 00:00:09,559
Speaker 1: Let's do it. Start the podcast podcast. All right, well

2
00:00:09,680 --> 00:00:12,119
Speaker 1: have the let's have that be what starts the podcast.

3
00:00:12,160 --> 00:00:16,279
Speaker 1: What we just said, let's start the podcast podcast. Let's

4
00:00:16,280 --> 00:00:19,720
Speaker 1: start the podcast. Well, I'm Robert Evans. Yep, I'm so

5
00:00:19,880 --> 00:00:23,600
Speaker 1: feeling I'm in. I never introduced myself. I'm Jamie. Who

6
00:00:23,640 --> 00:00:27,080
Speaker 1: are you, Jamie Loft? Anything more we need to say?

7
00:00:27,120 --> 00:00:29,720
Speaker 1: Are we done with the episode? Anderson's here? No, I

8
00:00:29,800 --> 00:00:34,400
Speaker 1: think that. Yeah, yeah, Anderson is here. Uh sure, Well,

9
00:00:34,800 --> 00:00:39,360
Speaker 1: you know what's happening in the world. Facebook is happening

10
00:00:39,360 --> 00:00:42,920
Speaker 1: to the world, and it's not great. Jamie. It's not great.

11
00:00:42,960 --> 00:00:45,480
Speaker 1: Sophie not not a fan of the Facebook. We left

12
00:00:45,520 --> 00:00:48,800
Speaker 1: off having gone through some of the Facebook papers, particularly

13
00:00:48,880 --> 00:00:53,120
Speaker 1: employees attacking their bosses after jan six, when it became

14
00:00:53,159 --> 00:00:55,800
Speaker 1: clear that the company they were working for was completely

15
00:00:55,840 --> 00:01:01,680
Speaker 1: morally indefensible. Attack they are. Yeah, they already knew. I

16
00:01:01,680 --> 00:01:05,200
Speaker 1: wouldn't call it attacking either. I would call it they

17
00:01:05,280 --> 00:01:09,480
Speaker 1: were pretty I mean, there's a guy like the quote there,

18
00:01:09,480 --> 00:01:12,280
Speaker 1: there was the guy who's like history won't judge as kindly. Uh,

19
00:01:12,640 --> 00:01:16,520
Speaker 1: the guy, and we didn't ban Trump into thus in fifteen.

20
00:01:16,600 --> 00:01:20,120
Speaker 1: That's what caused the Capital riot. I mean facts or facts?

21
00:01:20,200 --> 00:01:23,440
Speaker 1: Is that really attacking if you're just like, well, I

22
00:01:23,480 --> 00:01:28,639
Speaker 1: think that, Yeah, I think stating facts can be an attack. Well, okay,

23
00:01:28,720 --> 00:01:31,759
Speaker 1: put it on a T shirt. I mean for people

24
00:01:31,800 --> 00:01:35,240
Speaker 1: like this, you know, Um yeah, I think I think

25
00:01:35,240 --> 00:01:37,320
Speaker 1: stating facts can be an attack. And we ended Part

26
00:01:37,360 --> 00:01:40,880
Speaker 1: one by sharing some of the blistering criticisms of Facebook

27
00:01:40,880 --> 00:01:43,319
Speaker 1: employees against you know, management in the service itself. So

28
00:01:43,360 --> 00:01:45,800
Speaker 1: as we start part two, it's only proper that we

29
00:01:45,880 --> 00:01:49,520
Speaker 1: cover how Facebook responded to all of this internal criticism.

30
00:01:49,600 --> 00:01:51,920
Speaker 1: As I stated last episode, Facebook is in the midst

31
00:01:51,920 --> 00:01:54,480
Speaker 1: of a year's long drought of capable engineers and other

32
00:01:54,520 --> 00:01:57,080
Speaker 1: technical employees. They are having a lot of trouble hiring

33
00:01:57,120 --> 00:01:58,640
Speaker 1: all of the people that they need for all of

34
00:01:58,680 --> 00:02:01,680
Speaker 1: the things they're trying to do. Um. So, one of

35
00:02:01,720 --> 00:02:04,000
Speaker 1: the things is for a lot of these employees, when

36
00:02:04,000 --> 00:02:06,360
Speaker 1: they say things that are deeply critical, they can't just

37
00:02:06,480 --> 00:02:11,160
Speaker 1: dismiss the concerns of their employees outright because act like

38
00:02:12,000 --> 00:02:14,120
Speaker 1: if they were to do that, these people would get

39
00:02:14,160 --> 00:02:16,400
Speaker 1: angry and they need them, right. Facebook's not in the

40
00:02:16,400 --> 00:02:18,080
Speaker 1: strongest position when it comes to the people who are

41
00:02:18,080 --> 00:02:20,720
Speaker 1: good engineers. They have to walk a little bit of

42
00:02:20,720 --> 00:02:23,240
Speaker 1: a tight rope. However, if they were to actually do

43
00:02:23,320 --> 00:02:26,320
Speaker 1: anything about the actual meat of the concerns, it would

44
00:02:26,320 --> 00:02:29,280
Speaker 1: reduce profitability and in some cases destroy Facebook as it

45
00:02:29,280 --> 00:02:32,200
Speaker 1: currently exists. So they're not going to do anything, um,

46
00:02:32,320 --> 00:02:33,920
Speaker 1: which is meant that they've had to get kind of

47
00:02:33,919 --> 00:02:36,960
Speaker 1: creative with how they respond. So Mark and his fellow

48
00:02:37,160 --> 00:02:41,600
Speaker 1: bosses pivoted and argued that the damning critique like, yeah,

49
00:02:41,680 --> 00:02:43,799
Speaker 1: old Zucky z uck. So when when this all comes

50
00:02:43,800 --> 00:02:45,519
Speaker 1: out and people are like, boy, it sure seems like

51
00:02:45,560 --> 00:02:47,400
Speaker 1: all of your employees know that they're working for the

52
00:02:47,400 --> 00:02:51,960
Speaker 1: fucking death star. Um, Zuckerberg and his like mouthpieces made

53
00:02:51,960 --> 00:02:54,200
Speaker 1: a statement that like, all of these damning critiques from

54
00:02:54,200 --> 00:02:56,600
Speaker 1: people inside the company were actually evidence of the very

55
00:02:56,680 --> 00:03:00,240
Speaker 1: open culture inside Facebook, which encouraged workers to share their

56
00:03:00,240 --> 00:03:03,840
Speaker 1: opinions with management. Um. That's exactly what a company spokesperson

57
00:03:03,919 --> 00:03:06,480
Speaker 1: told The Atlantic when they asked about comments like history

58
00:03:06,480 --> 00:03:09,200
Speaker 1: will not judge us kindly. The fact that they're saying

59
00:03:09,200 --> 00:03:11,560
Speaker 1: will be damned by historians means that we really have

60
00:03:11,600 --> 00:03:16,320
Speaker 1: a healthy office culture. Um. Hashtag dust star, proud, dust Star,

61
00:03:16,400 --> 00:03:19,480
Speaker 1: proud everybody. Yeah. Yeah, it's like the fact that we

62
00:03:19,639 --> 00:03:22,839
Speaker 1: removed a stigma of working for the devil, right, I mean,

63
00:03:23,120 --> 00:03:25,760
Speaker 1: come on, the devil I would be proud to work

64
00:03:25,800 --> 00:03:28,240
Speaker 1: for because he's done some cool stuff, Like have you

65
00:03:28,280 --> 00:03:32,240
Speaker 1: ever been to Vegas? Nice town? Oh, I've been to Vegas.

66
00:03:32,280 --> 00:03:34,200
Speaker 1: I've seen I saw that. I saw the Backstreet Boys

67
00:03:34,200 --> 00:03:37,480
Speaker 1: in Vegas right before two of them were revealed to

68
00:03:37,520 --> 00:03:40,000
Speaker 1: be in Q and On. So really caught the end

69
00:03:40,000 --> 00:03:43,120
Speaker 1: of that, Locomo, Wow, I did not realize that a

70
00:03:43,280 --> 00:03:45,840
Speaker 1: sizeable percentage of the Backstreet Boys had gotten into Q

71
00:03:46,000 --> 00:03:50,040
Speaker 1: and On. That makes total sense. The Backstretoys, They're from Florida.

72
00:03:50,080 --> 00:03:53,400
Speaker 1: They're ultimately five men from Florida. So what can you do?

73
00:03:53,600 --> 00:03:56,600
Speaker 1: As the author of that article, The Atlantic article noted,

74
00:03:56,920 --> 00:04:00,000
Speaker 1: this stance allows Facebook to claim transparency while ignoring the

75
00:04:00,080 --> 00:04:02,720
Speaker 1: substance of the complaints and the implication of the complaints

76
00:04:02,760 --> 00:04:05,640
Speaker 1: that many of Facebook's employees believe their company operates without

77
00:04:05,640 --> 00:04:08,640
Speaker 1: a moral compass. All over America, people used Facebook to

78
00:04:08,760 --> 00:04:10,960
Speaker 1: organize convoys to d C and to fill the buses

79
00:04:11,000 --> 00:04:13,960
Speaker 1: they rented for their trips, and this was indeed done

80
00:04:14,240 --> 00:04:17,640
Speaker 1: in groups like the Lebanon main truth Seekers, where Kyle

81
00:04:17,640 --> 00:04:20,360
Speaker 1: fitz Simmons posted the following quote, this election was stolen,

82
00:04:20,400 --> 00:04:22,919
Speaker 1: and we're being slow walked towards Chinese ownership by an

83
00:04:23,080 --> 00:04:25,520
Speaker 1: establishment that is treasonous and all too willing to gaslight

84
00:04:25,520 --> 00:04:27,760
Speaker 1: the public into believing the theft was somehow the will

85
00:04:27,800 --> 00:04:29,839
Speaker 1: of the people. Would there be an interest locally in

86
00:04:29,920 --> 00:04:32,640
Speaker 1: organizing a caravan to Washington, d C. For the electoral

87
00:04:32,680 --> 00:04:37,760
Speaker 1: College vote count on January six? Um? Yeah, and Kyle

88
00:04:37,880 --> 00:04:40,680
Speaker 1: recently played guilt not guilty to eight federal charges, including

89
00:04:40,680 --> 00:04:43,920
Speaker 1: assault on a police officer. Mark Zuckerberg would argue that

90
00:04:43,960 --> 00:04:47,800
Speaker 1: like Facebook didn't play a significant role in organizing January six,

91
00:04:47,880 --> 00:04:50,240
Speaker 1: and couldn't have played a significant role in radicalizing this

92
00:04:50,320 --> 00:04:53,200
Speaker 1: guy and many other people. But the reality is that

93
00:04:53,800 --> 00:04:56,880
Speaker 1: for the people, like the people who managed part of

94
00:04:56,920 --> 00:04:59,720
Speaker 1: what led Kyle fitz Simmons to go assault people on

95
00:05:00,040 --> 00:05:02,880
Speaker 1: Nuary six, UM was the fact that he had been

96
00:05:02,960 --> 00:05:05,599
Speaker 1: radicalized by a social network that for years made the

97
00:05:05,600 --> 00:05:09,960
Speaker 1: conscious choice to amplify angry content and encourage anger because

98
00:05:09,960 --> 00:05:12,280
Speaker 1: it kept people on the site more right, Like all

99
00:05:12,320 --> 00:05:14,240
Speaker 1: of the anger that boiled up at January six that

100
00:05:14,320 --> 00:05:16,160
Speaker 1: came from a number of places, but one of those

101
00:05:16,200 --> 00:05:20,919
Speaker 1: places was social media. Because social media profited and specifically,

102
00:05:20,960 --> 00:05:25,160
Speaker 1: Facebook knowingly profited from making people angry. That was the business,

103
00:05:25,360 --> 00:05:27,120
Speaker 1: and of course it blew up in the real world.

104
00:05:27,279 --> 00:05:30,000
Speaker 1: I have a question, um, just out of your own

105
00:05:30,040 --> 00:05:34,840
Speaker 1: experience and observation, which is, how do you, um like,

106
00:05:34,880 --> 00:05:37,599
Speaker 1: if you're doing a side by side case study of

107
00:05:37,640 --> 00:05:40,960
Speaker 1: how Facebook responded to events like this versus how like

108
00:05:41,480 --> 00:05:47,400
Speaker 1: YouTube slash Google responded to radicalization? Are there like significant differences?

109
00:05:47,400 --> 00:05:50,960
Speaker 1: Did is there any did anyone do better or difference?

110
00:05:51,240 --> 00:05:56,960
Speaker 1: Twitter has done better than probably most of them YouTube.

111
00:05:57,000 --> 00:05:59,159
Speaker 1: I mean, and again I'm not saying this the Twitter

112
00:05:59,200 --> 00:06:01,880
Speaker 1: has done well, um or that YouTube has done well.

113
00:06:02,040 --> 00:06:05,719
Speaker 1: They've both done, particularly with coronavirus disinformation, a bit better

114
00:06:05,760 --> 00:06:08,440
Speaker 1: than Facebook. Um, and they were they were better in

115
00:06:08,480 --> 00:06:11,599
Speaker 1: general on not really YouTube as much, but like Twitter

116
00:06:11,720 --> 00:06:15,400
Speaker 1: was definitely has taken has been the most responsible of

117
00:06:15,440 --> 00:06:18,799
Speaker 1: the social networks around this stuff. Um. It did seem

118
00:06:18,800 --> 00:06:21,920
Speaker 1: like for a while there the various networks were kind

119
00:06:21,920 --> 00:06:24,120
Speaker 1: of like duking it out to see who could do

120
00:06:24,160 --> 00:06:27,080
Speaker 1: the absolute worst and damage the lives and it's it

121
00:06:27,160 --> 00:06:30,640
Speaker 1: seems like Facebook one. Yes, that I would say Facebook,

122
00:06:30,640 --> 00:06:34,919
Speaker 1: but and again Twitter chose to do a lot of

123
00:06:34,960 --> 00:06:38,039
Speaker 1: the same talxic things Facebook did, so did YouTube, and

124
00:06:38,040 --> 00:06:40,039
Speaker 1: they did it all for profit. A number of the

125
00:06:40,040 --> 00:06:43,120
Speaker 1: things we've criticized Facebook for, you can critique YouTube and Twitter. Four.

126
00:06:43,640 --> 00:06:46,920
Speaker 1: I would argue Twitter certainly has done more and more

127
00:06:46,920 --> 00:06:51,200
Speaker 1: effectively than Facebook, not enough that they're not being irresponsible,

128
00:06:51,200 --> 00:06:53,520
Speaker 1: because I would argue that Twitter has actually been extremely

129
00:06:53,560 --> 00:06:57,839
Speaker 1: irresponsible and knowingly so. Um. But I think Facebook, in

130
00:06:57,880 --> 00:07:01,159
Speaker 1: my analysis, Facebook has been the worst. Although I'm not

131
00:07:01,200 --> 00:07:03,920
Speaker 1: as I haven't gotten studied as much about like TikTok yet,

132
00:07:03,960 --> 00:07:07,800
Speaker 1: so we'll see. But but my analysis, you've got to

133
00:07:07,880 --> 00:07:12,720
Speaker 1: get on TikTok, pivot out of podcasting and into TikTok dances. Yeah,

134
00:07:12,800 --> 00:07:15,280
Speaker 1: I mean, I'm it's it's not the dances that concerned

135
00:07:15,280 --> 00:07:19,480
Speaker 1: me on TikTok. It's the minute long conspiracy theory videos

136
00:07:19,480 --> 00:07:21,920
Speaker 1: that have convinced a number of people that the Kardashians

137
00:07:21,960 --> 00:07:24,200
Speaker 1: are Armenian witches and had something to do with the

138
00:07:24,200 --> 00:07:27,040
Speaker 1: collapse of the astro worlds or the deaths in the

139
00:07:27,080 --> 00:07:31,760
Speaker 1: astro working. My my concern there is the dances that

140
00:07:31,840 --> 00:07:35,720
Speaker 1: go over those conspiracy videos and really marry the worst

141
00:07:35,720 --> 00:07:40,040
Speaker 1: of both worlds. Because I have seen dancing on there.

142
00:07:40,080 --> 00:07:43,800
Speaker 1: I have seen conspiracy videos that involved dancing incredible and

143
00:07:43,920 --> 00:07:47,360
Speaker 1: skincare routine. Have you ever seen a conspiracy video where

144
00:07:47,400 --> 00:07:50,400
Speaker 1: someone's also doing their skincare routine? Because that is a

145
00:07:50,560 --> 00:07:58,280
Speaker 1: thriving SOMETI I'm sure that's yeah. So well, I was like,

146
00:07:58,360 --> 00:08:03,960
Speaker 1: that is just a thing that on many platforms. Change

147
00:08:04,280 --> 00:08:10,920
Speaker 1: media companies are willfully bad at stopping radicalization because making

148
00:08:10,920 --> 00:08:14,280
Speaker 1: people angry and frightened is good for all of their

149
00:08:14,280 --> 00:08:17,960
Speaker 1: bottom lines, so they all knowingly participate in this. I

150
00:08:18,000 --> 00:08:21,680
Speaker 1: think Facebook has been the least responsible about it, but

151
00:08:21,760 --> 00:08:24,920
Speaker 1: that doesn't that shouldn't be taken as praise of anybody,

152
00:08:25,120 --> 00:08:28,080
Speaker 1: like saying Twitter. Saying Twitter has done the best is

153
00:08:28,120 --> 00:08:32,040
Speaker 1: saying like, well, we were all drunk driving, but John

154
00:08:32,160 --> 00:08:36,080
Speaker 1: could actually walk most of a straight line before vomiting,

155
00:08:36,240 --> 00:08:38,840
Speaker 1: so he was the least irresponsible of us who drunk

156
00:08:38,920 --> 00:08:40,800
Speaker 1: drove that night. Just to put it in terms that

157
00:08:40,840 --> 00:08:43,120
Speaker 1: I understand, I was it sounds like Twitter is the

158
00:08:43,120 --> 00:08:45,679
Speaker 1: Backstreet boy. That's like, look, I don't believe in Q

159
00:08:45,920 --> 00:08:50,280
Speaker 1: and on, but I see their points. That's that's kind

160
00:08:50,320 --> 00:08:53,960
Speaker 1: of um the vibe I'm getting fair enough. Um So,

161
00:08:54,240 --> 00:08:57,000
Speaker 1: when deciding what which posts should show up more often

162
00:08:57,000 --> 00:08:59,520
Speaker 1: in the feeds of other users. Facebook's algorithm ways a

163
00:08:59,600 --> 00:09:02,000
Speaker 1: number of actors. The end goal is always the same,

164
00:09:02,040 --> 00:09:03,960
Speaker 1: to get the most people to spend the most time

165
00:09:04,000 --> 00:09:06,920
Speaker 1: interacting with the site. For years, this was done by

166
00:09:06,960 --> 00:09:09,839
Speaker 1: calculating the different reactions of post god and weighing it

167
00:09:09,920 --> 00:09:13,360
Speaker 1: based on what responses people had to it. Again, for years,

168
00:09:13,400 --> 00:09:15,679
Speaker 1: the reaction that carried the most weight was anger. The

169
00:09:15,720 --> 00:09:18,320
Speaker 1: little snarling smiley face icon you could click under a

170
00:09:18,360 --> 00:09:21,200
Speaker 1: post it was at one point being waited five times

171
00:09:21,200 --> 00:09:24,160
Speaker 1: more than just like a like really like the again

172
00:09:24,200 --> 00:09:26,000
Speaker 1: when I'm when I'm saying this was all intentional. They

173
00:09:26,000 --> 00:09:29,720
Speaker 1: were like people who respond angrily to post that keeps

174
00:09:29,760 --> 00:09:31,640
Speaker 1: them on the site more. That's they spend the most

175
00:09:31,679 --> 00:09:33,880
Speaker 1: time engaging with things that make them angry. So, when

176
00:09:33,880 --> 00:09:38,040
Speaker 1: it comes to determining what by which methody like how

177
00:09:38,280 --> 00:09:41,320
Speaker 1: we choose to have the algorithm present people with posts,

178
00:09:41,760 --> 00:09:44,199
Speaker 1: the post that are making people angriest is the post

179
00:09:44,240 --> 00:09:46,640
Speaker 1: our algorithm will send to the most people. That's a

180
00:09:46,679 --> 00:09:50,080
Speaker 1: conscious choice. That's a conscious Yeah, it's so funny, how

181
00:09:50,120 --> 00:09:52,920
Speaker 1: I mean not funny, it's tragic and upsetting, But just

182
00:09:53,080 --> 00:09:55,680
Speaker 1: how specific the Facebook audience is that it's like you

183
00:09:55,679 --> 00:09:57,600
Speaker 1: would have to be the kind of person who would

184
00:09:57,960 --> 00:10:02,120
Speaker 1: be like, I'd better react angry as specific as possible.

185
00:10:02,120 --> 00:10:04,280
Speaker 1: In my feedback to this post, which is farm bill

186
00:10:04,440 --> 00:10:08,040
Speaker 1: moms and yeah it's boomers, it's boomers, and yeah, um

187
00:10:08,760 --> 00:10:12,040
Speaker 1: and yeah, they just kind of knowingly set off a

188
00:10:12,080 --> 00:10:14,720
Speaker 1: bomb and a lot of people's fucking brains. Um, they're

189
00:10:14,760 --> 00:10:19,560
Speaker 1: addicted to telling on themselves for Yeah, why why Facebook

190
00:10:19,600 --> 00:10:22,160
Speaker 1: has something called the Integrity Department and these are the

191
00:10:22,160 --> 00:10:25,400
Speaker 1: people with the unenviable task of trying to fight misinformation

192
00:10:25,440 --> 00:10:28,800
Speaker 1: and radicalization on the platform. They noted in July that

193
00:10:28,920 --> 00:10:33,880
Speaker 1: is so embarrassing. Yeah, just going on the first day

194
00:10:33,960 --> 00:10:37,000
Speaker 1: be like, I work for the Facebook Integrity Department, Like, yeah,

195
00:10:37,080 --> 00:10:40,360
Speaker 1: good fucking luck. Yeah, I work for thee in your life.

