1
00:00:01,120 --> 00:00:05,200
Speaker 1: Friends and neighbors, fellow conspiracy realist, welcome back to stuff

2
00:00:05,240 --> 00:00:07,920
Speaker 1: they don't want you to know. We have a classic

3
00:00:08,039 --> 00:00:15,000
Speaker 1: episode for you today that has only aged better with time.

4
00:00:15,680 --> 00:00:19,960
Speaker 1: You may recall back in twenty nineteen, we started talking

5
00:00:20,120 --> 00:00:24,760
Speaker 1: about the idea of a deep fake. What is a

6
00:00:24,840 --> 00:00:25,400
Speaker 1: deep fake?

7
00:00:26,079 --> 00:00:30,040
Speaker 2: Oh, it's a generated version of something that looks real.

8
00:00:30,680 --> 00:00:35,560
Speaker 2: And it's strange how even since twenty nineteen, the way

9
00:00:35,640 --> 00:00:40,280
Speaker 2: we talk about this concept and even the vocabulary that

10
00:00:40,320 --> 00:00:43,479
Speaker 2: we use surrounding it has a it's changed completely. It's

11
00:00:43,479 --> 00:00:46,279
Speaker 2: been altered by artificial intelligence and the way we talk

12
00:00:46,320 --> 00:00:50,479
Speaker 2: about that thing. But it's a Yeah, it's just a

13
00:00:50,520 --> 00:00:53,080
Speaker 2: fake thing, right, It's a it looks like Matt, but

14
00:00:53,200 --> 00:00:54,000
Speaker 2: it's not Matt.

15
00:00:55,560 --> 00:01:01,400
Speaker 1: Yeah. Photographs have been faked since the invention of photography,

16
00:01:01,480 --> 00:01:06,360
Speaker 1: which we talked about there pretty often. And I just again,

17
00:01:06,720 --> 00:01:09,480
Speaker 1: you know, we get a lot of messages and people

18
00:01:09,600 --> 00:01:13,480
Speaker 1: ask me, hey, Ben, are you guys ever gonna successfully

19
00:01:13,520 --> 00:01:18,000
Speaker 1: predict a good or a positive thing? We hope so.

20
00:01:18,720 --> 00:01:21,440
Speaker 1: But we were right about a lot of evil things,

21
00:01:21,560 --> 00:01:24,720
Speaker 1: and deep fakes are one of them. It was really

22
00:01:24,760 --> 00:01:29,160
Speaker 1: interesting to listen back to this episode together and to

23
00:01:29,560 --> 00:01:34,200
Speaker 1: realize just despite your point about nomenclature to realize just

24
00:01:34,319 --> 00:01:39,120
Speaker 1: how disturbingly accurate a lot of this was.

25
00:01:40,360 --> 00:01:43,560
Speaker 2: Yeah, so let's get into it. We assure you these

26
00:01:43,600 --> 00:01:47,240
Speaker 2: are the real US is having this conversation.

27
00:01:49,000 --> 00:01:53,480
Speaker 3: From UFOs to psychic powers and government conspiracies. History is

28
00:01:53,560 --> 00:01:57,880
Speaker 3: riddled with unexplained events. You can turn back now or

29
00:01:57,960 --> 00:02:00,960
Speaker 3: learn this stuff they don't want you to know. A

30
00:02:01,000 --> 00:02:04,080
Speaker 3: production of iHeart Radios How Stuff Works.

31
00:02:13,040 --> 00:02:16,440
Speaker 2: Welcome back to the show. My name's Matt, my name

32
00:02:16,480 --> 00:02:17,119
Speaker 2: is no.

33
00:02:17,480 --> 00:02:19,560
Speaker 1: They call me Ben. We were joined as always with

34
00:02:19,600 --> 00:02:23,240
Speaker 1: our super producer Paul Michigan, trol Deck and most importantly,

35
00:02:23,360 --> 00:02:26,400
Speaker 1: you are you. You are here and that makes this

36
00:02:26,840 --> 00:02:29,639
Speaker 1: stuff they don't want you to know. As we are

37
00:02:29,680 --> 00:02:33,040
Speaker 1: barreling toward the end of the year, if that is

38
00:02:33,160 --> 00:02:34,840
Speaker 1: you believe in the current calendar?

39
00:02:36,240 --> 00:02:38,639
Speaker 2: Yes, right, I mean.

40
00:02:39,040 --> 00:02:42,200
Speaker 1: Is this the real life? Is this just fantasy?

41
00:02:44,320 --> 00:02:45,560
Speaker 4: We're gonna do three part harmony.