196
00:10:40,400 --> 00:10:42,839
Speaker 1: My job is to go door to door and apologize

197
00:10:42,840 --> 00:10:45,920
Speaker 1: to people after we bomb them. We have gift baskets

198
00:10:46,080 --> 00:10:50,480
Speaker 1: for the survivors, you know, Like it's that that's the gig. Really. Um, yeah,

199
00:10:50,640 --> 00:10:53,360
Speaker 1: I send edible arrangements to people who have been droned

200
00:10:53,400 --> 00:10:57,600
Speaker 1: striight like, oh Jesus awful. There's was one of my

201
00:10:57,640 --> 00:11:01,200
Speaker 1: favorite follows on twitters, Brook Minkowski Um, who used to

202
00:11:01,240 --> 00:11:03,079
Speaker 1: work for Facebook and was like one of the people

203
00:11:03,080 --> 00:11:06,559
Speaker 1: early on who was trying to warn them about disinformation

204
00:11:06,559 --> 00:11:09,720
Speaker 1: and radicalization on the platform years ago and left because

205
00:11:09,720 --> 00:11:11,720
Speaker 1: like it was clear they didn't actually give a shit.

206
00:11:12,360 --> 00:11:15,040
Speaker 1: UM And and a lot of the Integrity department people

207
00:11:15,040 --> 00:11:18,600
Speaker 1: are actually like really good people who are a little

208
00:11:18,600 --> 00:11:21,160
Speaker 1: bit optimistic and kind of young and coming and like, Okay,

209
00:11:21,200 --> 00:11:23,280
Speaker 1: I'll make it's my job to make this huge and

210
00:11:23,320 --> 00:11:27,880
Speaker 1: important thing a lot safer. Um. And these people get

211
00:11:28,160 --> 00:11:31,320
Speaker 1: chewed up and spit out very very quickly. Um. And

212
00:11:31,760 --> 00:11:34,600
Speaker 1: the members of the integrity team, Um, we're kind of

213
00:11:34,640 --> 00:11:38,520
Speaker 1: analyzing the impact of weighing angry content so much. And

214
00:11:38,640 --> 00:11:41,640
Speaker 1: some of them noted in July the extra weight given

215
00:11:41,679 --> 00:11:45,240
Speaker 1: to the anger reaction was a huge problem. They recommended

216
00:11:45,240 --> 00:11:47,800
Speaker 1: the company stop weighing at extra in order to stop

217
00:11:47,800 --> 00:11:50,720
Speaker 1: the spread of harmful content. Their own tests showed that

218
00:11:50,840 --> 00:11:53,160
Speaker 1: dialing the weight of anger back to zero so it

219
00:11:53,240 --> 00:11:55,800
Speaker 1: was no more influential than any other reaction, would stop

220
00:11:55,840 --> 00:11:58,960
Speaker 1: rege inducing content from being shared and spread nearly as widely.

221
00:11:59,280 --> 00:12:03,040
Speaker 1: This led to a five percent reduction and hate speech, misinformation, bullying,

222
00:12:03,040 --> 00:12:05,160
Speaker 1: and posts with violent threats. And when you consider how

223
00:12:05,160 --> 00:12:08,000
Speaker 1: many billions of Facebook posts there are that's a lot

224
00:12:08,160 --> 00:12:11,280
Speaker 1: less nasty shit, um, some of which is going to

225
00:12:11,320 --> 00:12:13,440
Speaker 1: translate into real world violence. And again this was kind

226
00:12:13,440 --> 00:12:15,120
Speaker 1: of a limited study, so who knows how it would

227
00:12:15,120 --> 00:12:18,000
Speaker 1: have actually affected things in the long run. So Facebook

228
00:12:18,080 --> 00:12:22,200
Speaker 1: made this well less money question mark. Yeah, this actually

229
00:12:22,280 --> 00:12:24,520
Speaker 1: was kind of a win for them. Um. Facebook did

230
00:12:24,640 --> 00:12:27,760
Speaker 1: make this change. They pushed it out in September. UM,

231
00:12:27,760 --> 00:12:30,400
Speaker 1: and the employees responsible deserve real credit. Again, there's people

232
00:12:30,400 --> 00:12:34,080
Speaker 1: within Um, there's people within Facebook who did things that

233
00:12:34,160 --> 00:12:38,760
Speaker 1: really actually we're good Like changing this was probably like

234
00:12:38,920 --> 00:12:41,920
Speaker 1: made the world a bit healthier. That said, the fact

235
00:12:41,920 --> 00:12:44,839
Speaker 1: that it had been weighted this way for years, you

236
00:12:45,160 --> 00:12:48,440
Speaker 1: don't undo that just by dialing it back now. For

237
00:12:48,520 --> 00:12:51,520
Speaker 1: one thing, anger has become such an aspect of the

238
00:12:51,600 --> 00:12:55,080
Speaker 1: culture of Facebook that even without weighing the anger emoji,

239
00:12:55,160 --> 00:12:57,160
Speaker 1: most of the content that goes viral is still stuff

240
00:12:57,200 --> 00:12:59,400
Speaker 1: that makes pisces people off, because that's just become what

241
00:12:59,440 --> 00:13:02,319
Speaker 1: Facebook is, because that's what they selected for for years.

242
00:13:02,920 --> 00:13:05,880
Speaker 1: Like it Also, like, who knows, like if they've done

243
00:13:05,920 --> 00:13:08,360
Speaker 1: this years ago, if they never waited anger more, it

244
00:13:08,440 --> 00:13:10,840
Speaker 1: might be a very different platform with like a very

245
00:13:10,840 --> 00:13:13,400
Speaker 1: different impact on the brains of for example, our aunts

246
00:13:13,400 --> 00:13:16,840
Speaker 1: and uncles. Um I think that that's really interesting too,

247
00:13:16,880 --> 00:13:20,800
Speaker 1: because that timeline lines up pretty quickly with or pretty

248
00:13:20,800 --> 00:13:22,800
Speaker 1: exactly with where it feels like a lot of younger

249
00:13:22,840 --> 00:13:27,600
Speaker 1: people were leaving that platform and the platforms became associated

250
00:13:27,679 --> 00:13:30,559
Speaker 1: with older people. Because I feel like I don't think

251
00:13:30,559 --> 00:13:34,880
Speaker 1: I was using Facebook consistently after twenties seventeen. I want

252
00:13:34,880 --> 00:13:37,240
Speaker 1: to say it was maybe my last Facebook year. Yeah

253
00:13:37,320 --> 00:13:41,200
Speaker 1: I stopped. I mean I I stopped visiting it super

254
00:13:41,200 --> 00:13:45,800
Speaker 1: regularly a while back. Um yeah, maybe around So in

255
00:13:45,920 --> 00:13:49,880
Speaker 1: April of Facebook employees came up with another recommendation. This

256
00:13:49,920 --> 00:13:52,920
Speaker 1: one wouldn't be as successful as as changing you know,

257
00:13:53,000 --> 00:13:57,000
Speaker 1: the reaction of the algorithm to the angry reaction spurred

258
00:13:57,000 --> 00:13:58,960
Speaker 1: by the lockdown on the sudden surge of q and

259
00:13:58,960 --> 00:14:02,680
Speaker 1: on Boogaloo and he lockdown groups urging real world violence.

260
00:14:02,920 --> 00:14:05,880
Speaker 1: It was suggested by internal employees that the news feed

261
00:14:05,880 --> 00:14:09,040
Speaker 1: algorithm de prioritize the posting of content based on the

262
00:14:09,080 --> 00:14:12,040
Speaker 1: behavior of people's Facebook friends. So the basic idea is this,

263
00:14:12,200 --> 00:14:16,160
Speaker 1: um a new what Facebook was doing was you would

264
00:14:16,240 --> 00:14:19,200
Speaker 1: if normally like the way you'd think it would work, right,

265
00:14:19,280 --> 00:14:21,760
Speaker 1: is that like your friends post something and you see

266
00:14:21,760 --> 00:14:23,880
Speaker 1: that in your news feed, right, like the posts of

267
00:14:23,880 --> 00:14:25,960
Speaker 1: the people that you've chosen to follow and say are

268
00:14:26,000 --> 00:14:28,400
Speaker 1: your friends? Right? That's how you you would you would

269
00:14:28,400 --> 00:14:30,080
Speaker 1: want it to work. Is that what worked? At one

270
00:14:30,080 --> 00:14:32,920
Speaker 1: point they made a change a few years back where

271
00:14:33,160 --> 00:14:35,800
Speaker 1: they started sending you things not because someone you followed

272
00:14:35,840 --> 00:14:38,840
Speaker 1: had said something, but because they'd like to think um,

273
00:14:38,960 --> 00:14:41,800
Speaker 1: or they'd commented like not so even commented just like

274
00:14:41,800 --> 00:14:43,840
Speaker 1: liked a thing, Like if they'd reacted to a thing

275
00:14:44,240 --> 00:14:47,360
Speaker 1: you would get you would get that sent to your

276
00:14:47,400 --> 00:14:51,240
Speaker 1: news feed UM. And members of the Integrity Team start

277
00:14:51,280 --> 00:14:54,160
Speaker 1: to recognize, like, this has some problems um in it

278
00:14:54,240 --> 00:14:56,160
Speaker 1: from one thing, it results in a lot of people

279
00:14:56,200 --> 00:15:00,520
Speaker 1: getting exposed to dangerous bullshit UM. So they uh, they

280
00:15:01,160 --> 00:15:04,080
Speaker 1: start looking into like the impact of this and how

281
00:15:04,520 --> 00:15:06,880
Speaker 1: how just sharing the kind of things your friends are

282
00:15:06,920 --> 00:15:09,400
Speaker 1: reacting to influences what you see and what that does

283
00:15:09,440 --> 00:15:13,520
Speaker 1: to you on Facebook. The Integrity Team experimented without changing

284
00:15:13,520 --> 00:15:15,840
Speaker 1: this might work, and their early experiments found that fixing

285
00:15:15,880 --> 00:15:19,920
Speaker 1: this would reduce the spread of violence, insight and content UM.

286
00:15:20,000 --> 00:15:23,640
Speaker 1: For one thing, what they found is that like normally

287
00:15:23,800 --> 00:15:27,400
Speaker 1: if you hadn't seen someone like a post about something

288
00:15:27,400 --> 00:15:29,720
Speaker 1: that was maybe like violent or aggressive, your conspiratory, like

289
00:15:29,760 --> 00:15:32,160
Speaker 1: a flat earth post, your appost urging the execution of

290
00:15:32,160 --> 00:15:34,680
Speaker 1: an elected leader. If you hadn't seen anyone that you

291
00:15:34,760 --> 00:15:37,440
Speaker 1: knew react to that post, even if you saw it,

292
00:15:37,480 --> 00:15:39,880
Speaker 1: you wouldn't comment on it or share it. But they

293
00:15:39,880 --> 00:15:41,920
Speaker 1: found that, like if you just saw that a friend

294
00:15:41,920 --> 00:15:43,840
Speaker 1: had liked it, you were more likely to share it,

295
00:15:43,880 --> 00:15:47,360
Speaker 1: which increases exponentially the spread of this kind of violent content.

296
00:15:47,680 --> 00:15:50,760
Speaker 1: And it's it's this idea, like the whole people weren't

297
00:15:50,760 --> 00:15:53,240
Speaker 1: stopped being afraid to be racist at a certain points

298
00:15:53,400 --> 00:15:55,360
Speaker 1: as much as they had been earlier, and it led

299
00:15:55,360 --> 00:15:58,080
Speaker 1: to this surge in real world violence. It is kind

300
00:15:58,080 --> 00:16:01,160
Speaker 1: of the same thing people felt by seeing their friends

301
00:16:01,200 --> 00:16:03,800
Speaker 1: react to this, they felt permission to react to it too.

302
00:16:03,800 --> 00:16:05,240
Speaker 1: In a way, maybe they would have liked, well, I

303
00:16:05,280 --> 00:16:07,360
Speaker 1: don't want to like, maybe I'm interested in flat earthship,

304
00:16:07,360 --> 00:16:08,720
Speaker 1: but I'm just going to ignore this because like I

305
00:16:08,760 --> 00:16:10,720
Speaker 1: don't want to seem like a kok that is so

306
00:16:10,840 --> 00:16:14,560
Speaker 1: fucking upsetting and and fascinating in the way that it

307
00:16:14,600 --> 00:16:16,840
Speaker 1: affects your mind. Is is Yeah, there there was a

308
00:16:16,880 --> 00:16:23,880
Speaker 1: time where you would if you were you know, racist, misogynist, homophobic,

309
00:16:23,920 --> 00:16:26,440
Speaker 1: whatever you were, but you just didn't talk about it.

310
00:16:26,480 --> 00:16:29,680
Speaker 1: But then all of a sudden, there's this confirmation that like, hey,

311
00:16:29,680 --> 00:16:32,040
Speaker 1: this person you know and see all the time feels

312
00:16:32,040 --> 00:16:34,880
Speaker 1: the same fucking way you do. So why be quiet

313
00:16:34,960 --> 00:16:38,640
Speaker 1: about it. Let's discuss. Like it's just that's so dark.

314
00:16:38,800 --> 00:16:41,480
Speaker 1: It's really dark. And so the the the integrity team

315
00:16:41,520 --> 00:16:44,040
Speaker 1: sees this and they're like, we should change this. Um,

316
00:16:44,120 --> 00:16:46,280
Speaker 1: we should only show We shouldn't be showing people just

317
00:16:46,360 --> 00:16:48,680
Speaker 1: like the reactions their friends have had to content because

318
00:16:48,720 --> 00:16:52,720
Speaker 1: it seems to be bad for everybody. Um. And they

319
00:16:52,760 --> 00:16:54,480
Speaker 1: do find in some of their you know, because when

320
00:16:54,480 --> 00:16:57,080
Speaker 1: the experiment, they're like that, we'll take this country or

321
00:16:57,120 --> 00:16:59,120
Speaker 1: this city and we'll we'll roll this change out in

322
00:16:59,120 --> 00:17:01,520
Speaker 1: this limited geograph location to like try and see how

323
00:17:01,560 --> 00:17:03,920
Speaker 1: it might affect its scale. And they do this and

324
00:17:03,960 --> 00:17:07,000
Speaker 1: they see that like, oh, changing this significantly reduces the

325
00:17:07,040 --> 00:17:10,320
Speaker 1: spread of specifically violence and citing contents. So they're like, hey,

326
00:17:10,359 --> 00:17:14,879
Speaker 1: we should roll this out service wide. Zuckerberg himself steps in.

327
00:17:14,960 --> 00:17:19,119
Speaker 1: According to Francis Hoggan, the whistleblower and quote rejected this

328
00:17:19,200 --> 00:17:21,280
Speaker 1: intervention that could have reduced the risk of violence in

329
00:17:21,320 --> 00:17:25,920
Speaker 1: the election from The Atlantic Quote, an internal message characterizing

330
00:17:25,960 --> 00:17:28,639
Speaker 1: Zuckerberg's reasoning says he wanted to avoid new features that

331
00:17:28,680 --> 00:17:31,960
Speaker 1: would get in the way of meaningful social interactions. But

332
00:17:32,000 --> 00:17:35,600
Speaker 1: according to Facebook's definition, its employees say engagement is considered

333
00:17:35,640 --> 00:17:38,240
Speaker 1: meaningful even if it entails bullying, hate speech, and re

334
00:17:38,320 --> 00:17:42,080
Speaker 1: shares of harmful content. The episode, like Facebook's response to

335
00:17:42,080 --> 00:17:45,360
Speaker 1: the incitement that proliferated between the election in January six,

336
00:17:45,520 --> 00:17:48,800
Speaker 1: reveals a fundamental problem with the platform. Facebook's mega scale

337
00:17:48,800 --> 00:17:51,080
Speaker 1: allows the company to influence the speech and thought patterns

338
00:17:51,080 --> 00:17:53,200
Speaker 1: of billions of people. What the world is seeing now

339
00:17:53,200 --> 00:17:55,560
Speaker 1: through the window provided by reams of internal documents is

340
00:17:55,600 --> 00:17:58,439
Speaker 1: that Facebook catalogs and studies the harm it inflicts on people,

341
00:17:58,680 --> 00:18:01,800
Speaker 1: and then it keeps harming anyway. See that that is

342
00:18:01,840 --> 00:18:06,200
Speaker 1: so that's always um so interesting to hear. And by interesting,

343
00:18:06,240 --> 00:18:12,560
Speaker 1: I mean, you know, psychologically harmful, but uh, because it's like, yes,

344
00:18:12,680 --> 00:18:14,760
Speaker 1: that is a fundamental flaw of the platform, but that's

345
00:18:14,800 --> 00:18:17,639
Speaker 1: also very entrenched into like what the DNA of the

346
00:18:17,680 --> 00:18:23,480
Speaker 1: platform always was, which was based on harshly judging other people.

347
00:18:23,600 --> 00:18:28,000
Speaker 1: Like that's why Mark Zuckerberg created Facebook, was to harshly

348
00:18:28,080 --> 00:18:33,439
Speaker 1: judge women in his community. So it's like, I I

349
00:18:33,520 --> 00:18:36,600
Speaker 1: know that it is you know, on a bajillion scale

350
00:18:36,600 --> 00:18:39,440
Speaker 1: at this point, but I'm always kind of stunned at

351
00:18:40,080 --> 00:18:43,280
Speaker 1: how people are like, oh, it's so weird that this

352
00:18:43,359 --> 00:18:45,679
Speaker 1: went you know, the way that it did. It's like, well,

353
00:18:45,920 --> 00:18:50,239
Speaker 1: to an extent um, it was always like that, and

354
00:18:50,280 --> 00:18:52,600
Speaker 1: maybe it was like cosplaying as not being like that,

355
00:18:52,680 --> 00:18:55,640
Speaker 1: and for certain people there were eras in Facebook where

356
00:18:55,640 --> 00:18:58,399
Speaker 1: your user experience wouldn't be like that. But it you know,

357
00:18:58,680 --> 00:19:00,960
Speaker 1: this goes back almost twenty is at this point of

358
00:19:02,480 --> 00:19:06,840
Speaker 1: this being in the DNA of this um this ship show. Yeah,

359
00:19:06,960 --> 00:19:10,440
Speaker 1: and it's it's really bleak. Um. It's just really bleak.

360
00:19:10,480 --> 00:19:13,760
Speaker 1: And it also goes to show like the It was

361
00:19:13,800 --> 00:19:16,040
Speaker 1: one of the things Zuckerberg will say repeatedly when he

362
00:19:16,440 --> 00:19:18,640
Speaker 1: talks about when he when he does admit like, yes,

363
00:19:18,640 --> 00:19:20,960
Speaker 1: there are problems and there have been like negatives associated

364
00:19:20,960 --> 00:19:23,280
Speaker 1: with the site, and we're we're aware of that. They're humbling,

365
00:19:23,320 --> 00:19:25,399
Speaker 1: but like, you know, you also have to include all

366
00:19:25,440 --> 00:19:27,040
Speaker 1: the good that we're doing, all of the meaning and

367
00:19:27,040 --> 00:19:28,639
Speaker 1: the way he always phrases this is like all of

368
00:19:28,640 --> 00:19:31,480
Speaker 1: the meaningful social interactions that wouldn't have happened otherwise. And

369
00:19:31,480 --> 00:19:34,160
Speaker 1: then you realize when every time he says that meaningful

370
00:19:34,200 --> 00:19:37,439
Speaker 1: social interactions that have taken place on Facebook, when he

371
00:19:37,480 --> 00:19:41,400
Speaker 1: says he's including as these internal blocuments, he includes bullying

372
00:19:41,440 --> 00:19:43,600
Speaker 1: and people like making death threats and like talking about

373
00:19:43,600 --> 00:19:46,560
Speaker 1: their desire to murder people like, that's a meaningful interaction.

374
00:19:46,600 --> 00:19:49,399
Speaker 1: People getting angry and trying to inside violence together is

375
00:19:49,440 --> 00:19:53,400
Speaker 1: a meaningful social interaction, which I guess yet I mean

376
00:19:53,560 --> 00:19:57,320
Speaker 1: not meaningless that it that has meaning. Plan meetings were

377
00:19:57,520 --> 00:20:01,359
Speaker 1: meaningful social interactions, you know, Um, you gotta give the

378
00:20:01,440 --> 00:20:07,399
Speaker 1: KKK that. Uh, the Nuremberg rally was meaning interacting. The

379
00:20:07,480 --> 00:20:10,639
Speaker 1: last meaningful interaction I had on Twitter led to like

380
00:20:11,040 --> 00:20:14,160
Speaker 1: a rebound. I was dating coming to my grandma's funeral,

381
00:20:14,280 --> 00:20:21,920
Speaker 1: blackout drunk, so you know, just man, it's been too

382
00:20:21,960 --> 00:20:24,640
Speaker 1: long since I've shown up at a funeral just too

383
00:20:24,720 --> 00:20:28,960
Speaker 1: drunk to stand. It was still a one of my

384
00:20:29,000 --> 00:20:31,639
Speaker 1: favorite mag with my family to this day. They're like,

385
00:20:31,680 --> 00:20:33,840
Speaker 1: who is this guy? And I'm like, I don't really know.

386
00:20:35,280 --> 00:20:39,639
Speaker 1: He's drunk. He came on the megabus. He did the

387
00:20:39,680 --> 00:20:46,080
Speaker 1: megabus getting drunk from a camel back on the megabus. Yeah,

388
00:20:46,119 --> 00:20:48,040
Speaker 1: that would be when I used to do a lot

389
00:20:48,080 --> 00:20:50,320
Speaker 1: of bus trips, like when I was traveling and stuff.

390
00:20:50,320 --> 00:20:52,040
Speaker 1: That would be one of the tactics as you feel

391
00:20:52,080 --> 00:20:56,919
Speaker 1: like a thermist or a camel back with like cranberry juice,

392
00:20:57,040 --> 00:21:01,520
Speaker 1: six liquor and and just oh, I mean get that.

393
00:21:02,200 --> 00:21:05,680
Speaker 1: It's I'm not above getting funked up on a mega

394
00:21:06,600 --> 00:21:09,119
Speaker 1: you know, on your way to my grandma's funeral. That was.

395
00:21:09,440 --> 00:21:11,919
Speaker 1: That was a move. Me and my friends got like

396
00:21:12,119 --> 00:21:15,120
Speaker 1: wasted in San Francisco one day, just like going shopping

397
00:21:15,200 --> 00:21:17,359
Speaker 1: in broad daylight with a camel back, where we would

398
00:21:17,560 --> 00:21:20,479
Speaker 1: we would get a bottle of orange flavored Trader Joe's

399
00:21:20,600 --> 00:21:23,520
Speaker 1: Patron tequila, and we would get a half dozen lime

400
00:21:24,000 --> 00:21:26,720
Speaker 1: popsicles and you just throw the popsicles in with the

401
00:21:26,760 --> 00:21:29,040
Speaker 1: patron and the camel back, and throughout the day it

402
00:21:29,160 --> 00:21:31,760
Speaker 1: melts and you just have constant cold margarita. It's actually

403
00:21:31,800 --> 00:21:39,240
Speaker 1: fucking amazing that fucking we were ito. Yeah, I recommend

404
00:21:39,359 --> 00:21:43,080
Speaker 1: it heavily. You will get trashed and people don't notice

405
00:21:43,359 --> 00:21:45,959
Speaker 1: dude walking around with a camel back in San Francisco.

406
00:21:46,520 --> 00:21:49,880
Speaker 1: Nobody gives a ship. Oh my god, you're basically camouflaged.

407
00:21:50,080 --> 00:21:54,479
Speaker 1: Yeah there, you know who else is camouflaged, The products

408
00:21:54,480 --> 00:21:57,520
Speaker 1: and services that support this podcast, camouflaged to be more

409
00:21:57,600 --> 00:22:00,320
Speaker 1: likable to you by being wrapped in a package of

410
00:22:00,320 --> 00:22:03,080
Speaker 1: of the three of us. That's how ads work. I

411
00:22:03,119 --> 00:22:05,360
Speaker 1: thought you were saying that you were taking ads from

412
00:22:05,400 --> 00:22:08,440
Speaker 1: the U. S. Army recruitment center. Again, I mean, it's

413
00:22:08,560 --> 00:22:13,480
Speaker 1: entirely possible. Um but but but at the moment, we're

414
00:22:13,480 --> 00:22:16,399
Speaker 1: just camouflaging. I don't know whoever, whoever comes on next,

415
00:22:16,440 --> 00:22:19,639
Speaker 1: whoever comes on next, you'll feel more positively about because

416
00:22:19,880 --> 00:22:23,280
Speaker 1: of our presence here. That's how ads work. It's good stuff.

417
00:22:29,160 --> 00:22:32,720
Speaker 1: Oh we're back. My goodness, What a good time we're

418
00:22:32,720 --> 00:22:36,200
Speaker 1: all having today. How are you doing? You make it okay?

419
00:22:36,920 --> 00:22:39,480
Speaker 1: You made it sounds sarcastic. I'm having I am having

420
00:22:39,480 --> 00:22:41,680
Speaker 1: a good time. I'm glad. I'm happy that you're having

421
00:22:41,680 --> 00:22:44,040
Speaker 1: a good time. That's that's my only goal for for

422
00:22:44,080 --> 00:22:46,359
Speaker 1: this show and for you. That's a good time. See

423
00:22:46,400 --> 00:22:48,680
Speaker 1: now you're doubling down on it, and I'm getting insecure

424
00:22:48,800 --> 00:22:50,919
Speaker 1: doubling down. And I'm also talking more and more like

425
00:22:50,960 --> 00:22:53,879
Speaker 1: an NPR talking head as I as I get quieter

426
00:22:54,200 --> 00:22:55,800
Speaker 1: by the bed. Now I'm going to start having a

427
00:22:55,880 --> 00:22:58,399
Speaker 1: panic attack. I've never heard you talk. I know. This

428
00:22:58,480 --> 00:22:59,919
Speaker 1: is how I talk to my cats when I may

429
00:23:00,040 --> 00:23:04,119
Speaker 1: grey at them there. Honestly, I feel like we do

430
00:23:04,200 --> 00:23:06,080
Speaker 1: have that dynamic. I feel like I'm a cat that

431
00:23:06,119 --> 00:23:08,680
Speaker 1: you get angry at something. Yeah, because you you jump

432
00:23:08,720 --> 00:23:12,080
Speaker 1: on my desk and knock over my zvia. It's infurious attention,

433
00:23:12,320 --> 00:23:15,159
Speaker 1: I know. But I've got to work to keep you

434
00:23:15,200 --> 00:23:17,480
Speaker 1: in expensive cat food. I only feed my cats the

435
00:23:17,560 --> 00:23:19,840
Speaker 1: nice wet food. I would rather have your attention than

436
00:23:20,280 --> 00:23:26,080
Speaker 1: really nice food. That's not what my cats say. Um. So,

437
00:23:26,280 --> 00:23:28,920
Speaker 1: there's just a shipload to say about how Facebook negatively

438
00:23:29,040 --> 00:23:32,800
Speaker 1: impacts the increasingly violent political discourse in the United States

439
00:23:32,800 --> 00:23:35,199
Speaker 1: and how they helped make January six happen. But I

440
00:23:35,240 --> 00:23:37,399
Speaker 1: think the way I'd like to illustrate the harm of

441
00:23:37,600 --> 00:23:40,640
Speaker 1: Facebook next is a bit less political. Um it also

442
00:23:40,640 --> 00:23:43,240
Speaker 1: occurs in a different Facebook product. I'm talking about Facebook,

443
00:23:43,240 --> 00:23:45,280
Speaker 1: the company generally want her to Facebook. But now we're

444
00:23:45,280 --> 00:23:48,360
Speaker 1: going to talk about Instagram. In Part one, I mentioned

445
00:23:48,359 --> 00:23:51,600
Speaker 1: that young people felt that removing likes from Instagram temporarily

446
00:23:51,640 --> 00:23:55,120
Speaker 1: corresponded with a decrease in social anxiety. The impact of Instagram,

447
00:23:55,160 --> 00:23:57,840
Speaker 1: specifically on the mental health of kids and teens can

448
00:23:57,840 --> 00:24:01,199
Speaker 1: be incredibly significant. One of the other Facebook internal studies

449
00:24:01,200 --> 00:24:03,480
Speaker 1: that was released as part of the Facebook papers was

450
00:24:03,520 --> 00:24:07,080
Speaker 1: conducted by researchers on Instagram UH. The study, which again

451
00:24:07,119 --> 00:24:09,080
Speaker 1: almost certainly would never have seen the light of day

452
00:24:09,119 --> 00:24:11,960
Speaker 1: if a whistleblower hadn't released, it found that thirty two

453
00:24:11,960 --> 00:24:14,880
Speaker 1: percent of teen girls reported Instagram made their made them

454
00:24:14,880 --> 00:24:18,600
Speaker 1: feel worse about their body. Twenty two million teenagers in

455
00:24:18,600 --> 00:24:21,560
Speaker 1: the United States log onto Instagram on like a daily basis.

456
00:24:21,920 --> 00:24:25,719
Speaker 1: So that's millions of teen girls feeling worse about their

457
00:24:25,720 --> 00:24:32,320
Speaker 1: body because of Instagram. I've never been less surprised. Well,

458
00:24:32,359 --> 00:24:36,520
Speaker 1: good news, it gets worse like no fucking kidding. These

459
00:24:36,560 --> 00:24:40,720
Speaker 1: researchers released their findings internally on in March of noting

460
00:24:40,800 --> 00:24:43,400
Speaker 1: that comparisons on Instagram can change how young women view

461
00:24:43,400 --> 00:24:46,720
Speaker 1: and describe themselves. Again not surprising. So company researchers have

462
00:24:46,760 --> 00:24:49,239
Speaker 1: been investigating the way that Instagram works though for for

463
00:24:49,320 --> 00:24:52,960
Speaker 1: quite a while years um about three years that they've

464
00:24:53,160 --> 00:24:55,800
Speaker 1: been doing this seriously, and their previous findings all back

465
00:24:55,880 --> 00:24:59,440
Speaker 1: up the same central issues. Photos sharing in particular is

466
00:24:59,520 --> 00:25:03,720
Speaker 1: harmful the teen girls nineteen report concluded we make body

467
00:25:03,760 --> 00:25:06,960
Speaker 1: image issues worse for one in three teen girls. Its

468
00:25:07,000 --> 00:25:10,920
Speaker 1: findings included this damning line, teens blame Instagram for increases

469
00:25:10,960 --> 00:25:14,639
Speaker 1: in anxiety and depression. This reaction was unprompted and consistent

470
00:25:14,680 --> 00:25:18,840
Speaker 1: across all groups. So like they almost always mentioned that

471
00:25:18,920 --> 00:25:21,480
Speaker 1: this app specifically makes them feel worse about their body,

472
00:25:21,480 --> 00:25:23,159
Speaker 1: and we don't have to prompt them at all, Like

473
00:25:23,200 --> 00:25:25,640
Speaker 1: this just comes up when they talk about Instagram. I mean,

474
00:25:25,720 --> 00:25:30,000
Speaker 1: that's that, truly, It's so Sophie. I don't know how

475
00:25:30,040 --> 00:25:32,040
Speaker 1: you feel. I mean, I truly think that because I've

476
00:25:32,040 --> 00:25:37,400
Speaker 1: been on Instagram since what like show earlier. I think

477
00:25:37,440 --> 00:25:39,560
Speaker 1: I liked that on earlier. It was around when we

478
00:25:39,560 --> 00:25:43,040
Speaker 1: were in high school. I truly think my my life

479
00:25:43,160 --> 00:25:46,480
Speaker 1: and my relationship to my body would be very different

480
00:25:46,880 --> 00:25:48,520
Speaker 1: had not been on that app for the better part

481
00:25:48,520 --> 00:25:51,280
Speaker 1: of a decade. Yeah, I mean, I mean, especially when

482
00:25:51,280 --> 00:25:55,000
Speaker 1: they introduced filters. Yeah, we're about to talk about So

483
00:25:56,600 --> 00:25:58,919
Speaker 1: here's here's the kicker, And by kicker, I mean the

484
00:25:58,960 --> 00:26:04,439
Speaker 1: bleakest part. In teens who reported suicidal thoughts, percent of

485
00:26:04,440 --> 00:26:06,639
Speaker 1: teens in the UK and six percent of the teams

486
00:26:06,640 --> 00:26:09,760
Speaker 1: in the United States claim their desire to kill themselves

487
00:26:09,960 --> 00:26:14,520
Speaker 1: started on Instagram. That's fucking disgusting and terror. That's pretty bleak.

488
00:26:15,040 --> 00:26:19,200
Speaker 1: More than I just like I wish I were more surprise. Yeah,

489
00:26:20,440 --> 00:26:22,680
Speaker 1: but it's good to have this data. Um. The data

490
00:26:22,720 --> 00:26:24,760
Speaker 1: shows that more than forty percent of Instagram so more

491
00:26:24,760 --> 00:26:28,200
Speaker 1: than of Instagram users are less than twenty two years old. UM,

492
00:26:28,240 --> 00:26:31,120
Speaker 1: which means you've got twenty two million teens logging onto

493
00:26:31,119 --> 00:26:33,359
Speaker 1: the service in the US every day. Six percent of

494
00:26:33,359 --> 00:26:37,000
Speaker 1: those people becoming suicidal as the result of Instagram is

495
00:26:37,040 --> 00:26:39,960
Speaker 1: one point three two million children who once started wanting

496
00:26:39,960 --> 00:26:43,720
Speaker 1: to kill themselves while using Instagram. Hey, everybody, Robert Evans here,

497
00:26:43,760 --> 00:26:46,040
Speaker 1: and I actually screwed up the math that I just cited,

498
00:26:46,080 --> 00:26:47,960
Speaker 1: which is often the case when I do math. So

499
00:26:48,000 --> 00:26:50,560
Speaker 1: anytime I do math of my own in an episode,

500
00:26:50,920 --> 00:26:54,640
Speaker 1: you're right to question me. UM. I was calculating six

501
00:26:54,680 --> 00:26:59,600
Speaker 1: percent of twenty two million basically, Um. But as the

502
00:27:00,000 --> 00:27:02,520
Speaker 1: auddy noted, it's six percent of kids who are suicidal

503
00:27:02,680 --> 00:27:05,040
Speaker 1: say that their suicidal feelings started on Instagram, So I

504
00:27:05,080 --> 00:27:09,360
Speaker 1: wanted to recalculate that, um about seventy to seventy six

505
00:27:09,400 --> 00:27:11,719
Speaker 1: kind of depending on the source. Percent of American teens

506
00:27:12,160 --> 00:27:16,560
Speaker 1: use Instagram. UM there are about forty two million teenagers

507
00:27:16,600 --> 00:27:22,000
Speaker 1: in the United States um SO I calculated from that,

508
00:27:22,160 --> 00:27:26,200
Speaker 1: and about eighteen percent of about nineteen percent of high

509
00:27:26,200 --> 00:27:31,320
Speaker 1: school students of of like teenagers UM seriously considered attempting suicide. So,

510
00:27:31,359 --> 00:27:34,600
Speaker 1: if we're just counting serious attempts at or people who

511
00:27:34,640 --> 00:27:40,040
Speaker 1: seriously considered attempting suicide, that's UM five millions, seven thousand,

512
00:27:40,080 --> 00:27:44,200
Speaker 1: six hundred teens who seriously considered suicide six percent of those.

513
00:27:44,280 --> 00:27:47,280
Speaker 1: If six percent of those kids had their suicidal feelings

514
00:27:47,280 --> 00:27:50,560
Speaker 1: start on Instagram, that's three hundred and forty four thousand,

515
00:27:50,640 --> 00:27:53,240
Speaker 1: seven hundred and thirty six children in the United States

516
00:27:53,280 --> 00:27:57,320
Speaker 1: who suicidal feelings started on Instagram. UM and I've furthermore

517
00:27:57,320 --> 00:28:00,359
Speaker 1: found that about nine percent of kids who seriously attempt

518
00:28:00,359 --> 00:28:04,960
Speaker 1: suicide or seriously consider suicide attempted. So of that three

519
00:28:05,000 --> 00:28:10,880
Speaker 1: hundred and uh thousand, seven thirty six American kid teens

520
00:28:11,359 --> 00:28:16,440
Speaker 1: whose suicidal feelings started on Instagram, about thirty one thousand,

521
00:28:17,080 --> 00:28:21,600
Speaker 1: uh and twenty six kids attempt suicide. Um SO, about

522
00:28:21,640 --> 00:28:25,480
Speaker 1: thirty one kids in the United States on an annual

523
00:28:25,520 --> 00:28:30,080
Speaker 1: basis attempts suicide because of suicidal feelings that started on Instagram.

524
00:28:30,200 --> 00:28:34,680
Speaker 1: So that is the more accurate look at the data.

525
00:28:35,040 --> 00:28:37,520
Speaker 1: And I apologize as always for the error. But what's

526
00:28:37,560 --> 00:28:41,000
Speaker 1: interesting is that these studies do document like Facebook has

527
00:28:40,800 --> 00:28:45,000
Speaker 1: is as physically harmful at scale, is like a wide

528
00:28:45,080 --> 00:28:48,760
Speaker 1: variety of narcotics, like most narcotics probably are less harmful

529
00:28:48,800 --> 00:28:52,360
Speaker 1: at scale physically than Instagram. Um, I think weeds certainly

530
00:28:52,440 --> 00:28:56,400
Speaker 1: is um so my god, if if every teenager was

531
00:28:56,440 --> 00:29:00,600
Speaker 1: smoking weed instead of doom scrolling Instagram, the world would

532
00:29:00,640 --> 00:29:03,160
Speaker 1: just be so funny. If they were, they were chained

533
00:29:03,200 --> 00:29:09,760
Speaker 1: smoking cigars instead of being on Instagram. It's so weird

534
00:29:09,760 --> 00:29:13,800
Speaker 1: because I think about, like how I don't know whatever.

535
00:29:14,040 --> 00:29:16,200
Speaker 1: Like I'm in my late twenties, so I feel like

536
00:29:16,240 --> 00:29:18,400
Speaker 1: I have like a little bit of memory of like

537
00:29:19,160 --> 00:29:25,000
Speaker 1: what life was like before you were constantly being encouraged

538
00:29:25,040 --> 00:29:29,160
Speaker 1: to compare yourself to every single person you've ever met

539
00:29:29,160 --> 00:29:31,520
Speaker 1: in your life, regardless of whether you know who they are,

540
00:29:31,560 --> 00:29:36,720
Speaker 1: how they are whatever. Um, and I just call me nostalgic,

541
00:29:36,960 --> 00:29:40,400
Speaker 1: but uh, I liked how I felt better. Yeah, Like

542
00:29:41,080 --> 00:29:43,680
Speaker 1: it's so absurd how much I know about people I

543
00:29:43,720 --> 00:29:46,680
Speaker 1: don't give a shit about, and how bad it makes

544
00:29:46,680 --> 00:29:49,960
Speaker 1: me feel to know about the curated lives of people

545
00:29:49,960 --> 00:29:51,760
Speaker 1: that I don't give a shit about, and how I

546
00:29:51,880 --> 00:29:54,960
Speaker 1: let that actively affect my daily life. And it's just yeah,

547
00:29:55,000 --> 00:29:58,000
Speaker 1: it's just fucking miserable. It is. It's horrible. It's horrible.

548
00:29:58,320 --> 00:30:01,280
Speaker 1: Um that said, I like Florida on the application, so

549
00:30:01,360 --> 00:30:05,040
Speaker 1: you know, it's sure. Now Here's why despite the documented

550
00:30:05,120 --> 00:30:09,360
Speaker 1: harm that Instagram does, nothing's ever going to change. Um.

551
00:30:09,400 --> 00:30:12,680
Speaker 1: As I stated, million US teens used to Graham daily,

552
00:30:13,120 --> 00:30:16,880
Speaker 1: only five million log on to Facebook. So Instagram is

553
00:30:16,920 --> 00:30:20,720
Speaker 1: almost five times is popular among teenagers as Facebook, where

554
00:30:20,800 --> 00:30:26,200
Speaker 1: kids are leaving in droves. So Facebook Mark Zuckerberg's invention

555
00:30:26,400 --> 00:30:28,840
Speaker 1: is now definitively just the terrain of the olds. And

556
00:30:28,840 --> 00:30:30,720
Speaker 1: Facebook knows that kids are never going to come back,

557
00:30:30,760 --> 00:30:33,680
Speaker 1: because that's not how how it being a kid works

558
00:30:33,720 --> 00:30:35,800
Speaker 1: like that, You don't get them back. They're going to

559
00:30:35,840 --> 00:30:38,640
Speaker 1: continue to do new ship. Eventually they'll leave Instagram for

560
00:30:38,760 --> 00:30:41,280
Speaker 1: something else. You know, That's just the way it fucking goes.