42
00:02:45,680 --> 00:02:48,720
Speaker 1: No, no, okay, here, let's start this way. What's the

43
00:02:49,080 --> 00:02:53,280
Speaker 1: what's the very last thing you watched on video before

44
00:02:53,320 --> 00:02:57,680
Speaker 1: we recorded today, like VHS, No, just video, just any

45
00:02:57,880 --> 00:02:59,120
Speaker 1: video on any screen.

46
00:03:00,800 --> 00:03:03,720
Speaker 2: I can't really remember, but right now I'm feeling lost

47
00:03:03,800 --> 00:03:06,400
Speaker 2: in the woods all right.

48
00:03:06,480 --> 00:03:07,680
Speaker 4: Is that another queen reference?

49
00:03:08,400 --> 00:03:11,600
Speaker 2: That's a reference for all the parents out there who've watched.

50
00:03:11,240 --> 00:03:14,760
Speaker 4: First Into Is that the song that actually held up?

51
00:03:15,400 --> 00:03:15,840
Speaker 2: I liked.

52
00:03:15,960 --> 00:03:18,280
Speaker 4: People were saying that the music was a little lackluster.

53
00:03:19,040 --> 00:03:24,160
Speaker 2: I'm sorry. When when I get older, everything is going

54
00:03:24,240 --> 00:03:27,120
Speaker 2: to make sense. That is performed by Josh Gadd is

55
00:03:27,160 --> 00:03:28,320
Speaker 2: one of the best songs written.

56
00:03:28,520 --> 00:03:30,079
Speaker 1: Who's Who's Josh Gadd?

57
00:03:30,360 --> 00:03:31,680
Speaker 2: He plays the Little Snowman.

58
00:03:32,720 --> 00:03:38,200
Speaker 1: Yeah, okay, so so you saw that, which is entirely

59
00:03:38,240 --> 00:03:38,880
Speaker 1: a deep fake.

60
00:03:38,920 --> 00:03:42,040
Speaker 4: Basically it's CGI. You know, it's like e'sing real going

61
00:03:42,120 --> 00:03:45,760
Speaker 4: on there. I watched The Irishman, which is an interesting

62
00:03:46,280 --> 00:03:49,040
Speaker 4: thing to bring up when when we're talking about today's

63
00:03:49,040 --> 00:03:54,720
Speaker 4: topic because it uses anti aging de aging technology, which

64
00:03:54,760 --> 00:03:57,640
Speaker 4: sort of a hybrid of like video and CGI to

65
00:03:57,760 --> 00:04:01,280
Speaker 4: make the actors look younger in the earlier parts of

66
00:04:01,320 --> 00:04:03,640
Speaker 4: the film where they're you know, their younger selves. The

67
00:04:03,640 --> 00:04:06,920
Speaker 4: movie takes place over decades, and it still has a

68
00:04:06,960 --> 00:04:09,320
Speaker 4: little bit of that Uncanny Valley kind of vibe. It

69
00:04:09,360 --> 00:04:11,760
Speaker 4: looks a little just too perfect, almost like a cut

70
00:04:11,800 --> 00:04:14,400
Speaker 4: scene from a final fantasy game or something like that.

71
00:04:14,960 --> 00:04:17,800
Speaker 4: But you eventually stopped noticing it. So that was the

72
00:04:17,880 --> 00:04:19,960
Speaker 4: last I've been watching that I had to watch in

73
00:04:20,000 --> 00:04:22,159
Speaker 4: two sittings. It's quite long, and I, of course Sasey

74
00:04:22,200 --> 00:04:22,839
Speaker 4: would not approve.

75
00:04:23,160 --> 00:04:26,559
Speaker 1: I watched The Irishman on the recommendation of our own

76
00:04:26,640 --> 00:04:30,240
Speaker 1: Paul Michion control decan who has fantastic, impeccable taste in

77
00:04:30,480 --> 00:04:33,640
Speaker 1: film and also has a film of his own out

78
00:04:33,720 --> 00:04:36,640
Speaker 1: of It now on Amazon Prime. You okay with me

79
00:04:36,680 --> 00:04:37,360
Speaker 1: pluging that, Paul.

80
00:04:37,440 --> 00:04:40,400
Speaker 2: It's called Annie, Get Your Gun. There are some people

81
00:04:40,440 --> 00:04:43,240
Speaker 2: here at the apartment complex and they mean business.

82
00:04:44,320 --> 00:04:50,080
Speaker 1: It does take place in the city, Annie in the City.

83
00:04:50,440 --> 00:04:53,520
Speaker 1: Do check it out. You'll astute listeners will notice some

84
00:04:53,640 --> 00:04:57,440
Speaker 1: cameos from some of our podcast.