561
00:30:41,360 --> 00:30:45,200
Speaker 1: Unless unless the thirty year nostalgic cycle is like Facebook

562
00:30:45,240 --> 00:30:48,480
Speaker 1: is actually back now, it's actually and I don't think

563
00:30:48,520 --> 00:30:50,920
Speaker 1: it gave anybody a good experience enough to have that

564
00:30:51,040 --> 00:30:53,720
Speaker 1: it's not the fucking teenage mutant ninja turtle. Yeah. No,

565
00:30:53,760 --> 00:30:56,120
Speaker 1: one's getting dopamine hits. That's a good Yeah, it's like

566
00:30:56,200 --> 00:30:59,200
Speaker 1: flame and hot cheet does. Nobody's thinking fondly back to

567
00:30:59,360 --> 00:31:02,240
Speaker 1: scrolling Facebook when they were seven. They're thinking back to

568
00:31:02,400 --> 00:31:07,880
Speaker 1: I don't know, SpongeBob square Pants. Um oh, but at

569
00:31:07,880 --> 00:31:10,840
Speaker 1: the moment, Instagram is very popular with teens, and Facebook

570
00:31:10,840 --> 00:31:12,479
Speaker 1: knows that if they're going to continue to grow and

571
00:31:12,520 --> 00:31:15,600
Speaker 1: maintain their cultural dominance, they have to keep bringing in

572
00:31:15,640 --> 00:31:18,640
Speaker 1: the teens. They have to keep Instagram as profitable and

573
00:31:18,720 --> 00:31:22,360
Speaker 1: as as addictive as it currently is. Um. And that's

574
00:31:22,400 --> 00:31:24,160
Speaker 1: why they bought Instagram in the first place. They only

575
00:31:24,160 --> 00:31:25,880
Speaker 1: paid like a billion dollars for it. It It was an

576
00:31:25,880 --> 00:31:31,720
Speaker 1: incredible investment um. And they spend in it in a day. Yeah,

577
00:31:31,800 --> 00:31:34,800
Speaker 1: that's that's cheap as hell for something as influential and

578
00:31:34,880 --> 00:31:38,680
Speaker 1: huge as Instagram. Yeah. I wonder do you know what

579
00:31:38,720 --> 00:31:41,680
Speaker 1: it's worth now? I would guess significantly more than a

580
00:31:41,680 --> 00:31:44,479
Speaker 1: billion dollars um, But I don't entirely know a lot

581
00:31:44,520 --> 00:31:47,920
Speaker 1: of value. But Facebook's like a trillion dollar company now. Yeah,

582
00:31:47,960 --> 00:31:51,200
Speaker 1: they're very sucks and it's well, but face but that

583
00:31:51,240 --> 00:31:55,080
Speaker 1: includes Instagram, you know. Okay, so yeah, and among you know,

584
00:31:55,160 --> 00:31:57,880
Speaker 1: teens are one of the most valuable demographics to have

585
00:31:58,000 --> 00:32:02,400
Speaker 1: for advertisers, and Instagram is where quarantine. The number it's

586
00:32:02,560 --> 00:32:07,960
Speaker 1: estimated value is two billion. Yeah, that's a good investmentsment,

587
00:32:08,440 --> 00:32:13,040
Speaker 1: good investment if money you gotta yeah. Um So. The

588
00:32:13,080 --> 00:32:15,360
Speaker 1: fact that so much is at stake with Instagram, the

589
00:32:15,360 --> 00:32:17,640
Speaker 1: fact that it's such a central part of company having

590
00:32:17,680 --> 00:32:19,480
Speaker 1: any kind of future, is part of why Mark and

591
00:32:19,600 --> 00:32:22,480
Speaker 1: company have been so compelled to lie about it. None

592
00:32:22,520 --> 00:32:24,720
Speaker 1: of this stuff that we've been talking about was released

593
00:32:24,760 --> 00:32:27,400
Speaker 1: when Facebook researchers got it, of course not. They wouldn't

594
00:32:27,400 --> 00:32:30,920
Speaker 1: want anyone to know this ship. In March, Mark took

595
00:32:30,960 --> 00:32:33,200
Speaker 1: to Congress, where he was criticized for his plans to

596
00:32:33,200 --> 00:32:36,400
Speaker 1: create a new Instagram service for children under thirteen. He

597
00:32:36,440 --> 00:32:40,280
Speaker 1: was asked if he'd studied how Instagram infects affects children,

598
00:32:40,360 --> 00:32:43,120
Speaker 1: and he said, I believe the answer is yes. So

599
00:32:43,240 --> 00:32:46,880
Speaker 1: not yes, I think we've studied that. He told them. Then,

600
00:32:47,240 --> 00:32:49,560
Speaker 1: the research we've seen is that using social apps to

601
00:32:49,560 --> 00:32:52,360
Speaker 1: connect with other people can have positive mental health benefits.

602
00:32:52,880 --> 00:32:57,480
Speaker 1: And I'm sure there's something that he's gotten paid researchers

603
00:32:57,520 --> 00:32:59,360
Speaker 1: to come up with that he can make that case

604
00:32:59,400 --> 00:33:02,280
Speaker 1: off of. Um, I'm sure in certain situations it may

605
00:33:02,320 --> 00:33:04,560
Speaker 1: even be true. There are ways you can use social

606
00:33:04,560 --> 00:33:07,200
Speaker 1: media that are good. Dear, myth I've legitimately smiled or

607
00:33:07,240 --> 00:33:09,640
Speaker 1: had my heart warmed by things that happened on social media.

608
00:33:09,640 --> 00:33:12,720
Speaker 1: It doesn't not happen. Um, And I do think that

609
00:33:12,760 --> 00:33:14,840
Speaker 1: there is a case for like I mean, and it's

610
00:33:15,160 --> 00:33:17,520
Speaker 1: you can't credit Mark Zuckerberg with it, but just I

611
00:33:17,560 --> 00:33:20,080
Speaker 1: mean going back to fucking like live journal days of

612
00:33:20,240 --> 00:33:23,640
Speaker 1: just like friendships that have deepened as a result of

613
00:33:24,480 --> 00:33:28,200
Speaker 1: social media. That's definitely a thing. But the costs that

614
00:33:28,320 --> 00:33:32,440
Speaker 1: way the benefits there by quite a bit. Yeah, Um,

615
00:33:32,440 --> 00:33:35,880
Speaker 1: it's it's great. So um. So Mark goes on to say,

616
00:33:36,200 --> 00:33:38,640
Speaker 1: you know, I I think we've got research that shows

617
00:33:38,640 --> 00:33:40,680
Speaker 1: that can have positive mental health effects. You know, I

618
00:33:40,720 --> 00:33:44,080
Speaker 1: think we've studied whether or not how it affects children. Um.

619
00:33:44,120 --> 00:33:46,520
Speaker 1: But he doesn't talk about He leaves out all the

620
00:33:46,560 --> 00:33:49,200
Speaker 1: stuff that all the statistics, like about all the kids

621
00:33:49,240 --> 00:33:52,920
Speaker 1: whose suicidal ideation starts on Instagram, they had that data

622
00:33:52,960 --> 00:33:55,400
Speaker 1: when he went before Congress. He just didn't mention it.

623
00:33:55,440 --> 00:33:58,120
Speaker 1: They hadn't told anyone that ship, Like, he didn't say

624
00:33:58,160 --> 00:34:00,560
Speaker 1: a goddamn word about it. Yeah, he was like, yeah,

625
00:34:00,560 --> 00:34:01,840
Speaker 1: I think we've looked into it, and you know, there's

626
00:34:01,840 --> 00:34:03,760
Speaker 1: some ways in which it can be healthy not And

627
00:34:03,880 --> 00:34:07,000
Speaker 1: also one point three million American kids became suicidal because

628
00:34:07,040 --> 00:34:10,480
Speaker 1: of our app Like he did not throw that anything,

629
00:34:10,719 --> 00:34:13,560
Speaker 1: Like did he throw that? I mean truly, I'm like

630
00:34:13,680 --> 00:34:15,640
Speaker 1: up in the air of like did he not say

631
00:34:15,680 --> 00:34:18,920
Speaker 1: that because he didn't want people to know? Or did

632
00:34:18,960 --> 00:34:20,480
Speaker 1: he just say that because he heard it and he

633
00:34:20,520 --> 00:34:23,319
Speaker 1: didn't care and he forgot, Like you just don't know

634
00:34:23,400 --> 00:34:26,480
Speaker 1: what that guy that is so fucking evil. Woah, it's

635
00:34:26,480 --> 00:34:29,080
Speaker 1: pretty great. Um, and we'll talk more about that later.

636
00:34:29,360 --> 00:34:33,400
Speaker 1: In May, Instagram boss Adam Mussari told reporters that he

637
00:34:33,400 --> 00:34:36,279
Speaker 1: thought any impact on teen well being by Instagram was

638
00:34:36,360 --> 00:34:39,759
Speaker 1: likely quote quite small, based on the internal research he'd seen. Again,

639
00:34:39,880 --> 00:34:42,320
Speaker 1: I haven't released this research. He's saying, Oh, we have research,

640
00:34:42,360 --> 00:34:44,080
Speaker 1: and it says that any kind of impact on well

641
00:34:44,120 --> 00:34:47,520
Speaker 1: being is pretty small. And again, the actual research by

642
00:34:47,560 --> 00:34:50,480
Speaker 1: this point show of kids in the UK and six

643
00:34:50,520 --> 00:34:52,400
Speaker 1: percent of kids in the United States were moved to

644
00:34:52,400 --> 00:34:54,920
Speaker 1: thoughts of suicide by Instagram, which I would not call small.

645
00:34:55,480 --> 00:34:57,480
Speaker 1: I would not call I wouldn't I wouldn't necessarily say

646
00:34:57,480 --> 00:35:00,400
Speaker 1: it's huge. But that is not a small impact. Um, No,

647
00:35:00,719 --> 00:35:06,920
Speaker 1: that is like thousands and thousands, and yeah, that's significant.

648
00:35:07,480 --> 00:35:10,080
Speaker 1: The Wall Street Journal caught up with Massari after the

649
00:35:10,080 --> 00:35:12,799
Speaker 1: Facebook papers leaked, so they were able to like drill

650
00:35:12,880 --> 00:35:15,240
Speaker 1: him on this a bit, and he said a bit more. Quote,

651
00:35:15,560 --> 00:35:17,520
Speaker 1: in no way do I mean to diminish these issues.

652
00:35:17,560 --> 00:35:20,120
Speaker 1: Some of the issues mentioned in this story aren't necessarily widespread,

653
00:35:20,160 --> 00:35:23,160
Speaker 1: but their impact on people may be huge, which is like,

654
00:35:23,600 --> 00:35:27,680
Speaker 1: again a perfect nonstatement. That's right. They're like, but what

655
00:35:27,719 --> 00:35:30,600
Speaker 1: about the thing we couldn't possibly gauge at all versus

656
00:35:30,640 --> 00:35:33,480
Speaker 1: the thing that we did and we're actively distancing ourselves from.

657
00:35:33,520 --> 00:35:36,399
Speaker 1: I mean those statistics that's like at least one kid

658
00:35:36,440 --> 00:35:40,480
Speaker 1: in every classroom, Like that is gigantic. And when you

659
00:35:40,520 --> 00:35:43,200
Speaker 1: read the responses of guys like Massari and compare them

660
00:35:43,200 --> 00:35:45,520
Speaker 1: the responsive guy like people like Mark Zuckerberg and official

661
00:35:45,520 --> 00:35:47,680
Speaker 1: corporate spokes people, it's it's very clear that they're working

662
00:35:47,719 --> 00:35:51,120
Speaker 1: from the same playbook, that they're very disciplined in their responses.

663
00:35:51,440 --> 00:35:54,359
Speaker 1: Because Massari does try to tell the journal um that

664
00:35:54,440 --> 00:35:57,160
Speaker 1: he he thinks Facebook was late to realizing there were

665
00:35:57,200 --> 00:35:59,600
Speaker 1: drawbacks and connecting people in such large numbers. But then

666
00:35:59,600 --> 00:36:01,360
Speaker 1: he says, I've been pushing very hard for us to

667
00:36:01,360 --> 00:36:05,319
Speaker 1: embrace our responsibilities more broadly, which again says nothing. He

668
00:36:05,400 --> 00:36:08,560
Speaker 1: then pivots from that to stating that he's actually really

669
00:36:08,600 --> 00:36:10,600
Speaker 1: proud of the research they've done in the mental health

670
00:36:10,600 --> 00:36:13,000
Speaker 1: effects on teams, which again they didn't share with anybody,

671
00:36:13,040 --> 00:36:15,520
Speaker 1: and I would argue light about by omission in front

672
00:36:15,560 --> 00:36:18,839
Speaker 1: of Congress. Um, He's proud of this because he says

673
00:36:18,880 --> 00:36:21,840
Speaker 1: that shows Facebook employees are asking tough questions about the platform.

674
00:36:22,080 --> 00:36:24,440
Speaker 1: Quote for me, this isn't dirty laundry. I'm actually very

675
00:36:24,480 --> 00:36:26,840
Speaker 1: proud of this research, which is the same thing Zuckerberg

676
00:36:26,880 --> 00:36:30,200
Speaker 1: said about his own employees damning the service after six Right,

677
00:36:30,200 --> 00:36:32,840
Speaker 1: I'm gonna say that's the same exact thing as the

678
00:36:33,080 --> 00:36:36,440
Speaker 1: as the as the like actually bad work, you know,

679
00:36:36,480 --> 00:36:38,759
Speaker 1: talking about how that working for the Death Star is bad,

680
00:36:39,000 --> 00:36:42,520
Speaker 1: um is like evidence of Oh, the Death Stars actually

681
00:36:42,560 --> 00:36:46,520
Speaker 1: has a really open work culture. Like No, I don't know,

682
00:36:46,960 --> 00:36:49,400
Speaker 1: I feel like there are a few. There are not

683
00:36:49,480 --> 00:36:54,160
Speaker 1: many CEOs that are good at flipping a narrative. But

684
00:36:54,480 --> 00:36:58,759
Speaker 1: Mark Zuckerberg is particularly bad at it. Yeah, and it's

685
00:36:58,800 --> 00:37:00,480
Speaker 1: I mean, part of why they can be bad at

686
00:37:00,480 --> 00:37:03,160
Speaker 1: it is it doesn't really matter um, or at least

687
00:37:03,200 --> 00:37:06,440
Speaker 1: it hasn't fucking so far. Um. But the patterns I

688
00:37:06,480 --> 00:37:09,360
Speaker 1: mean not enough to get a better figurehead like, Yeah,

689
00:37:09,920 --> 00:37:13,879
Speaker 1: the patterns pretty clear here. Um. When a scandal comes out,

690
00:37:13,960 --> 00:37:17,279
Speaker 1: deny it until the information that can't be denied leaks out,

691
00:37:17,600 --> 00:37:20,600
Speaker 1: and then claim that whatever is happening at the site,

692
00:37:20,640 --> 00:37:23,680
Speaker 1: whatever like our information you had about how harmful it is,

693
00:37:23,680 --> 00:37:25,479
Speaker 1: is a positive because it means that you were trying

694
00:37:25,520 --> 00:37:28,040
Speaker 1: to do stuff about it, even if you actually rejected

695
00:37:28,120 --> 00:37:30,320
Speaker 1: taking action based on the data you had and refused

696
00:37:30,360 --> 00:37:33,680
Speaker 1: to share it with anybody else. Massari and Zuckerberg were

697
00:37:33,680 --> 00:37:36,400
Speaker 1: also careful to reiterate that any harms from Instagram had

698
00:37:36,400 --> 00:37:38,560
Speaker 1: to be weighed against its benefits, which I haven't found

699
00:37:38,600 --> 00:37:40,919
Speaker 1: a ton of documentation on. In fact, as the Wall

700
00:37:40,920 --> 00:37:44,480
Speaker 1: Street Journal rights. In five presentations over eighteen months to

701
00:37:44,560 --> 00:37:47,719
Speaker 1: this spring, the researchers Facebook researchers conducted what they called

702
00:37:47,760 --> 00:37:50,240
Speaker 1: a teen mental health deep dive and follow up studies.

703
00:37:50,480 --> 00:37:52,440
Speaker 1: They came to the conclusion that some of the problems

704
00:37:52,440 --> 00:37:55,280
Speaker 1: were specific to Instagram and not social media. More broadly,

705
00:37:55,560 --> 00:37:58,640
Speaker 1: This is especially true concerning so called social comparison, which

706
00:37:58,640 --> 00:38:00,560
Speaker 1: is when people assess their own value you in relation

707
00:38:00,600 --> 00:38:03,920
Speaker 1: to the attractiveness, wealth, and success of others. Social comparison

708
00:38:04,040 --> 00:38:07,120
Speaker 1: is worse on Instagram, states Facebook's deep dive into Team

709
00:38:07,160 --> 00:38:10,479
Speaker 1: Girl body image issues in noting that TikTok, a short

710
00:38:10,600 --> 00:38:13,520
Speaker 1: video app, is grounded in performance, while users on Snapchat,

711
00:38:13,600 --> 00:38:16,279
Speaker 1: a rifle photo and video sharing app, are sheltered by

712
00:38:16,360 --> 00:38:20,160
Speaker 1: jokey features that keep the focus on the face. In contrast,

713
00:38:20,200 --> 00:38:23,520
Speaker 1: Instagram more focuses more heavily on the body and lifestyle.

714
00:38:23,880 --> 00:38:27,439
Speaker 1: March Internal Research states it warns that the Explore page,

715
00:38:27,480 --> 00:38:30,840
Speaker 1: which serves users photos and videos curated by an algorithm,

716
00:38:31,040 --> 00:38:33,440
Speaker 1: can send users deep into content that can be harmful.

717
00:38:33,800 --> 00:38:37,080
Speaker 1: Aspects of Instagram exacerbate each other to create a perfect storm.

718
00:38:37,160 --> 00:38:40,719
Speaker 1: The research states, Yeah, I mean, again, not a shocking

719
00:38:40,760 --> 00:38:46,960
Speaker 1: revelation over here there. It is. I mean, I and

720
00:38:47,000 --> 00:38:50,480
Speaker 1: I do think that that let's TikTok and Snapchat get

721
00:38:50,480 --> 00:38:55,839
Speaker 1: off easy. They're like they're certain there is absolutely uh

722
00:38:56,080 --> 00:38:58,719
Speaker 1: toxic body image culture on there, And I feel like

723
00:38:58,800 --> 00:39:02,720
Speaker 1: finspo will thrive on any platform it fucking gloms itself onto.

724
00:39:02,800 --> 00:39:06,520
Speaker 1: But Instagram is particularly bad because it's like where so

725
00:39:06,560 --> 00:39:11,080
Speaker 1: many lifestyle people have launched and and there's so many

726
00:39:11,200 --> 00:39:14,520
Speaker 1: headless women on Instagram. It's it is shocking. There's so

727
00:39:14,560 --> 00:39:18,319
Speaker 1: many like not like um, not like you macheted my

728
00:39:18,360 --> 00:39:21,879
Speaker 1: head off, but like you're not encouraged to show your

729
00:39:21,920 --> 00:39:26,680
Speaker 1: head by the algorithm, which sounds weird, but it is true.

730
00:39:26,800 --> 00:39:29,879
Speaker 1: The less like it is just very focused on how

731
00:39:29,960 --> 00:39:33,400
Speaker 1: you physically look. And then there's also this tendency to

732
00:39:33,480 --> 00:39:37,480
Speaker 1: like tear people apart if they have edited their body

733
00:39:37,520 --> 00:39:39,400
Speaker 1: to look a certain way, when it's like, well that

734
00:39:39,680 --> 00:39:42,400
Speaker 1: the algorithm rewards editing your body to look a certain

735
00:39:42,440 --> 00:39:45,319
Speaker 1: way and to do all this, and it's you do

736
00:39:45,440 --> 00:39:47,600
Speaker 1: bring up a good point where it's like going, it's

737
00:39:47,640 --> 00:39:52,960
Speaker 1: frustrating that it's important to critique Facebook in relation to

738
00:39:53,000 --> 00:39:57,160
Speaker 1: its competitors like TikTok and snapchat. Um. That can lead

739
00:39:57,200 --> 00:40:00,160
Speaker 1: to the uncomfortable situation of like seeming to praise them

740
00:40:00,160 --> 00:40:02,160
Speaker 1: when they haven't done a good job. They just haven't

741
00:40:02,200 --> 00:40:05,360
Speaker 1: been as irresponsible. It's kind of like attacking like Chevron.

742
00:40:05,880 --> 00:40:08,280
Speaker 1: If you look at all of the overall harms, including

743
00:40:08,320 --> 00:40:11,160
Speaker 1: like their impact and like covering up climate change. Maybe

744
00:40:11,200 --> 00:40:13,000
Speaker 1: the worst of the big oil and gas companies. I

745
00:40:13,000 --> 00:40:14,839
Speaker 1: don't know, it's debatable, but it's like if you're if

746
00:40:14,840 --> 00:40:18,320
Speaker 1: you're criticizing Chevron specifically, it doesn't you're not saying that

747
00:40:18,600 --> 00:40:20,960
Speaker 1: BP is great. You're just being like, well, these are

748
00:40:20,960 --> 00:40:23,400
Speaker 1: the guys specifically that did this bad thing, and they

749
00:40:23,440 --> 00:40:26,560
Speaker 1: were the leaders in this specific terrible thing. Other bad

750
00:40:26,600 --> 00:40:28,960
Speaker 1: things are going on, but we can't, like, the episode

751
00:40:28,960 --> 00:40:31,240
Speaker 1: can't be about how bad everyone. We're talking about Facebook

752
00:40:31,320 --> 00:40:33,399
Speaker 1: right now, we have these documents from inside Facebook. I'm

753
00:40:33,400 --> 00:40:36,279
Speaker 1: sure versions of this are happening everywhere else. Listeners, in

754
00:40:36,320 --> 00:40:41,520
Speaker 1: your everyday life, just don't use Facebook as a yardstick

755
00:40:41,800 --> 00:40:44,760
Speaker 1: for morality, you know, because they just end up letting

756
00:40:44,760 --> 00:40:47,080
Speaker 1: a lot of people off for a lot of funked

757
00:40:47,120 --> 00:40:50,200
Speaker 1: up stuff. I would say in your regular life, don't

758
00:40:50,280 --> 00:40:53,959
Speaker 1: use Facebook. Is all the sentence we need it there. Um. Wow,

759
00:40:55,920 --> 00:40:58,359
Speaker 1: So you asked me you were talking earlier about like

760
00:40:58,840 --> 00:41:01,160
Speaker 1: because Mark went up and of Congress and was like, yeah,

761
00:41:01,200 --> 00:41:03,160
Speaker 1: I think we've got research on this, and I've definitely

762
00:41:03,160 --> 00:41:06,480
Speaker 1: seen research that says it's good for kids. We know everything.