85
00:04:57,000 --> 00:05:00,240
Speaker 2: Cohort, especially Ben Well.

86
00:05:00,600 --> 00:05:03,800
Speaker 1: I think my favorite turn is probably one of our

87
00:05:03,839 --> 00:05:08,640
Speaker 1: super producers, Chandler Mays. He's really the scene stealer for me.

88
00:05:09,480 --> 00:05:12,720
Speaker 1: So check it out, tell us what you think, and

89
00:05:14,480 --> 00:05:16,960
Speaker 1: we can assure you, to the best of our ability

90
00:05:17,040 --> 00:05:19,359
Speaker 1: that the people who appear to be on screen saying

91
00:05:19,400 --> 00:05:23,000
Speaker 1: and doing things are actually on screen saying and doing things.

92
00:05:23,440 --> 00:05:27,839
Speaker 1: This is not a revolutionary thought. For centuries, people used

93
00:05:27,880 --> 00:05:31,719
Speaker 1: to say scene is believing, and this meant that while

94
00:05:31,839 --> 00:05:36,719
Speaker 1: people can make up anything in conversation or writing. Actually

95
00:05:36,800 --> 00:05:42,560
Speaker 1: witnessing something with your own eyes presented inarguable proof of

96
00:05:42,600 --> 00:05:48,839
Speaker 1: an event. This began to change with the rise of photography.

97
00:05:49,400 --> 00:05:53,000
Speaker 1: Photographs could be faked. During the heyday of spiritualism in

98
00:05:53,040 --> 00:05:57,320
Speaker 1: the West, it was very common for mediums or people

99
00:05:57,360 --> 00:06:00,680
Speaker 1: purporting to be mediums to fake photographs.

100
00:06:00,360 --> 00:06:02,919
Speaker 2: The old double exposure right right.

101
00:06:02,800 --> 00:06:06,440
Speaker 1: Which was an unfamiliar alien technology to the casual observer.

102
00:06:06,800 --> 00:06:09,640
Speaker 1: We cannot blame people for believing that they saw it

103
00:06:09,640 --> 00:06:12,120
Speaker 1: with their own eyes. They were unaware of the trickery

104
00:06:12,200 --> 00:06:15,240
Speaker 1: that could or could not be involved. And then things

105
00:06:15,360 --> 00:06:19,720
Speaker 1: escalate further with the dawn of moving pictures. That's when directors, filmmakers,

106
00:06:19,760 --> 00:06:23,919
Speaker 1: and other industry professionals begin working assiduously to make the

107
00:06:24,200 --> 00:06:29,520
Speaker 1: unreal seem real for fantasy's sake. For fantasy's sake, yes exactly, Matt,

108
00:06:29,680 --> 00:06:33,200
Speaker 1: to bring the fantasies of the human mind to life

109
00:06:33,320 --> 00:06:37,200
Speaker 1: of a sort on screen. Our species also quickly realized

110
00:06:37,240 --> 00:06:41,360
Speaker 1: the power inherent and film, and at times history has

111
00:06:41,520 --> 00:06:45,000
Speaker 1: hinged on specific images or even just bits of video.

112
00:06:45,320 --> 00:06:48,800
Speaker 1: So today's question, what happens in a world where seeing

113
00:06:48,920 --> 00:06:52,360
Speaker 1: is no longer believing? What happens when the line between

114
00:06:52,440 --> 00:06:57,599
Speaker 1: film fiction and film fact blur. Let's start with some

115
00:06:57,720 --> 00:07:01,720
Speaker 1: of the most immediately important video the world, the news.

116
00:07:02,240 --> 00:07:03,599
Speaker 1: Here are the facts.

117
00:07:04,000 --> 00:07:06,640
Speaker 4: So, according to a twenty eighteen survey from the Pew

118
00:07:06,800 --> 00:07:10,000
Speaker 4: Research Center, forty seven percent of Americans prefer watching the

119
00:07:10,040 --> 00:07:14,680
Speaker 4: news rather than reading or listening to it. Thirty four percent,

120
00:07:14,800 --> 00:07:17,040
Speaker 4: on the other hand, prefer to read the news and

121
00:07:17,160 --> 00:07:19,080
Speaker 4: nineteen percent to listen to it.

122
00:07:19,560 --> 00:07:22,760
Speaker 2: And I know those people who prefer their news on video,

123
00:07:23,480 --> 00:07:25,920
Speaker 2: most of them still really want to watch it on

124
00:07:26,000 --> 00:07:29,160
Speaker 2: a big screen, on a maybe not a huge screen,

125
00:07:29,440 --> 00:07:33,920
Speaker 2: but a bigger screen like a television, rather than internet video,

126
00:07:34,000 --> 00:07:36,080
Speaker 2: you know, maybe on your phone or on your computer.