763
00:41:06,520 --> 00:41:08,759
Speaker 1: I just stated that quote. I just read everything like

764
00:41:08,840 --> 00:41:12,080
Speaker 1: that's that's in those internal studies. Um, we know that

765
00:41:12,120 --> 00:41:14,000
Speaker 1: Mark saw this. We know that it was viewed by

766
00:41:14,000 --> 00:41:17,360
Speaker 1: top Facebook leaders because it was mentioned in a presentation

767
00:41:17,400 --> 00:41:20,480
Speaker 1: that was given to Mark Zuckerberg himself. We know that

768
00:41:20,560 --> 00:41:24,960
Speaker 1: when in August, Senators Richard Blumenthal and Marshall Blackburn sent

769
00:41:25,000 --> 00:41:27,160
Speaker 1: a letter to Mark Zuckerberg asking him to release his

770
00:41:27,200 --> 00:41:30,440
Speaker 1: internal research on how platforms impact child mental health, we

771
00:41:30,560 --> 00:41:32,600
Speaker 1: know that he sent back a six page letter that

772
00:41:32,640 --> 00:41:35,880
Speaker 1: included none of the studies we've just mentioned. Instead, the

773
00:41:35,920 --> 00:41:38,040
Speaker 1: study said that it was hard to conduct research on

774
00:41:38,080 --> 00:41:40,399
Speaker 1: Instagram and that there was no consensus about how much

775
00:41:40,440 --> 00:41:43,200
Speaker 1: screen time is too much. Meanwhile, their own data show

776
00:41:43,280 --> 00:41:46,680
Speaker 1: that Instagram users who reported feeling unattractive said that the

777
00:41:46,719 --> 00:41:50,200
Speaker 1: feeling began while they were on Instagram. Facebook's own internal

778
00:41:50,200 --> 00:41:52,759
Speaker 1: reports showed that their users reported wanting to spend less

779
00:41:52,760 --> 00:41:55,680
Speaker 1: time on Instagram but couldn't make themselves. And here's a

780
00:41:55,719 --> 00:41:58,719
Speaker 1: quote that makes it sound like Heroin teens told us

781
00:41:58,719 --> 00:42:00,560
Speaker 1: they don't like the amount of time they spend on

782
00:42:00,600 --> 00:42:02,360
Speaker 1: the app, but feel like they have to be present.

783
00:42:02,440 --> 00:42:04,640
Speaker 1: They often feel addicted and know that what they're seeing

784
00:42:04,719 --> 00:42:06,640
Speaker 1: is bad for their mental health but feel unable to

785
00:42:06,680 --> 00:42:11,080
Speaker 1: stop themselves. That's Facebook writing about Instagram like that's that's

786
00:42:11,200 --> 00:42:13,480
Speaker 1: their own people saying this like this is not some

787
00:42:13,640 --> 00:42:16,880
Speaker 1: activists getting in here, you know. That's so I mean,

788
00:42:17,080 --> 00:42:20,280
Speaker 1: it's I guess, good on them, regardless of the level

789
00:42:20,320 --> 00:42:24,239
Speaker 1: of self awareness going on there. Um, that's I mean.

790
00:42:24,239 --> 00:42:26,399
Speaker 1: And what I was thinking about earlier when it comes

791
00:42:26,440 --> 00:42:29,719
Speaker 1: to any time Zuckerberg is in front of Congress or

792
00:42:29,760 --> 00:42:32,080
Speaker 1: in front of political officials, I feel like for a

793
00:42:32,200 --> 00:42:36,680
Speaker 1: lot of people, the takeaway and the thing that gets

794
00:42:36,719 --> 00:42:41,760
Speaker 1: trending is how little political officials and members of Congress

795
00:42:42,000 --> 00:42:45,680
Speaker 1: understand about how the Internet works. And that's like the

796
00:42:45,719 --> 00:42:49,640
Speaker 1: funny story is like, oh, Mark Zuckerberg talked about an algorithm,

797
00:42:49,680 --> 00:42:51,839
Speaker 1: and they and you know, like this is this comes

798
00:42:51,920 --> 00:42:54,399
Speaker 1: up all the time, It comes up on v became

799
00:42:54,440 --> 00:42:57,520
Speaker 1: up on succession of just like how not internet literate

800
00:42:57,560 --> 00:43:00,840
Speaker 1: the majority of people who just i'd how the Internet

801
00:43:00,840 --> 00:43:04,239
Speaker 1: works are And it's like it almost becomes like a

802
00:43:04,320 --> 00:43:07,120
Speaker 1: he he ha ha, old guy doesn't know how algorithm works.

803
00:43:07,120 --> 00:43:09,560
Speaker 1: But it's like, well, the consequence of that is that

804
00:43:09,600 --> 00:43:12,560
Speaker 1: it ends up making Mark Zuckerberg look way cooler than

805
00:43:12,600 --> 00:43:14,920
Speaker 1: he is and it also doesn't address the problem at

806
00:43:14,920 --> 00:43:19,799
Speaker 1: all of like, no, Mark Zuckerberg is omitting something gigantic here,

807
00:43:20,040 --> 00:43:23,320
Speaker 1: and the majority of our you know, lawmakers in Congress

808
00:43:23,680 --> 00:43:27,560
Speaker 1: don't have the fucking, you know, cultural vocabulary to even

809
00:43:27,719 --> 00:43:31,919
Speaker 1: understand that. And that is like and and I guess

810
00:43:31,960 --> 00:43:34,160
Speaker 1: it's like it makes for a couple of memes, but

811
00:43:34,239 --> 00:43:38,280
Speaker 1: it's just like, no, this is bad. Can you commit

812
00:43:38,320 --> 00:43:44,799
Speaker 1: to cancel Finsta? Do you remember that who that was? Sad? Right?

813
00:43:44,840 --> 00:43:46,640
Speaker 1: Cancel fits? I mean that I think that's the most

814
00:43:46,760 --> 00:43:50,000
Speaker 1: recent one where it's like, okay, yeah, that is you know,

815
00:43:50,040 --> 00:43:55,360
Speaker 1: objectively funny. But but like the consequence of that is

816
00:43:55,480 --> 00:43:57,919
Speaker 1: I mean, that's ultimately a win for Instagram. And that's

817
00:43:57,920 --> 00:44:00,640
Speaker 1: a win for Facebook because it makes them look like

818
00:44:00,680 --> 00:44:03,719
Speaker 1: they're operating on a level of the fucking government doesn't understand.

819
00:44:04,000 --> 00:44:06,800
Speaker 1: And meanwhile, you know, one kid in every classroom is

820
00:44:06,840 --> 00:44:11,040
Speaker 1: suicidal as a result of the inability of law of

821
00:44:11,320 --> 00:44:15,240
Speaker 1: like law making officials to understand the effect that this has.

822
00:44:15,280 --> 00:44:18,799
Speaker 1: It's just it makes me real mad, Robert. And one

823
00:44:18,800 --> 00:44:21,480
Speaker 1: of the nune things about this is that while these

824
00:44:21,560 --> 00:44:26,080
Speaker 1: lawmakers don't understand and sound like idiots talking to Mark

825
00:44:26,160 --> 00:44:29,400
Speaker 1: Zuckerberg his own employees. These researchers who are part of

826
00:44:29,400 --> 00:44:32,480
Speaker 1: the Integrity team, these researchers studying the impact of Instagram

827
00:44:32,480 --> 00:44:35,680
Speaker 1: on teens know exactly how harmful it is, and they

828
00:44:35,719 --> 00:44:38,239
Speaker 1: are grappling in real time with like the damage their

829
00:44:38,239 --> 00:44:41,520
Speaker 1: product is doing two children. Members of these teams reported

830
00:44:41,560 --> 00:44:44,120
Speaker 1: frustration at the fact that their colleagues often refused to

831
00:44:44,120 --> 00:44:47,399
Speaker 1: take their findings seriously. One former researcher told The Wall

832
00:44:47,400 --> 00:44:50,640
Speaker 1: Street Journal that we're standing directly between people and their

833
00:44:50,640 --> 00:44:54,080
Speaker 1: bonuses when they try to reduce the harmful aspects of Instagram,

834
00:44:54,120 --> 00:44:57,120
Speaker 1: because like anything that reduces the harm is going to

835
00:44:57,239 --> 00:45:00,319
Speaker 1: reduce its popularity. It's going to cut down on time site,

836
00:45:00,320 --> 00:45:03,560
Speaker 1: it's going to cut down on users, and so everyone

837
00:45:03,600 --> 00:45:05,920
Speaker 1: else at Facebook. But the Integrity team gets paid by

838
00:45:05,960 --> 00:45:09,160
Speaker 1: how much they increase engagement on the site. That's where

839
00:45:09,160 --> 00:45:11,799
Speaker 1: the bonuses come from. So if you're trying to make

840
00:45:11,800 --> 00:45:15,239
Speaker 1: Facebook less dangerous, a huge number of the most influential

841
00:45:15,239 --> 00:45:19,120
Speaker 1: employees have a financially invested financial interest in you failing.

842
00:45:19,239 --> 00:45:21,400
Speaker 1: So that's yeah, And what that boils down to is

843
00:45:21,440 --> 00:45:25,040
Speaker 1: like there is a financial incentive for you convincing a

844
00:45:25,120 --> 00:45:28,360
Speaker 1: thirteen year old girl that she should get fucking fillers

845
00:45:28,440 --> 00:45:33,719
Speaker 1: in her, Like, what the fuck? It's just now all

846
00:45:33,719 --> 00:45:35,520
Speaker 1: I'm going to try to sell the teens because it's

847
00:45:35,560 --> 00:45:39,120
Speaker 1: ad break time, Jamie, and I would like to Okay, okay,

848
00:45:39,360 --> 00:45:43,279
Speaker 1: if you're if you're a teenager, you don't need Instagram.

849
00:45:43,280 --> 00:45:45,480
Speaker 1: That ship is bad for you. You certainly don't need Facebook.

850
00:45:45,560 --> 00:45:48,279
Speaker 1: That's where the olds are. But you know what's come

851
00:45:48,320 --> 00:45:50,480
Speaker 1: back around to being cool and what I think your

852
00:45:50,520 --> 00:45:55,440
Speaker 1: generation needs to get on board with smoking huge comical cigars.

853
00:45:55,480 --> 00:45:57,799
Speaker 1: I'm talking the ones that are like nine inches long

854
00:45:57,880 --> 00:46:01,600
Speaker 1: and like three inches white. Massive teens listening right now,

855
00:46:01,920 --> 00:46:04,520
Speaker 1: go out to your local cigar shop. Do not dribe

856
00:46:04,719 --> 00:46:08,319
Speaker 1: a sad ador. The worst looking you find outside the

857
00:46:08,320 --> 00:46:13,399
Speaker 1: cigar shall pay him for huge, cheap cigars and new

858
00:46:13,520 --> 00:46:19,200
Speaker 1: Instagram smoking horrible cigars. This is this is well, this

859
00:46:19,239 --> 00:46:22,960
Speaker 1: is what's going to bring buying teenagers hook zoomers, zoomers.

860
00:46:23,360 --> 00:46:25,799
Speaker 1: The air is not getting any cleaner. Right, You're all

861
00:46:25,800 --> 00:46:28,600
Speaker 1: gonna we're all gonna choke to death on wildfires. You

862
00:46:28,680 --> 00:46:31,760
Speaker 1: might as well burn down a big fat Macanudoh, Robert,

863
00:46:31,760 --> 00:46:33,680
Speaker 1: can I tell you I bought a teenager a white

864
00:46:33,680 --> 00:46:37,319
Speaker 1: claw the other. Good for you, Thank you. I felt good.

865
00:46:37,360 --> 00:46:39,560
Speaker 1: I felt like I did a public service. Yeah, teens,

866
00:46:39,719 --> 00:46:45,240
Speaker 1: go buy those big, fat, ridiculous lunatic cigars. Bribe, bribe

867
00:46:45,400 --> 00:46:49,040
Speaker 1: for it. You're just preparing yourself for climate change. All right,

868
00:46:49,160 --> 00:46:52,560
Speaker 1: here's the other white claw teens. Okay, I mean white

869
00:46:52,560 --> 00:46:55,040
Speaker 1: Cloud goes great with a huge shitty cigar, Jamie, No,

870
00:46:55,200 --> 00:47:00,160
Speaker 1: it did. Absolutely smoking is bad for you, Andy, as

871
00:47:00,160 --> 00:47:03,120
Speaker 1: a white smoking a cigar, you puff it, so it's

872
00:47:03,120 --> 00:47:15,480
Speaker 1: healthy you. Alright, here's some ads. Alright, we're back. We

873
00:47:15,560 --> 00:47:18,440
Speaker 1: are we all just we all just enjoyed a couple

874
00:47:18,480 --> 00:47:22,920
Speaker 1: of really comically large cigars. Um, we did not have

875
00:47:23,280 --> 00:47:27,399
Speaker 1: those ridiculous long asylum cigars. It was great. Why why

876
00:47:27,440 --> 00:47:30,680
Speaker 1: are you fixated on this what is happening? Because I

877
00:47:30,680 --> 00:47:32,520
Speaker 1: I find that sketch from I think you should leave

878
00:47:32,560 --> 00:47:35,279
Speaker 1: while the little girls are talking about smoking five maccanodos

879
00:47:35,280 --> 00:47:38,080
Speaker 1: to unwind at the end of the day. Actually, I

880
00:47:38,120 --> 00:47:42,520
Speaker 1: mean yeah, but like I love when you're reveal yourself

881
00:47:42,520 --> 00:47:46,000
Speaker 1: to be a basic bitch. I am a basic bittix

882
00:47:46,560 --> 00:47:49,160
Speaker 1: A right, That's why I'm thinking about cigar. I love that.

883
00:47:49,280 --> 00:47:51,399
Speaker 1: I love that we're in the middle of a podcast,

884
00:47:51,640 --> 00:47:55,360
Speaker 1: and uh, you can't get off that well. I also

885
00:47:55,400 --> 00:47:58,640
Speaker 1: think making children do things that's bad for them as funny,

886
00:47:58,760 --> 00:48:03,560
Speaker 1: but not this way, not the face flashes send dan flashes.

887
00:48:03,680 --> 00:48:06,120
Speaker 1: I mean they've also I think the teens are rejecting

888
00:48:06,120 --> 00:48:11,520
Speaker 1: in f t s pretty widely, Jamie. So when Facebook

889
00:48:11,520 --> 00:48:13,840
Speaker 1: does try to make the case that their products are benign,

890
00:48:13,960 --> 00:48:17,080
Speaker 1: they like to bring up studies from the Oxford Internet Institute,

891
00:48:17,080 --> 00:48:20,320
Speaker 1: which is a project of Oxford University, which show minimal

892
00:48:20,440 --> 00:48:23,920
Speaker 1: or no correlation between social media use and depression. Uh.

893
00:48:23,960 --> 00:48:26,240
Speaker 1: The Wall Street Journal actually reached out to the Oxford

894
00:48:26,239 --> 00:48:29,200
Speaker 1: researcher responsible for some of these studies, who right away

895
00:48:29,320 --> 00:48:33,359
Speaker 1: was like wasn't like, oh, yes, they're right, everything's fine. Um,

896
00:48:33,400 --> 00:48:35,920
Speaker 1: he was like, actually, Facebook needs to be much more

897
00:48:35,960 --> 00:48:38,200
Speaker 1: open with the research that they're doing because they have

898
00:48:38,280 --> 00:48:41,319
Speaker 1: better data than than we can get, than researchers can get,

899
00:48:41,560 --> 00:48:44,120
Speaker 1: and so our actual information that they're citing is hampered

900
00:48:44,120 --> 00:48:46,200
Speaker 1: by the fact that they're not sharing what they're finding.

901
00:48:46,600 --> 00:48:48,920
Speaker 1: And who knows how things can change and our conclusions

902
00:48:48,920 --> 00:48:52,440
Speaker 1: could change if we had access to all of that data. Um.

903
00:48:52,480 --> 00:48:54,799
Speaker 1: He even told the Wall spre Street Journal. People talk

904
00:48:54,800 --> 00:48:57,400
Speaker 1: about Instagram like it's a drug, but we can't study

905
00:48:57,440 --> 00:49:00,360
Speaker 1: the active ingredient, which you'll notice is not him saying

906
00:49:00,760 --> 00:49:03,279
Speaker 1: it's fine, it's him being like, yeah, I really wish

907
00:49:03,280 --> 00:49:07,000
Speaker 1: we could actually study this better. Um, it's difficult, right.

908
00:49:07,080 --> 00:49:10,279
Speaker 1: And also he's referring to it like drugs, which is

909
00:49:10,719 --> 00:49:14,719
Speaker 1: the comparable scale of how it manifesting. Okay, Yeah, he's

910
00:49:14,719 --> 00:49:18,000
Speaker 1: certainly not being like everything's fine. Um, I think that's clear.

911
00:49:18,719 --> 00:49:22,520
Speaker 1: He's truly like constantly, Mr Policeman. I gave you all

912
00:49:22,520 --> 00:49:25,680
Speaker 1: the closing the situation and just no one gives a ship.

913
00:49:25,719 --> 00:49:29,719
Speaker 1: It is very funny and like that that movie, And

914
00:49:29,760 --> 00:49:31,920
Speaker 1: that's what I was trying to say, that it's hilarious. Ye.

915
00:49:32,120 --> 00:49:34,440
Speaker 1: So we focused a lot on these episodes about how

916
00:49:34,440 --> 00:49:37,520
Speaker 1: Facebook has harmed people and institutions in the United States,

917
00:49:37,719 --> 00:49:40,280
Speaker 1: but as we've covered in past episodes, the social network

918
00:49:40,280 --> 00:49:42,799
Speaker 1: has been responsible for helping to incite ethnic cleansings and

919
00:49:42,840 --> 00:49:46,279
Speaker 1: mass racial violence in places like Myanmar and India. Mob

920
00:49:46,360 --> 00:49:50,240
Speaker 1: violence against Muslims in India and cited by viral Facebook misinformation,

921
00:49:50,560 --> 00:49:53,319
Speaker 1: led one researcher in February of two thousand nineteen to

922
00:49:53,400 --> 00:49:56,080
Speaker 1: create yet another fake account to try and experience social

923
00:49:56,080 --> 00:50:00,000
Speaker 1: media as a person in Kerala, India, might um from

924
00:50:00,040 --> 00:50:02,440
Speaker 1: New York Times quote. For the next three weeks, the

925
00:50:02,440 --> 00:50:05,600
Speaker 1: account operated by a simple rule follow all the recommendations

926
00:50:05,640 --> 00:50:08,600
Speaker 1: generated by Facebook's algorithm to join groups, watch videos, and

927
00:50:08,640 --> 00:50:11,120
Speaker 1: explore new pages on the site. The result was an

928
00:50:11,120 --> 00:50:14,440
Speaker 1: inundation of hate speech, misinformation, and celebrations of violence, which

929
00:50:14,480 --> 00:50:17,600
Speaker 1: were documented in an internal Facebook report published later that month.

930
00:50:18,640 --> 00:50:21,640
Speaker 1: And this is from the Facebook researcher following this test

931
00:50:21,760 --> 00:50:24,360
Speaker 1: users news feed. I've seen more images of dead people

932
00:50:24,400 --> 00:50:26,040
Speaker 1: in the past three weeks than I've seen in my

933
00:50:26,200 --> 00:50:31,880
Speaker 1: entire life total. What a great site Mark built. Facebook's

934
00:50:31,880 --> 00:50:38,560
Speaker 1: new tagline the place for corpses. Oh yeah, my goodness.

935
00:50:38,600 --> 00:50:40,960
Speaker 1: I mean and so I know that we we have

936
00:50:41,520 --> 00:50:46,960
Speaker 1: discussed Facebook's role in in uh super charging ethnic cleansings,

937
00:50:47,040 --> 00:50:50,880
Speaker 1: but that is just that is so. Yeah, it's not great, Jamie.

938
00:50:51,000 --> 00:50:54,080
Speaker 1: Someone wrote that down, Robert and one wrote that down

939
00:50:54,120 --> 00:50:57,600
Speaker 1: and hit published it's not greater or because India is

940
00:50:57,640 --> 00:51:02,520
Speaker 1: Facebook's biggest customer, forty million Indians use one or more

941
00:51:02,560 --> 00:51:06,840
Speaker 1: Facebook products. Um, that's that's a ship of people. Yeah

942
00:51:06,960 --> 00:51:10,600
Speaker 1: three ll um. That is something that I think is

943
00:51:10,640 --> 00:51:14,239
Speaker 1: important to remember and something that I lose sight of

944
00:51:14,320 --> 00:51:19,920
Speaker 1: sometimes is like Facebook is not a super popular platform

945
00:51:20,080 --> 00:51:24,520
Speaker 1: for people of all ages in North America. But that's

946
00:51:24,560 --> 00:51:26,960
Speaker 1: not the case. Yeah, and it is just it is

947
00:51:27,000 --> 00:51:29,920
Speaker 1: the Internet for a lot of these people. Um, Like

948
00:51:30,040 --> 00:51:31,840
Speaker 1: that is the way that that is the whole of

949
00:51:31,880 --> 00:51:34,160
Speaker 1: how they consume the Internet in a lot of cases.

950
00:51:34,200 --> 00:51:36,320
Speaker 1: I mean maybe with like YouTube or something mixed in,

951
00:51:36,320 --> 00:51:37,919
Speaker 1: but they're probably getting a lot of their YouTube links

952
00:51:37,920 --> 00:51:41,239
Speaker 1: from their Facebook feed. Um. Now, the fact the fact

953
00:51:41,239 --> 00:51:43,759
Speaker 1: that India is the number one customer in terms of

954
00:51:43,760 --> 00:51:45,920
Speaker 1: like number of people for Facebook, I'm sure the United

955
00:51:45,960 --> 00:51:48,360
Speaker 1: States is still more profitable just because of like differences

956
00:51:48,400 --> 00:51:51,879
Speaker 1: in income and whatnot. Um, but this is a huge

957
00:51:51,920 --> 00:51:54,400
Speaker 1: part of their business. But despite that fact, they have

958
00:51:54,440 --> 00:51:57,239
Speaker 1: failed to invest very much in terms of meaningful resources

959
00:51:57,239 --> 00:52:00,480
Speaker 1: into having employees who speak the language, whereas is more

960
00:52:00,520 --> 00:52:05,680
Speaker 1: the problem the languages of India. See India super mixed

961
00:52:05,840 --> 00:52:08,239
Speaker 1: country right in terms of different like ethnic groups and

962
00:52:08,239 --> 00:52:11,920
Speaker 1: religious groups. They have twenty two officially recognized languages in

963
00:52:11,960 --> 00:52:13,920
Speaker 1: the country, and there's way more languages than that in

964
00:52:13,920 --> 00:52:16,279
Speaker 1: India that significant numbers of people speak. There's twenty two

965
00:52:16,280 --> 00:52:19,359
Speaker 1: officially recognized languages. Anyone who can travel there, and I've

966
00:52:19,360 --> 00:52:20,840
Speaker 1: spent a lot of time in India can tell you

967
00:52:20,840 --> 00:52:23,160
Speaker 1: that being able to effectively say hello and ask basic

968
00:52:23,239 --> 00:52:25,319
Speaker 1: questions of people can require a lot of research if

969
00:52:25,360 --> 00:52:28,800
Speaker 1: you're traveling a decent amount. But Facebook aren't twenty something

970
00:52:28,840 --> 00:52:31,440
Speaker 1: tourists on the problem for good Tan Dory and bond Lassies.

971
00:52:31,680 --> 00:52:34,560
Speaker 1: They have effectively taken control of the primary method of

972
00:52:34,560 --> 00:52:37,680
Speaker 1: communication and information distribution for hundreds of millions of people,

973
00:52:37,880 --> 00:52:40,239
Speaker 1: and they fed failed to hire folks who might know

974
00:52:40,280 --> 00:52:43,320
Speaker 1: if some of those people are deliberately inciting genocide against

975
00:52:43,320 --> 00:52:47,040
Speaker 1: other people in the country. Eight seven to Facebook's global

976
00:52:47,040 --> 00:52:50,480
Speaker 1: budget for identifying misinformation is spent on the United States.

977
00:52:50,920 --> 00:52:55,040
Speaker 1: The rest of the planet shares of their misinformation budget.

978
00:52:55,200 --> 00:52:57,560
Speaker 1: You want to guess what percentage of Facebook users? North

979
00:52:57,600 --> 00:53:03,400
Speaker 1: Americans make up? Ten percent of their budget goes on

980
00:53:03,480 --> 00:53:07,239
Speaker 1: ten percent of their users. Of like dealing with disinformation

981
00:53:08,160 --> 00:53:12,680
Speaker 1: something else, of dealing with disinformation specifically now, When this

982
00:53:12,760 --> 00:53:15,440
Speaker 1: leaked out, Facebook's response was that the information site it

983
00:53:15,560 --> 00:53:18,520
Speaker 1: was incomplete and did not include third party fact checkers.

984
00:53:18,520 --> 00:53:20,280
Speaker 1: They're like, well, this doesn't include all of the people

985
00:53:20,440 --> 00:53:22,799
Speaker 1: the third party companies we hire, except for the data

986
00:53:22,840 --> 00:53:25,400
Speaker 1: they show suggests that the majority of the effort and

987
00:53:25,480 --> 00:53:27,600
Speaker 1: money spent on third party fact checkers is for fact

988
00:53:27,680 --> 00:53:30,359
Speaker 1: checking stuff in the United States. Um. And of course

989
00:53:30,360 --> 00:53:32,960
Speaker 1: they did not elaborate on how including this information might

990
00:53:33,000 --> 00:53:35,000
Speaker 1: have changed the overall numbers, So my guess is not

991
00:53:35,040 --> 00:53:38,120
Speaker 1: by much of at all. Internal documents do show that

992
00:53:38,160 --> 00:53:41,040
Speaker 1: Facebook attempted to create changes to their platform to stop

993
00:53:41,040 --> 00:53:43,799
Speaker 1: the spread of the disinformation during the November election. In

994
00:53:43,840 --> 00:53:46,800
Speaker 1: myan mar those changes which also halted the spread of

995
00:53:46,840 --> 00:53:49,160
Speaker 1: disinformation put out by the military, which was a big

996
00:53:49,200 --> 00:53:51,719
Speaker 1: like the it was the military inciting ethnic cleansings and

997
00:53:51,760 --> 00:53:53,759
Speaker 1: like and trying to incite violence in order to like

998
00:53:54,280 --> 00:53:58,160
Speaker 1: lockdown political power ahead of this election. Um, so they

999
00:53:58,280 --> 00:54:00,680
Speaker 1: cut this significantly prior to the election. They see it

1000
00:54:00,680 --> 00:54:03,280
Speaker 1: as a problem. The institute changes similar to the changes

1001
00:54:03,320 --> 00:54:05,040
Speaker 1: they talked about putting up in the U. S. If

1002
00:54:05,040 --> 00:54:07,200
Speaker 1: things went badly with the election and these worked, it

1003
00:54:07,320 --> 00:54:11,200
Speaker 1: dropped dramatically, Yeah and again and it's it's that is

1004
00:54:11,360 --> 00:54:13,920
Speaker 1: that is good. I'm glad that was done. But they

1005
00:54:13,960 --> 00:54:17,360
Speaker 1: only respond to give me a second, exclusively, give me

1006
00:54:17,400 --> 00:54:20,360
Speaker 1: a second, Jamie, because prior to the election, they institute

1007
00:54:20,360 --> 00:54:22,799
Speaker 1: these changes which are significant. It reduces the number of

1008
00:54:22,800 --> 00:54:26,160
Speaker 1: inflammatory posts by twenty five point one percent and reduces

1009
00:54:26,200 --> 00:54:29,319
Speaker 1: the spread of photo posts containing disinformation by forty eight

1010
00:54:29,360 --> 00:54:34,279
Speaker 1: point five pc. This is huge. That's that's that's really significant. UM.

1011
00:54:34,320 --> 00:54:37,240
Speaker 1: As soon as the election was done, Facebook reversed those changes,

1012
00:54:37,239 --> 00:54:40,479
Speaker 1: presumably because they were bad for money. Three months after

1013
00:54:40,520 --> 00:54:43,680
Speaker 1: the election, the Myanmar military launched a vicious coup. Violence

1014
00:54:43,719 --> 00:54:46,440
Speaker 1: there continues to this moment. In response, Facebook created a

1015
00:54:46,440 --> 00:54:49,520
Speaker 1: special policy to stop people from praising violence in the country,

1016
00:54:49,840 --> 00:54:52,960
Speaker 1: one which presumably reduces the spread of content by freedom

1017
00:54:53,000 --> 00:54:55,800
Speaker 1: fighters resisting the military as much as it reduces content

1018
00:54:55,840 --> 00:54:59,279
Speaker 1: spread by the military. It's obviously too much to say

1019
00:54:59,280 --> 00:55:01,799
Speaker 1: that Facebook call is to coup and Myanmar ship's been.

1020
00:55:01,840 --> 00:55:03,360
Speaker 1: I mean, there's a lot going on there. I'm not

1021
00:55:03,400 --> 00:55:06,200
Speaker 1: gonna I'm not pretending that this is like it's just Facebook,

1022
00:55:06,239 --> 00:55:10,520
Speaker 1: but a major contributing factor. It was an insignificant UM

1023
00:55:10,560 --> 00:55:13,680
Speaker 1: and the fact that they knew how much their policies

1024
00:55:13,719 --> 00:55:17,120
Speaker 1: were helping and reverse them after the election, reversing this

1025
00:55:17,160 --> 00:55:20,400
Speaker 1: effect and leading to an increase in inflammatory content because

1026
00:55:20,440 --> 00:55:23,799
Speaker 1: it profited them more is damning, right, That's the thing

1027
00:55:23,840 --> 00:55:28,200
Speaker 1: that's damning. UM Around the world. Facebook's contribution to violence

1028
00:55:28,200 --> 00:55:30,640
Speaker 1: maybe greatest in places where the company has huge reach

1029
00:55:30,680 --> 00:55:33,560
Speaker 1: but pays little attention. In Sri Lanka, people were able

1030
00:55:33,600 --> 00:55:36,360
Speaker 1: to automatically add hundreds of thousands of users to Facebook

1031
00:55:36,360 --> 00:55:40,160
Speaker 1: groups that spread violent content. In Ethiopia, a nationalist militia

1032
00:55:40,440 --> 00:55:43,760
Speaker 1: coordinated call sur violence openly on the app. The company

1033
00:55:43,760 --> 00:55:45,760
Speaker 1: claims that it has reduced the amount of hate speech

1034
00:55:45,760 --> 00:55:48,160
Speaker 1: people see globally by half this year. But even if

1035
00:55:48,200 --> 00:55:50,400
Speaker 1: that is true, how much hate was spread during the

1036
00:55:50,480 --> 00:55:52,600
Speaker 1: years where they ignored the rest of the world. How

1037
00:55:52,640 --> 00:55:55,400
Speaker 1: many killings, how many militant groups seeded with new recruits,

1038
00:55:55,480 --> 00:55:58,760
Speaker 1: how many pieces of extermination? Is propaganda spread while Facebook

1039
00:55:58,800 --> 00:56:03,120
Speaker 1: just wasn't paying attention? The actual answer is likely incalculable.

1040
00:56:03,160 --> 00:56:05,080
Speaker 1: But here's The New York Times again reporting on that

1041
00:56:05,120 --> 00:56:09,680
Speaker 1: test account in Kerala, India, perfect turn of friends. Yeah yeah.

1042
00:56:09,800 --> 00:56:12,160
Speaker 1: Ten days after the researcher opened the fake account to

1043
00:56:12,160 --> 00:56:15,560
Speaker 1: study misinformation, a suicide bombing and the disputed border region

1044
00:56:15,560 --> 00:56:17,360
Speaker 1: of cashmir set off a round of violence and a

1045
00:56:17,400 --> 00:56:21,960
Speaker 1: spike and accusations misinformation and conspiracies between Indian and Pakistani nationals.

1046
00:56:22,320 --> 00:56:25,239
Speaker 1: After the attack, anti Pakistan content began to circulate in

1047
00:56:25,239 --> 00:56:28,160
Speaker 1: the Facebook recommendation groups that the researcher had joined. Many

1048
00:56:28,200 --> 00:56:30,520
Speaker 1: of the groups she noted had tens of thousands of followers.

1049
00:56:30,640 --> 00:56:33,600
Speaker 1: A different report by Facebook published in December two thousand nineteen,

1050
00:56:33,600 --> 00:56:36,719
Speaker 1: found Indian Facebook users tended to join large groups, with

1051
00:56:36,760 --> 00:56:39,080
Speaker 1: the company's median group size at a hundred and forty

1052
00:56:39,080 --> 00:56:42,360
Speaker 1: thousand members. In a separate report produced after the elections,

1053
00:56:42,360 --> 00:56:46,160
Speaker 1: Facebook found that over fort of top views or impressions

1054
00:56:46,160 --> 00:56:48,440
Speaker 1: in the Indian state of West Bengal were fake or

1055
00:56:48,480 --> 00:56:51,640
Speaker 1: inauthentic when one in authentic account had amassed more than

1056
00:56:51,680 --> 00:56:55,000
Speaker 1: thirty million impressions. A report in March showed that many

1057
00:56:55,040 --> 00:56:57,960
Speaker 1: of the problems cited during the two thousand nineteen elections persisted.

1058
00:56:58,239 --> 00:57:01,720
Speaker 1: In the internal document called Adverse Serial Harmful Networks India

1059
00:57:01,800 --> 00:57:04,400
Speaker 1: Case Study, a Facebook researcher wrote that there were groups

1060
00:57:04,400 --> 00:57:08,120
Speaker 1: and pages replete with inflammatory and misleading anti Muslim content

1061
00:57:08,200 --> 00:57:10,560
Speaker 1: on Facebook. The report said that there were a number

1062
00:57:10,560 --> 00:57:14,080
Speaker 1: of dehumanizing posts comparing Muslims to pigs and dogs, and

1063
00:57:14,120 --> 00:57:16,720
Speaker 1: misinformation claimed that the Koran, the Holy Book of Islam,

1064
00:57:16,800 --> 00:57:19,520
Speaker 1: calls from men to rape their female family members, so

1065
00:57:20,120 --> 00:57:23,640
Speaker 1: that's significant. Like the scale at which the ship spreads

1066
00:57:23,880 --> 00:57:29,600
Speaker 1: is huge, and and I I mean, I don't even

1067
00:57:29,720 --> 00:57:33,360
Speaker 1: I mean I feel like I know the answer if

1068
00:57:33,400 --> 00:57:37,640
Speaker 1: if the hate is existing on that scale unmitigated. But

1069
00:57:39,440 --> 00:57:42,520
Speaker 1: who is working to? Like how many people does Facebook

1070
00:57:42,560 --> 00:57:46,760
Speaker 1: have working on? Is there is there an integrity team

1071
00:57:47,120 --> 00:57:51,600
Speaker 1: for this regent? Like technically yes, the question is how

1072
00:57:51,600 --> 00:57:54,000
Speaker 1: many of them and how many of the languages there

1073
00:57:54,000 --> 00:57:57,240
Speaker 1: are represented by the team. And it's not many exactly,

1074
00:57:57,320 --> 00:57:59,880
Speaker 1: Like it's not many. Can't have a global company and

1075
00:58:00,040 --> 00:58:03,560
Speaker 1: not have global representation or ship like this is going

1076
00:58:03,640 --> 00:58:06,520
Speaker 1: to happen, Like it's just it's actually, you know what.

1077
00:58:06,600 --> 00:58:08,760
Speaker 1: It kind of reminds me of Jamie. I was looking

1078
00:58:08,760 --> 00:58:10,360
Speaker 1: at this, and I was thinking about the East India

1079
00:58:10,400 --> 00:58:14,280
Speaker 1: Trading Company UM. When the East India Company took over

1080
00:58:14,400 --> 00:58:17,360
Speaker 1: large chunks of India UM. They took it over from

1081
00:58:17,360 --> 00:58:21,520
Speaker 1: a regime. The government, the monarchical government that had been

1082
00:58:21,520 --> 00:58:24,040
Speaker 1: in charge in that area prior, was not a good government,

1083
00:58:24,080 --> 00:58:25,920
Speaker 1: right because they number one, they lost that war. But

1084
00:58:25,960 --> 00:58:28,440
Speaker 1: like they weren't a very good government, they were a government,

1085
00:58:28,480 --> 00:58:31,320
Speaker 1: so they did do things like provide aid and famines

1086
00:58:31,360 --> 00:58:34,080
Speaker 1: and disasters and have people whose job it was to

1087
00:58:34,120 --> 00:58:36,680
Speaker 1: like handle stuff like that and like handle like make

1088
00:58:36,720 --> 00:58:38,920
Speaker 1: sure that like place the stuff was getting where it

1089
00:58:38,960 --> 00:58:42,040
Speaker 1: needed to go during like calamities and whatnot, and doing

1090
00:58:42,080 --> 00:58:44,640
Speaker 1: things specifically that helped people. But we're not we're not

1091
00:58:44,720 --> 00:58:46,720
Speaker 1: profitable because a big chunk of what a government does

1092
00:58:46,880 --> 00:58:50,040
Speaker 1: isn't directly profitable. It's just helping to like keep people

1093
00:58:50,040 --> 00:58:54,160
Speaker 1: alive and keep the roads open and whatnot, right sustain humanity.

1094
00:58:54,520 --> 00:58:57,400
Speaker 1: When the East India Company took over, they were governing

1095
00:58:57,440 --> 00:58:59,640
Speaker 1: and in control of this region and this has actually

1096
00:58:59,680 --> 00:59:01,800
Speaker 1: been all I think is their first place. But they

1097
00:59:01,840 --> 00:59:04,080
Speaker 1: don't have any responsibility. They don't have teams who are

1098
00:59:04,080 --> 00:59:06,080
Speaker 1: dedicated to making sure people aren't starving. They don't have

1099
00:59:06,080 --> 00:59:08,439
Speaker 1: people who are dedicated to actually keeping the roads open

1100
00:59:08,440 --> 00:59:11,280
Speaker 1: in any way that isn't necessary for directly the trade

1101
00:59:11,280 --> 00:59:14,280
Speaker 1: that profits them. They don't do those things because they're

1102
00:59:14,320 --> 00:59:16,760
Speaker 1: not They're governing effectively, but they're not a government. And

1103
00:59:16,800 --> 00:59:20,040
Speaker 1: there's been a lot of talk about how Facebook is

1104
00:59:20,440 --> 00:59:23,640
Speaker 1: a is effectively like a nation, a digital nation of

1105
00:59:23,680 --> 00:59:26,040
Speaker 1: like three billion people, and Mark Zuckerberg has the power

1106
00:59:26,080 --> 00:59:28,400
Speaker 1: of a dictator. And one of the problems with that

1107
00:59:28,520 --> 00:59:31,480
Speaker 1: is that for all of their faults. Governments have a

1108
00:59:31,560 --> 00:59:35,320
Speaker 1: responsibility to do things for people that are like necessary

1109
00:59:35,360 --> 00:59:38,840
Speaker 1: to stop them, like to deal with like calamities and whatnot.

1110
00:59:39,240 --> 00:59:42,360
Speaker 1: Facebook has no such responsibility, and so when people were

1111
00:59:42,400 --> 00:59:44,760
Speaker 1: not paying attention to Sri Lanka to West Bengal to

1112
00:59:44,960 --> 00:59:48,640
Speaker 1: um Um to Myanmar, they didn't do anything. Um And

1113
00:59:48,680 --> 00:59:50,520
Speaker 1: as we know, like forty in the in a in

1114
00:59:50,560 --> 00:59:53,120
Speaker 1: a region where there are millions and millions of people,

1115
00:59:53,520 --> 00:59:57,800
Speaker 1: forty percent of the views were fake and authentic content,

1116
00:59:58,280 --> 01:00:01,480
Speaker 1: you know, like because don't give a ship what's spreading

1117
01:00:01,520 --> 01:00:03,800
Speaker 1: because they don't have to, because they don't have to

1118
01:00:03,800 --> 01:00:06,720
Speaker 1: deal with the consequences unless it pisses people off, as

1119
01:00:06,720 --> 01:00:09,560
Speaker 1: opposed to a government where it's like, well, yeah, we

1120
01:00:09,640 --> 01:00:11,840
Speaker 1: are made up of the people who live here and

1121
01:00:11,880 --> 01:00:14,440
Speaker 1: if things go badly enough, it can't not affect us.

1122
01:00:14,680 --> 01:00:16,840
Speaker 1: I'm not trying to be again, not like with TikTok,

1123
01:00:16,840 --> 01:00:19,400
Speaker 1: I'm not trying to praise the concept of governance. But

1124
01:00:19,480 --> 01:00:23,560
Speaker 1: it is better than what Facebook's doing right right, It's

1125
01:00:23,640 --> 01:00:27,040
Speaker 1: it's I think that that is like a very I'd

1126
01:00:27,080 --> 01:00:29,680
Speaker 1: never considered looking at it that way, but but viewing

1127
01:00:29,680 --> 01:00:33,760
Speaker 1: it as this kind of digital dictatorship that a colonial dictatorship.

1128
01:00:33,800 --> 01:00:38,960
Speaker 1: It's colonized people's information. Um, really like information streams, it's

1129
01:00:39,000 --> 01:00:42,280
Speaker 1: colonized the way people communicate, but it has no responsibility

1130
01:00:42,320 --> 01:00:44,720
Speaker 1: to them if they aren't white and wealthy. Well yeah,

1131
01:00:44,760 --> 01:00:47,840
Speaker 1: and and yeah, and marginalize people in the same ways

1132
01:00:47,840 --> 01:00:51,920
Speaker 1: that actual dictatorships do in terms of how much attention

1133
01:00:52,000 --> 01:00:55,120
Speaker 1: is being given, how are are people being hired to

1134
01:00:55,800 --> 01:00:58,520
Speaker 1: support and represent this area? And of course the answer

1135
01:00:58,760 --> 01:01:04,960
Speaker 1: is no. Course the result of that is extreme human

1136
01:01:05,440 --> 01:01:09,360
Speaker 1: consequence and and harm. And it's so and it's like,

1137
01:01:09,840 --> 01:01:12,360
Speaker 1: it's just so striking to me that it still feels

1138
01:01:12,400 --> 01:01:18,600
Speaker 1: like in terms of the laws that exist that control.

1139
01:01:18,680 --> 01:01:22,160
Speaker 1: I mean that that even attempt to address the amount

1140
01:01:22,200 --> 01:01:25,920
Speaker 1: of influence and control that a gigantic digital network like

1141
01:01:25,920 --> 01:01:31,120
Speaker 1: like Facebook has. Um, you know that Facebook. I mean

1142
01:01:31,320 --> 01:01:33,960
Speaker 1: unless people are yelling at them, and unless unless their

1143
01:01:34,000 --> 01:01:36,440
Speaker 1: bottom line is threatened, they're never going to respond to

1144
01:01:36,480 --> 01:01:40,400
Speaker 1: stuff like this. Like it. That's that's been made clear

1145
01:01:40,560 --> 01:01:44,280
Speaker 1: for decades at this point. It's great. I love it

1146
01:01:45,920 --> 01:01:49,080
Speaker 1: so well. I'm all worked out. Yeah. A great seal

1147
01:01:49,240 --> 01:01:52,720
Speaker 1: of the disinformation that goes throughout India on Facebook comes

1148
01:01:52,760 --> 01:01:56,200
Speaker 1: from the RSS, which is an Indian fascist organization closely

1149
01:01:56,240 --> 01:01:58,440
Speaker 1: tied to the b j P, which is the current

1150
01:01:58,520 --> 01:02:00,760
Speaker 1: ruling right wing party. And when I say fascist, I

1151
01:02:00,760 --> 01:02:02,640
Speaker 1: mean like some of the founders of the r S

1152
01:02:02,840 --> 01:02:06,120
Speaker 1: s we're actual like friends with Nazis and they were

1153
01:02:06,120 --> 01:02:09,200
Speaker 1: heavily influenced by that ship in like the thirties. Both

1154
01:02:09,440 --> 01:02:12,880
Speaker 1: organizations are profoundly anti Muslim and the r S s

1155
01:02:13,000 --> 01:02:16,200
Speaker 1: IS propaganda has been tied to numerous acts of violence.

1156
01:02:16,600 --> 01:02:19,600
Speaker 1: Facebook refuses to designate them a dangerous organization because of

1157
01:02:19,680 --> 01:02:22,600
Speaker 1: quote political sensitivities that might harm their ability to make

1158
01:02:22,640 --> 01:02:25,800
Speaker 1: money in India. Facebook is the best friend many far

1159
01:02:25,920 --> 01:02:28,680
Speaker 1: right and fascist political parties have ever had. Take the

1160
01:02:28,760 --> 01:02:33,160
Speaker 1: Polish Confederation Party. There your standard right wing extremists, anti immigrant,

1161
01:02:33,200 --> 01:02:36,440
Speaker 1: anti lockdown, anti vaccine, anti LGBT. The head of their

1162
01:02:36,440 --> 01:02:41,200
Speaker 1: social media team, Thomas garbage Check uh sorry Toms, told

1163
01:02:41,240 --> 01:02:44,400
Speaker 1: The Washington Post that Facebook's hate algorithm, in his words,

1164
01:02:44,600 --> 01:02:47,000
Speaker 1: had been a huge boon to their digital efforts. Like

1165
01:02:47,040 --> 01:02:49,640
Speaker 1: he calls it a hate Alglerman says this is great

1166
01:02:49,680 --> 01:02:52,600
Speaker 1: for us um expanding it. Like I think we're good

1167
01:02:52,600 --> 01:02:56,000
Speaker 1: with emotional messages and thus there ship spreads well on Facebook.

1168
01:02:56,520 --> 01:03:00,000
Speaker 1: Quote from The Washington post. In one April two nineteen

1169
01:03:00,080 --> 01:03:02,560
Speaker 1: document detailing a research trip to the European Union, a

1170
01:03:02,560 --> 01:03:05,840
Speaker 1: Facebook team reported feedback from European politicians that an algorithm

1171
01:03:05,920 --> 01:03:08,600
Speaker 1: changed the previous year built by Facebook at chief executive

1172
01:03:08,600 --> 01:03:11,640
Speaker 1: Mark Zuckerberg's a effort to foster more meaningful interactions on

1173
01:03:11,680 --> 01:03:15,520
Speaker 1: the platform had changed politics for the worst. This change,

1174
01:03:15,560 --> 01:03:17,960
Speaker 1: Mark claim, was meant to make interactions more meaningful, but

1175
01:03:18,000 --> 01:03:19,840
Speaker 1: it was really just a tweak to the algorithm that

1176
01:03:19,880 --> 01:03:22,640
Speaker 1: made comments that provoked anger and argument even more viral.

1177
01:03:23,320 --> 01:03:25,600
Speaker 1: Um and I'm gonna quote from the post again here.

1178
01:03:25,720 --> 01:03:27,800
Speaker 1: In two thousand and eighteen, Facebook made a big change

1179
01:03:27,800 --> 01:03:31,280
Speaker 1: to that formula to provote meaningful social interactions. These changes

1180
01:03:31,280 --> 01:03:33,120
Speaker 1: were built as a design to make the news feed

1181
01:03:33,120 --> 01:03:35,160
Speaker 1: more focused on posts from family and friends and less

1182
01:03:35,160 --> 01:03:38,120
Speaker 1: from brands, businesses, and the media. The process weighted the

1183
01:03:38,160 --> 01:03:40,960
Speaker 1: profit probability that a post would produce an interaction such

1184
01:03:40,960 --> 01:03:43,680
Speaker 1: as a like, emoji or comment, more heavily than other factors,

1185
01:03:44,000 --> 01:03:46,640
Speaker 1: but that appeared to backfire. Hogan, who this week took

1186
01:03:46,640 --> 01:03:49,160
Speaker 1: her campaign against her former employer to Europe, voiced a

1187
01:03:49,160 --> 01:03:52,720
Speaker 1: concern that Facebook's algorithm amplifies the extreme anger and hate

1188
01:03:52,760 --> 01:03:54,800
Speaker 1: is the easiest way to grow on Facebook, she told

1189
01:03:54,800 --> 01:03:57,640
Speaker 1: British lawmakers, many of whom have their jobs because of

1190
01:03:57,680 --> 01:03:59,920
Speaker 1: how easy it is to make people ship go by

1191
01:04:00,080 --> 01:04:06,120
Speaker 1: roll when it comes. I mean that that that's not

1192
01:04:06,160 --> 01:04:09,960
Speaker 1: true for yes, Um, what's again? We're we're focusing on

1193
01:04:09,960 --> 01:04:11,720
Speaker 1: Facebook here in part because I do think it's more

1194
01:04:11,760 --> 01:04:13,280
Speaker 1: severe in a lot of ways there, but also just

1195
01:04:13,320 --> 01:04:15,760
Speaker 1: because like they're the ones who had a big leak,

1196
01:04:15,800 --> 01:04:18,640
Speaker 1: and so we have this data. So we're not just saying, yeah,

1197
01:04:18,760 --> 01:04:21,240
Speaker 1: look at Facebook obviously hate spreading. They were saying no, no no,

1198
01:04:21,400 --> 01:04:24,800
Speaker 1: we have numbers. We have their numbers about how fucking

1199
01:04:24,880 --> 01:04:28,760
Speaker 1: bad the problem is. I guess that that is the difference. Yeah, yeah,

1200
01:04:29,080 --> 01:04:32,440
Speaker 1: and and yeah, we have evidence that the system is

1201
01:04:32,440 --> 01:04:34,480
Speaker 1: well aware of the Yeah. I would love to be

1202
01:04:34,520 --> 01:04:37,640
Speaker 1: talking about Twitter too. It's just maybe Twitter just never

1203
01:04:37,680 --> 01:04:40,920
Speaker 1: bothered to get those kind of numbers. Who knows. Um.

1204
01:04:40,920 --> 01:04:43,520
Speaker 1: This caused what experts describe as a social civil war

1205
01:04:43,600 --> 01:04:47,120
Speaker 1: in Poland. Like this change, one internal report concluded, we

1206
01:04:47,120 --> 01:04:49,640
Speaker 1: can choose to be idle and keep feeding users fast food,

1207
01:04:49,680 --> 01:04:51,560
Speaker 1: but that only works for so long. And you've already

1208
01:04:51,560 --> 01:04:53,240
Speaker 1: caught onto the fact that fast food is linked to

1209
01:04:53,280 --> 01:04:55,680
Speaker 1: obesity and therefore and it's short term value is not

1210
01:04:55,720 --> 01:04:59,000
Speaker 1: worth the long term cost. So he's being like, we're

1211
01:04:59,040 --> 01:05:03,880
Speaker 1: poisoning people and it's addictive, like you know McDonald's, but

1212
01:05:03,960 --> 01:05:06,040
Speaker 1: like people are going to give it up in the

1213
01:05:06,040 --> 01:05:09,760
Speaker 1: same way that McDonald started to suffer a couple of

1214
01:05:09,840 --> 01:05:12,440
Speaker 1: years back, because like, they don't they don't like the

1215
01:05:12,440 --> 01:05:15,400
Speaker 1: way this makes them feel. Actually, it's fun for a moment.

1216
01:05:15,440 --> 01:05:19,240
Speaker 1: We just got to get a Morgan Spurlock for Facebook. Baby,

1217
01:05:19,480 --> 01:05:22,640
Speaker 1: we just got to get where's the supersize me for faceboo?

1218
01:05:23,000 --> 01:05:28,400
Speaker 1: Entire society is the Morgan Spurlock for Facebook? January sixties more. Yeah,

1219
01:05:29,760 --> 01:05:31,080
Speaker 1: I was gonna say it was like I feel like

1220
01:05:31,160 --> 01:05:37,640
Speaker 1: it's I mean, whatever, not to say that McDonald's isn't

1221
01:05:37,840 --> 01:05:41,680
Speaker 1: a hell of a drug, but like this is not

1222
01:05:42,000 --> 01:05:45,800
Speaker 1: the same. I mean, I it's stronger because it's your

1223
01:05:46,560 --> 01:05:50,160
Speaker 1: fucking brain and self image and the view of yourself.

1224
01:05:50,200 --> 01:05:54,360
Speaker 1: And I feel like that is the most strong manipulation

1225
01:05:54,440 --> 01:05:57,480
Speaker 1: that any given system, person, whatever can have on you

1226
01:05:57,520 --> 01:06:00,360
Speaker 1: as controlling the way that you see yourself. It's not

1227
01:06:01,600 --> 01:06:05,959
Speaker 1: the same in terms of like involuntary baseness. I feel

1228
01:06:05,960 --> 01:06:08,960
Speaker 1: like it's something that you very much participate in. Yeah,

1229
01:06:09,440 --> 01:06:13,840
Speaker 1: um it's bad. Yeah, um it's good. I think it's good.

1230
01:06:14,160 --> 01:06:17,520
Speaker 1: That's what I think, Jamie. Facebook has been aggressive. You've

1231
01:06:17,560 --> 01:06:20,280
Speaker 1: called me today to see good, actually to read all

1232
01:06:20,320 --> 01:06:22,680
Speaker 1: this and then say so that's fine, let's never talk

1233
01:06:22,720 --> 01:06:26,640
Speaker 1: of it again. Um. Anyway, Facebook has been aggressive at

1234
01:06:26,640 --> 01:06:30,040
Speaker 1: rebutting the allegations that their product leads to polarization. Their

1235
01:06:30,040 --> 01:06:32,560
Speaker 1: spokeswoman brought up a study which she said shows that

1236
01:06:32,640 --> 01:06:36,160
Speaker 1: academic research doesn't support the idea that Facebook or social

1237
01:06:36,160 --> 01:06:39,440
Speaker 1: media more generally is the primary cause of polarization. To

1238
01:06:39,520 --> 01:06:42,440
Speaker 1: ignore for the moment that not the primary cause doesn't

1239
01:06:42,480 --> 01:06:45,520
Speaker 1: mean isn't a significant cause. And let's look at this study.

1240
01:06:45,800 --> 01:06:49,640
Speaker 1: The spokeswoman was referencing cross country trends and effective Polarization

1241
01:06:49,920 --> 01:06:54,280
Speaker 1: in August study from researchers at Stanford and Brown University.

1242
01:06:54,360 --> 01:06:57,000
Speaker 1: This study opens by noting it includes data for only

1243
01:06:57,040 --> 01:06:59,960
Speaker 1: twelve countries and that all but Britain and Germany exhibited

1244
01:07:00,000 --> 01:07:04,120
Speaker 1: a positive trend trend towards more polarization. So right off

1245
01:07:04,120 --> 01:07:06,720
Speaker 1: the bat, there's some things to question about this study. Um,

1246
01:07:06,760 --> 01:07:09,000
Speaker 1: which is number one, they're saying that, like, oh, Britain

1247
01:07:09,040 --> 01:07:11,320
Speaker 1: hasn't gotten more polarized, which is like have you have

1248
01:07:11,400 --> 01:07:15,080
Speaker 1: you been there? Have you talked to but I yeah,

1249
01:07:15,200 --> 01:07:18,040
Speaker 1: not the not the don't live there, but not what

1250
01:07:18,160 --> 01:07:21,480
Speaker 1: I've been hearing. And here's the thing, when you look

1251
01:07:21,520 --> 01:07:24,360
Speaker 1: at Facebook is basically using this citing this is like

1252
01:07:24,400 --> 01:07:26,920
Speaker 1: evidence that like, look, we're fine, social media is not.

1253
01:07:27,040 --> 01:07:29,439
Speaker 1: The study from this very credible study says that we're

1254
01:07:29,440 --> 01:07:33,560
Speaker 1: not the cause of polarization, so everything's good. The study

1255
01:07:33,800 --> 01:07:37,000
Speaker 1: doesn't quite back them up on this um right off

1256
01:07:37,040 --> 01:07:40,680
Speaker 1: the bat. One of the authors provides it like notes this,

1257
01:07:40,960 --> 01:07:42,280
Speaker 1: and this is from a right up by one of

1258
01:07:42,280 --> 01:07:43,880
Speaker 1: the authors on the study on a website called tech

1259
01:07:43,880 --> 01:07:46,040
Speaker 1: policy dot com where he's talking about the study and

1260
01:07:46,040 --> 01:07:49,040
Speaker 1: what it says. A flat or de or declining trend

1261
01:07:49,160 --> 01:07:51,160
Speaker 1: over the forty years of our sample does not rule

1262
01:07:51,160 --> 01:07:53,920
Speaker 1: out the possibility that countries have seen rising polarization in

1263
01:07:53,960 --> 01:07:56,840
Speaker 1: the most recent years. Britain, for example, shows a slight

1264
01:07:56,920 --> 01:07:59,760
Speaker 1: overall decline, but a clear increasing trend post two thousand

1265
01:07:59,760 --> 01:08:02,640
Speaker 1: in Brexit. So he's saying that, like, we don't have

1266
01:08:02,720 --> 01:08:05,720
Speaker 1: as much data from like more recent polarization, and that

1267
01:08:05,800 --> 01:08:07,680
Speaker 1: may be a reason why this study is less accurate,

1268
01:08:07,720 --> 01:08:10,040
Speaker 1: why some of our our statements are not do not

1269
01:08:10,160 --> 01:08:12,920
Speaker 1: conform with like what people have observed. He goes on

1270
01:08:13,000 --> 01:08:15,560
Speaker 1: to note the data do not provide much support for

1271
01:08:15,560 --> 01:08:18,360
Speaker 1: the hypothesis that digital technology is the central driver of

1272
01:08:18,400 --> 01:08:21,640
Speaker 1: effective polarization. The Internet has diffused widely in all the

1273
01:08:21,680 --> 01:08:24,240
Speaker 1: countries we looked at, and under simple stories where this

1274
01:08:24,320 --> 01:08:26,559
Speaker 1: is the key driver, we would have expected polarization to

1275
01:08:26,560 --> 01:08:29,559
Speaker 1: have risen everywhere as well. In our data, neither diffusion

1276
01:08:29,800 --> 01:08:32,839
Speaker 1: of internet nor penetration of digital news are significantly correlated

1277
01:08:32,880 --> 01:08:36,679
Speaker 1: with increasing polarization. Similarly, we found little association with changes

1278
01:08:36,720 --> 01:08:40,000
Speaker 1: in inequality or trade. One explanatory factor that looks more

1279
01:08:40,040 --> 01:08:42,800
Speaker 1: promising is increasing racial diversity. The non white share of

1280
01:08:42,800 --> 01:08:45,040
Speaker 1: the population has increased faster in the US than in

1281
01:08:45,080 --> 01:08:47,559
Speaker 1: almost any other country, and our sample in other countries

1282
01:08:47,600 --> 01:08:49,679
Speaker 1: like New Zealand and Canada, where it has risen sharply

1283
01:08:49,680 --> 01:08:53,559
Speaker 1: have seen that rising polarization as well. So I have

1284
01:08:53,640 --> 01:08:56,200
Speaker 1: some significant arguments with him here, including the fact that,

1285
01:08:56,600 --> 01:09:00,439
Speaker 1: as he notes here, his study only looks at Western nations.

1286
01:09:00,439 --> 01:09:02,840
Speaker 1: With the exception of Japan, all of the nations in

1287
01:09:02,840 --> 01:09:06,160
Speaker 1: the study are European or the United States. In Canada,

1288
01:09:06,640 --> 01:09:08,840
Speaker 1: um and so they have all have had prior to

1289
01:09:08,880 --> 01:09:11,960
Speaker 1: two thousand, higher penetrations of the Internet and non Internet

1290
01:09:11,960 --> 01:09:14,280
Speaker 1: mass media. Like outside of this, if you're trying to

1291
01:09:14,320 --> 01:09:17,240
Speaker 1: determine the impact of social media elements of what social

1292
01:09:17,280 --> 01:09:19,320
Speaker 1: media has done, we're in present in places like Fox

1293
01:09:19,360 --> 01:09:23,200
Speaker 1: News in the United States years before Facebook ever existed. Um,

1294
01:09:23,240 --> 01:09:25,080
Speaker 1: And that was not the case in places like me

1295
01:09:25,160 --> 01:09:26,840
Speaker 1: and mar in India, which are not a part of

1296
01:09:26,840 --> 01:09:29,600
Speaker 1: this study. So right off the bat, it's problematic to

1297
01:09:29,640 --> 01:09:32,440
Speaker 1: try and study the impact of social media on polarization

1298
01:09:32,800 --> 01:09:35,840
Speaker 1: only in countries that already had robust mass media before

1299
01:09:35,880 --> 01:09:37,560
Speaker 1: social media came into the effect. Which is not to

1300
01:09:37,600 --> 01:09:40,280
Speaker 1: say that I agree with their conclusion, because I think

1301
01:09:40,280 --> 01:09:43,160
Speaker 1: there's other flaws with this study, but it one of

1302
01:09:43,160 --> 01:09:45,320
Speaker 1: the flaws is just that like hundreds of millions of

1303
01:09:45,360 --> 01:09:47,920
Speaker 1: their users exist in countries where they did not The

1304
01:09:47,960 --> 01:09:50,360
Speaker 1: study was not done, um, where they were not looking

1305
01:09:50,479 --> 01:09:52,519
Speaker 1: at these places, which is a flaw which is just

1306
01:09:52,640 --> 01:09:56,240
Speaker 1: like dependent on yeah, and that's dependent on most readers,

1307
01:09:56,360 --> 01:09:59,120
Speaker 1: just conflating, you know, in North America and Europe with

1308
01:09:59,200 --> 01:10:00,960
Speaker 1: the center of the fucking and and again I have

1309
01:10:01,080 --> 01:10:03,759
Speaker 1: issues about like, Okay, well you're saying that racial diversity

1310
01:10:03,800 --> 01:10:05,639
Speaker 1: is more of a thing. But like, where is the

1311
01:10:05,800 --> 01:10:09,200
Speaker 1: where is the propaganda, Where is the hate speech about

1312
01:10:09,240 --> 01:10:12,639
Speaker 1: racial diversity spreading? Is it spreading on social media? Like, yes,

1313
01:10:12,680 --> 01:10:14,960
Speaker 1: it is. I can say that as an expert. Um.

1314
01:10:15,000 --> 01:10:17,720
Speaker 1: It's also just like again, not that this study is

1315
01:10:17,760 --> 01:10:21,080
Speaker 1: even bad or not useful. It is one study, um.

1316
01:10:21,240 --> 01:10:24,879
Speaker 1: And again we have internal Facebook studies that make claims

1317
01:10:24,920 --> 01:10:28,080
Speaker 1: that I would say throw some of this into question.

1318
01:10:28,560 --> 01:10:30,920
Speaker 1: But again, this is just how a corporation is going

1319
01:10:30,960 --> 01:10:32,720
Speaker 1: to react. They're going to find a study that they

1320
01:10:32,720 --> 01:10:36,639
Speaker 1: can simplify in such a way that they claim that

1321
01:10:36,640 --> 01:10:39,320
Speaker 1: that they can claim there's not a problem because nobody,

1322
01:10:40,479 --> 01:10:41,880
Speaker 1: none of the people who they're going to be arguing

1323
01:10:41,880 --> 01:10:43,839
Speaker 1: with on Capitol Hill and precious to you, the journalists

1324
01:10:43,840 --> 01:10:45,720
Speaker 1: are going to actually drill into this and then talk

1325
01:10:45,760 --> 01:10:47,439
Speaker 1: to other experts. You're gonna reach out to members of

1326
01:10:47,479 --> 01:10:50,040
Speaker 1: that study and be like, how fair is this phrasing?

1327
01:10:50,040 --> 01:10:52,040
Speaker 1: How does that how does it deal with this information

1328
01:10:52,040 --> 01:10:54,160
Speaker 1: and this information As we saw earlier with the last study,

1329
01:10:54,160 --> 01:10:56,200
Speaker 1: when we people reached when the Wall Street Journal, to

1330
01:10:56,240 --> 01:10:58,639
Speaker 1: their credit, reached out to that scientist, he was like, well,

1331
01:10:58,680 --> 01:11:00,559
Speaker 1: actually they have better data than me. And I'd love

1332
01:11:00,640 --> 01:11:04,799
Speaker 1: to see it because maybe that'll change our conclusions. Um anyway,

1333
01:11:04,960 --> 01:11:08,759
Speaker 1: yeah uh. Mark Zuckerberg has been consistent in his argument

1334
01:11:08,760 --> 01:11:11,599
Speaker 1: that deliberately pushing divisive and violent content would be bad

1335
01:11:11,600 --> 01:11:14,400
Speaker 1: for Facebook. Quote. We make money from ads, and advertisers

1336
01:11:14,439 --> 01:11:16,360
Speaker 1: consistently tell us they don't want their ads next to

1337
01:11:16,400 --> 01:11:19,439
Speaker 1: harmful or angry content. While I was writing this article,

1338
01:11:19,479 --> 01:11:21,920
Speaker 1: I browsed over to one of my test Facebook accounts.

1339
01:11:22,080 --> 01:11:23,800
Speaker 1: The third ad on my feed was for a device

1340
01:11:23,840 --> 01:11:27,759
Speaker 1: to illegally turn a glock handgun into a fully automatic weapon. Um,

1341
01:11:28,360 --> 01:11:32,080
Speaker 1: just just his heads up. Yeah, one of my I

1342
01:11:32,080 --> 01:11:34,280
Speaker 1: have a couple of test feeds and it was like, hey,

1343
01:11:34,280 --> 01:11:37,280
Speaker 1: this button will turn your glock automatic, which is so

1344
01:11:37,320 --> 01:11:40,080
Speaker 1: many felonies. Jamie. If you even have that thing in

1345
01:11:40,120 --> 01:11:42,439
Speaker 1: a glock in your home, the FBI can put you

1346
01:11:42,479 --> 01:11:45,719
Speaker 1: away forever. I have to laugh. I have to laugh

1347
01:11:45,760 --> 01:11:48,400
Speaker 1: because that is really really scary. But yeah, it is

1348
01:11:48,439 --> 01:11:50,840
Speaker 1: like Mark being like, look, oh, no advertiser wants this

1349
01:11:50,880 --> 01:11:53,800
Speaker 1: to be a violent place by a machine gun on Facebook,

1350
01:11:53,840 --> 01:11:56,360
Speaker 1: you know, next to ads that are like t shirts

1351
01:11:56,360 --> 01:11:59,960
Speaker 1: about telling liberals and stuff, Like a machine gun ad

1352
01:12:00,000 --> 01:12:04,120
Speaker 1: advertiser maybe would be one that wouldn't take issue with that.

1353
01:12:04,720 --> 01:12:08,680
Speaker 1: Holy fucking hang the media shirts advertised to me on

1354
01:12:08,720 --> 01:12:12,439
Speaker 1: Facebook like my god go to go like fuck you

1355
01:12:12,640 --> 01:12:15,680
Speaker 1: Mark Um, I used to the last Well, when I

1356
01:12:15,760 --> 01:12:17,800
Speaker 1: quit Facebook a couple of years ago, I was I

1357
01:12:18,000 --> 01:12:21,920
Speaker 1: was getting Normi advertisements. I was getting good for you,

1358
01:12:22,080 --> 01:12:25,439
Speaker 1: those really scary ones that says, like those custom T

1359
01:12:25,520 --> 01:12:29,760
Speaker 1: shirts that say it's a Jamie Loftus thing, you wouldn't understand,

1360
01:12:31,400 --> 01:12:34,360
Speaker 1: and you wouldn't I would not know. And you know,

1361
01:12:34,840 --> 01:12:37,360
Speaker 1: the only time Facebook I can think of recently actually

1362
01:12:37,400 --> 01:12:40,360
Speaker 1: anticipated something I wanted is they keep showing me on

1363
01:12:40,400 --> 01:12:43,600
Speaker 1: all of the accounts that I've I've used videos of

1364
01:12:43,640 --> 01:12:46,759
Speaker 1: hydraulic presses crushing things, and I do love those videos.

1365
01:12:47,120 --> 01:12:50,519
Speaker 1: Those those are those are pretty pretty fun. And that's

1366
01:12:50,560 --> 01:12:55,000
Speaker 1: the meaningful social interactions that Mr Mark Zuckerberg was talking about,

1367
01:12:55,200 --> 01:12:58,160
Speaker 1: was the hydraulic press videos, and those are very comforted.

1368
01:12:58,240 --> 01:13:01,559
Speaker 1: On the good old Internet, which also wasn't all that great,

1369
01:13:01,600 --> 01:13:03,920
Speaker 1: but on the old Internet, which was it was a

1370
01:13:04,000 --> 01:13:05,439
Speaker 1: lot more fun. Though there would have just been a

1371
01:13:05,439 --> 01:13:07,679
Speaker 1: whole website that was just like, here's all the videos

1372
01:13:07,720 --> 01:13:11,600
Speaker 1: of hydraulic presses crushing things. Come watch this ship. There

1373
01:13:11,600 --> 01:13:14,639
Speaker 1: wouldn't have been any algorithm necessary. You could just scroll

1374
01:13:14,720 --> 01:13:20,719
Speaker 1: through videos. There's no friend function, it's just hydraulic press ship. Yeah,

1375
01:13:20,760 --> 01:13:24,920
Speaker 1: that's all I need, baby, I need. So back to

1376
01:13:24,960 --> 01:13:28,200
Speaker 1: the point, it is undeniable that any service on the

1377
01:13:28,240 --> 01:13:30,559
Speaker 1: scale of Facebook, again like three billion users, it's going

1378
01:13:30,600 --> 01:13:32,600
Speaker 1: to face some tough choices when it comes to the

1379
01:13:32,600 --> 01:13:35,479
Speaker 1: problem of regulating the speech of political movements and thinkers.

1380
01:13:35,800 --> 01:13:37,960
Speaker 1: As one employee wrote in an internal message, I am

1381
01:13:38,000 --> 01:13:40,840
Speaker 1: not comfortable making judgments about some parties being less good

1382
01:13:40,840 --> 01:13:43,360
Speaker 1: for society and less worthy of distribution based on where

1383
01:13:43,360 --> 01:13:47,280
Speaker 1: they fall in the ideological spectrum. That's true. Um, This

1384
01:13:47,360 --> 01:13:49,960
Speaker 1: is again part of the problem of not regulating them

1385
01:13:50,000 --> 01:13:53,040
Speaker 1: like a media company, like a newspaper or something. Um,

1386
01:13:53,120 --> 01:13:56,400
Speaker 1: Because by not making any choices, they're making an editorial choice,

1387
01:13:56,400 --> 01:13:59,800
Speaker 1: which is to allow this stuff to spread. Presumably. Actually,

1388
01:14:00,080 --> 01:14:01,760
Speaker 1: if you were actually being held to some kind of

1389
01:14:01,840 --> 01:14:04,519
Speaker 1: legal standard that again most of our media isn't anymore,

1390
01:14:04,720 --> 01:14:07,040
Speaker 1: you would at least have to be like, well, let's

1391
01:14:07,080 --> 01:14:10,400
Speaker 1: evaluate the truthfulness of some of these basic statements before

1392
01:14:10,439 --> 01:14:12,439
Speaker 1: president and and I would say that's where the judgment

1393
01:14:12,439 --> 01:14:15,280
Speaker 1: should come in on. But that's expensive. If Facebook is saying,

1394
01:14:15,520 --> 01:14:17,559
Speaker 1: we won't judge based on politics, but we will will

1395
01:14:17,640 --> 01:14:19,680
Speaker 1: We will judge based on whether or not something is

1396
01:14:19,760 --> 01:14:23,479
Speaker 1: counter factual that I think is morally defensible. But that's

1397
01:14:23,479 --> 01:14:26,160
Speaker 1: expensive as shit and they're never going to do that.

1398
01:14:26,439 --> 01:14:30,800
Speaker 1: Look spent more. Moral decisions are famously not cheap, and

1399
01:14:30,880 --> 01:14:33,120
Speaker 1: that is a lot of the reason why people do

1400
01:14:33,200 --> 01:14:36,439
Speaker 1: not do them. Yeah, um, it is true. That is

1401
01:14:36,479 --> 01:14:41,400
Speaker 1: not a profitable venture. Yeah, it's no, of course not um.

1402
01:14:41,439 --> 01:14:44,240
Speaker 1: And the other thing that's true is that Facebook already

1403
01:14:44,280 --> 01:14:46,759
Speaker 1: makes a lot of decisions about which politicians and parties

1404
01:14:46,800 --> 01:14:49,360
Speaker 1: are worthy of speech, and they make that decision based

1405
01:14:49,400 --> 01:14:51,439
Speaker 1: mostly on whether or not said public figures get a

1406
01:14:51,439 --> 01:14:55,080
Speaker 1: lot of engagement. Midway through last year, they deleted and

1407
01:14:55,640 --> 01:14:58,720
Speaker 1: like all of the different anarchist media groups that had

1408
01:14:58,720 --> 01:15:00,679
Speaker 1: in a lot of anti fascist group that had accounts

1409
01:15:00,680 --> 01:15:02,759
Speaker 1: on Facebook. Just across the board, they deleted Let crime

1410
01:15:02,840 --> 01:15:06,920
Speaker 1: Think and uh um they kicked off. It's going down

1411
01:15:07,080 --> 01:15:10,840
Speaker 1: like a rapper, I know, soul like yeah they I mean,

1412
01:15:11,040 --> 01:15:13,519
Speaker 1: I mean, it's nobody ever complains when best ship happens

1413
01:15:13,560 --> 01:15:16,040
Speaker 1: to anarchists except for anarchists. But yeah, they knuked a

1414
01:15:16,080 --> 01:15:19,599
Speaker 1: bunch of anarchist content. Um just kind of blankets saying

1415
01:15:19,600 --> 01:15:21,280
Speaker 1: it was dangerous. And I think it was because they

1416
01:15:21,320 --> 01:15:23,080
Speaker 1: just knuked the Proud Boys and they had to be

1417
01:15:23,120 --> 01:15:25,760
Speaker 1: shown to be fair. Um. But it has now come

1418
01:15:25,760 --> 01:15:28,320
Speaker 1: out that they have a whole program called x check

1419
01:15:28,560 --> 01:15:31,960
Speaker 1: or cross check, which is where they decide which political

1420
01:15:32,000 --> 01:15:37,759
Speaker 1: figures get to spread violent and false content without getting banned. Um,

1421
01:15:37,800 --> 01:15:40,960
Speaker 1: they had a yeah, based on engagement. They've claimed for

1422
01:15:41,040 --> 01:15:43,840
Speaker 1: years that everybody's accountable to site rules. But again the

1423
01:15:43,840 --> 01:15:47,200
Speaker 1: Facebook papers has revealed that, like, that's explicitly a lie

1424
01:15:47,240 --> 01:15:49,320
Speaker 1: and it's a life Facebook has told other people at

1425
01:15:49,400 --> 01:15:51,479
Speaker 1: high levels of Facebook. And I'm gonna quote from the

1426
01:15:51,479 --> 01:15:53,880
Speaker 1: Wall Street Journal here. The program, known as cross Checker

1427
01:15:54,040 --> 01:15:56,719
Speaker 1: x check, was initially intended as a quality control measure

1428
01:15:56,920 --> 01:16:00,520
Speaker 1: for actions taken against high profile accounts, including celebrities, politicians,

1429
01:16:00,560 --> 01:16:03,600
Speaker 1: and journalists. Today, at shields millions of v I P

1430
01:16:03,840 --> 01:16:07,040
Speaker 1: users from the company's normal enforcement process. The documents show

1431
01:16:07,320 --> 01:16:10,360
Speaker 1: some users are white listed, rendered immune from enforcement actions,

1432
01:16:10,360 --> 01:16:13,160
Speaker 1: while others are allowed to post rule violating material pending

1433
01:16:13,160 --> 01:16:16,280
Speaker 1: Facebook employee reviews that often never come. At times, the

1434
01:16:16,320 --> 01:16:19,320
Speaker 1: documents show ex check has protected public figures whose posts

1435
01:16:19,320 --> 01:16:22,719
Speaker 1: contain harassment or incitement to violence, violations that would typically

1436
01:16:22,760 --> 01:16:25,799
Speaker 1: lead to sanctions for regular users. In two thousand nineteen,

1437
01:16:25,840 --> 01:16:28,559
Speaker 1: it allowed international soccer star in neymar to show nude

1438
01:16:28,560 --> 01:16:30,479
Speaker 1: photos of a woman who had accused him of rape

1439
01:16:30,520 --> 01:16:32,519
Speaker 1: to tens of millions of his fans before the content

1440
01:16:32,600 --> 01:16:36,280
Speaker 1: was removed by Facebook. White Listed accounts shared inflammatory claims

1441
01:16:36,280 --> 01:16:39,479
Speaker 1: that Facebook's fact checkers dimmed deemed false, including that vaccines

1442
01:16:39,520 --> 01:16:42,439
Speaker 1: are deadly, that Hillary Clinton had covered up pedophile wings,

1443
01:16:42,640 --> 01:16:45,040
Speaker 1: and that then president Donald Trump had called all refugees

1444
01:16:45,040 --> 01:16:48,880
Speaker 1: seeking asylum animals. According to the documents, A two thousand

1445
01:16:48,960 --> 01:16:52,320
Speaker 1: nineteen review of Facebook's white listing procedures marked Attorney Client

1446
01:16:52,400 --> 01:16:55,920
Speaker 1: Privileged found favoritism to those users to be both widespread

1447
01:16:55,960 --> 01:16:59,360
Speaker 1: and not publicly defensible. We are not actually doing what

1448
01:16:59,400 --> 01:17:01,960
Speaker 1: we say we do publicly, said the confidential review. It

1449
01:17:02,040 --> 01:17:05,280
Speaker 1: called the company's actions a breach of trust, and added,

1450
01:17:05,360 --> 01:17:07,800
Speaker 1: unlike the rest of our community, these people violate our

1451
01:17:07,840 --> 01:17:09,960
Speaker 1: standards without any consequence, and they like to like their

1452
01:17:09,960 --> 01:17:12,000
Speaker 1: board members about whether or not this. They didn't lie

1453
01:17:12,000 --> 01:17:13,679
Speaker 1: about whether or not it was a thing. They said

1454
01:17:13,720 --> 01:17:15,360
Speaker 1: it was very small and just really I think the

1455
01:17:15,400 --> 01:17:17,360
Speaker 1: initial claim was, like we have to have something like

1456
01:17:17,400 --> 01:17:19,120
Speaker 1: this in place where people like President Trump. But it's

1457
01:17:19,160 --> 01:17:22,280
Speaker 1: a tiny number of people, and it's because they occupy

1458
01:17:22,360 --> 01:17:26,080
Speaker 1: some political position where we can't just as easily, you know,

1459
01:17:26,160 --> 01:17:29,200
Speaker 1: delete their account because it creates other problems because they're

1460
01:17:29,240 --> 01:17:30,960
Speaker 1: not as fringe as they need to be for this

1461
01:17:31,720 --> 01:17:36,240
Speaker 1: conduct to be. That was their justification, one sect, Jamie.

1462
01:17:36,320 --> 01:17:39,160
Speaker 1: That was their justification on the level of you can

1463
01:17:39,240 --> 01:17:43,360
Speaker 1: be unethical and still be illegal. And well, here's the

1464
01:17:43,400 --> 01:17:46,439
Speaker 1: thing they told their board. They only did this for

1465
01:17:46,479 --> 01:17:48,160
Speaker 1: a small number of users. You want to guess how

1466
01:17:48,160 --> 01:17:50,439
Speaker 1: small what that small number was. Oh, I love when

1467
01:17:50,479 --> 01:17:53,120
Speaker 1: Facebook say there's a small number. What is this? What

1468
01:17:53,280 --> 01:18:00,479
Speaker 1: is this? Eight million? That's so many? Yeah, oh dear,

1469
01:18:01,560 --> 01:18:04,960
Speaker 1: it's very funny. It's very funny. It's all good. I

1470
01:18:07,600 --> 01:18:11,439
Speaker 1: that is so. I mean, yeah, that they're just they're

1471
01:18:11,520 --> 01:18:15,160
Speaker 1: just can I say something controversial? Please? I don't like

1472
01:18:15,240 --> 01:18:18,519
Speaker 1: this company one bit. You don't, well, I feel like

1473
01:18:18,600 --> 01:18:22,080
Speaker 1: that's going a bit uh, that's going a bit far. Sorry,

1474
01:18:22,560 --> 01:18:26,800
Speaker 1: And I'm famously you know, I don't like making harsh

1475
01:18:26,880 --> 01:18:29,240
Speaker 1: judgments on others, but I'm starting to think that they

1476
01:18:29,320 --> 01:18:32,599
Speaker 1: might be doing some bad stuff over there. M hmmm, yeah,

1477
01:18:32,760 --> 01:18:35,920
Speaker 1: I would. You know, I don't like these people. I

1478
01:18:35,960 --> 01:18:38,240
Speaker 1: don't like these people at all. You know what, I

1479
01:18:38,280 --> 01:18:43,479
Speaker 1: do like Jamie ending podcast episodes. M m oh, I

1480
01:18:43,520 --> 01:18:46,080
Speaker 1: actually do like that. Yeah, that's the thing I'm best at.

1481
01:18:46,479 --> 01:18:50,880
Speaker 1: Do you want to plug your plug doubles passion? Um? Yeah,

1482
01:18:50,880 --> 01:18:55,799
Speaker 1: sure you can find. I'm gonna just open by plugging

1483
01:18:55,800 --> 01:19:01,280
Speaker 1: my Instagram account, a famously healthy platform that I'm addicted to,

1484
01:19:01,439 --> 01:19:03,720
Speaker 1: and I don't really have any concerns about it. I

1485
01:19:03,720 --> 01:19:06,679
Speaker 1: don't really think it's affecting my mental health at all. Um.

1486
01:19:06,720 --> 01:19:10,040
Speaker 1: So I'm over there and that's Jamie christ Superstar. I'm

1487
01:19:10,040 --> 01:19:12,559
Speaker 1: also on Twitter, which Robert can't stop saying is the

1488
01:19:12,560 --> 01:19:17,280
Speaker 1: healthiest of the platforms. Of all of the people who

1489
01:19:17,320 --> 01:19:22,120
Speaker 1: are drunk driving through intersections filled with children, Twitter has

1490
01:19:22,160 --> 01:19:25,559
Speaker 1: the least amount of human blood and gore underneath the

1491
01:19:25,600 --> 01:19:29,080
Speaker 1: grill of the car, like the Robert saying for all

1492
01:19:29,120 --> 01:19:32,080
Speaker 1: you Backstreet Boys heads, he's saying that Twitter is the

1493
01:19:32,160 --> 01:19:35,360
Speaker 1: Kevin Richardson of social media. I'm there as well. I'm

1494
01:19:35,439 --> 01:19:39,400
Speaker 1: saying the the drunk driving Twitter car made it a

1495
01:19:39,439 --> 01:19:42,920
Speaker 1: full fifteen feet further than the Facebook car before the

1496
01:19:42,920 --> 01:19:45,479
Speaker 1: sheer amount of blood being churned up into the engine,

1497
01:19:45,520 --> 01:19:48,200
Speaker 1: flooded the engine air intakes. But at the end of

1498
01:19:48,200 --> 01:19:51,160
Speaker 1: the day, we're all fucked. I'm on Twitter as well,

1499
01:19:51,400 --> 01:19:57,000
Speaker 1: at Jamie. Listen to my podcast, Yeah, you know. You

1500
01:19:57,000 --> 01:20:00,559
Speaker 1: can listen to my podcast, the Beachdel Cast. You listened

1501
01:20:00,600 --> 01:20:03,400
Speaker 1: to act Cast, that's about the Kathy Comics. You can

1502
01:20:03,439 --> 01:20:05,519
Speaker 1: listen to My year in Mensa. You can listen to

1503
01:20:05,560 --> 01:20:08,040
Speaker 1: the lead of podcast. You can listen to nothing you

1504
01:20:08,040 --> 01:20:11,599
Speaker 1: know what never led to a genocide in any country

1505
01:20:11,640 --> 01:20:14,880
Speaker 1: as far as I'm aware of, Jamie. Uh, the Kathy Comics.

1506
01:20:14,960 --> 01:20:17,400
Speaker 1: We'll see then you haven't listened to the whole series, really,

1507
01:20:17,560 --> 01:20:20,040
Speaker 1: is it? Oh? You know what? You know? Yeah, that's

1508
01:20:20,120 --> 01:20:23,000
Speaker 1: that's why the last the last episode is your Your

1509
01:20:23,040 --> 01:20:28,400
Speaker 1: Life Report from Sarah Jevo in episode eleven. Yeah, Irving

1510
01:20:28,479 --> 01:20:31,280
Speaker 1: really his his politics were not good. Yeah he was.

1511
01:20:31,400 --> 01:20:35,919
Speaker 1: He was. He was like weirdly into the Serbian nationalism. Um.

1512
01:20:36,520 --> 01:20:39,599
Speaker 1: Irving is like for the Kathy Comics, he's like, Okay,

1513
01:20:39,640 --> 01:20:43,760
Speaker 1: I'm about to make a wild parallel. But Irving is

1514
01:20:43,800 --> 01:20:48,000
Speaker 1: like the Barefoot Contessa's husband and that he looks so innocent,

1515
01:20:48,120 --> 01:20:50,679
Speaker 1: but then when you google him, you're like, wait, a second.

1516
01:20:51,040 --> 01:20:53,600
Speaker 1: This man is running on dark money. This guy is

1517
01:20:53,640 --> 01:20:55,559
Speaker 1: like was on Wall Street in the eighties. This is

1518
01:20:55,600 --> 01:21:00,000
Speaker 1: a bad man. That's he's basically like Jeffrey, the Barefoot

1519
01:21:00,000 --> 01:21:03,320
Speaker 1: Contessa's husband. The Barefoot Contessa is run on dark money.

1520
01:21:03,640 --> 01:21:06,000
Speaker 1: I know people don't like to hear it. They love her,

1521
01:21:06,600 --> 01:21:09,160
Speaker 1: but it's just true. It's objectively true, and that's what

1522
01:21:09,200 --> 01:21:11,559
Speaker 1: I would like to say at the end of the episode.

1523
01:21:11,760 --> 01:21:14,439
Speaker 1: I've never heard of the Barefoot Contessa and I don't

1524
01:21:14,479 --> 01:21:17,800
Speaker 1: know what you're talking about. I have not even one

1525
01:21:17,840 --> 01:21:20,760
Speaker 1: percent surprise, but that's okay. But you know what I

1526
01:21:20,800 --> 01:21:24,400
Speaker 1: do know about what I know about podcasts, and this

1527
01:21:24,439 --> 01:21:26,240
Speaker 1: one is done great ending