127
00:07:36,600 --> 00:07:38,840
Speaker 2: And that may be surprising to a lot of people listening.

128
00:07:38,880 --> 00:07:41,160
Speaker 2: I know it's a little surprising to me, right. I

129
00:07:41,160 --> 00:07:44,600
Speaker 2: think it's because, look, this is my understanding of it.

130
00:07:44,640 --> 00:07:46,679
Speaker 2: But I think I think a lot of the people

131
00:07:46,680 --> 00:07:49,800
Speaker 2: who end up responding to a Pew survey maybe are

132
00:07:49,960 --> 00:07:53,560
Speaker 2: leaning more towards people who would be watching their televisions,

133
00:07:53,920 --> 00:07:57,360
Speaker 2: or maybe a little skewing a little older maybe.

134
00:07:57,120 --> 00:08:00,800
Speaker 1: Or the kind of people who say, sure, take a survey.

135
00:08:01,080 --> 00:08:04,080
Speaker 2: Yeah, right, That's just my read of it.

136
00:08:04,160 --> 00:08:05,239
Speaker 1: I think it's a good point.

137
00:08:05,480 --> 00:08:09,400
Speaker 2: And then Pugh found just over four and ten, So

138
00:08:09,600 --> 00:08:13,760
Speaker 2: four out of ten United States adults prefer TV, compared

139
00:08:13,800 --> 00:08:16,200
Speaker 2: with about a third who prefer the web, and then

140
00:08:16,320 --> 00:08:21,120
Speaker 2: fourteen percent who prefer radio and seven only seven percent

141
00:08:21,200 --> 00:08:23,800
Speaker 2: who actually want to read the words.

142
00:08:23,400 --> 00:08:26,160
Speaker 1: Of the news right for one reason or another. That's

143
00:08:26,160 --> 00:08:30,679
Speaker 1: the other thing. We don't have a solid methodological grasp

144
00:08:30,880 --> 00:08:36,560
Speaker 1: of how these questions were accurate, like how they were built,

145
00:08:36,640 --> 00:08:38,640
Speaker 1: how they were framed, and there's so much that goes

146
00:08:38,640 --> 00:08:43,400
Speaker 1: into there. But these numbers feel pretty solid, if surprising.

147
00:08:44,360 --> 00:08:47,720
Speaker 1: I think it's safe to say that the four of

148
00:08:47,840 --> 00:08:54,000
Speaker 1: us recording obviously longtime listeners. You know this. We don't

149
00:08:54,320 --> 00:08:58,559
Speaker 1: just go home and turn on CNN, Fox or MSNBC

150
00:08:58,840 --> 00:09:01,840
Speaker 1: or something. We we get to the edges of stuff.

151
00:09:01,840 --> 00:09:03,280
Speaker 1: We look into the weird things.

152
00:09:03,600 --> 00:09:06,640
Speaker 2: Yeah, generally, I just really quickly and maybe we do

153
00:09:06,679 --> 00:09:10,320
Speaker 2: a quick pull here. Generally, if I'm if I'm consuming

154
00:09:10,320 --> 00:09:12,880
Speaker 2: the news, it is through a written article.

155
00:09:13,280 --> 00:09:16,280
Speaker 4: What about you, guys, Yeah, I typically read stuff because

156
00:09:16,280 --> 00:09:18,559
Speaker 4: of the way we research for the show and other shows.

157
00:09:19,200 --> 00:09:20,920
Speaker 4: It tends to be a kind of our bread and butter.

158
00:09:21,080 --> 00:09:25,319
Speaker 4: Occasionally watch documentary, read parts of books. We don't always

159
00:09:25,360 --> 00:09:27,679
Speaker 4: have the luxury of reading, you know, an entire book

160
00:09:27,720 --> 00:09:31,200
Speaker 4: for one podcast episode, but definitely excerpts and chapters. And

161
00:09:31,200 --> 00:09:33,720
Speaker 4: then occasionally, you know, stuff like YouTube videos, which tend

162
00:09:33,760 --> 00:09:36,199
Speaker 4: to sort of show you where the kind of Internet

163
00:09:36,240 --> 00:09:39,240
Speaker 4: culture or the zeitgeist is, you know, coming down on

164
00:09:39,240 --> 00:09:40,640
Speaker 4: one side or the other of an issue.

165
00:09:41,040 --> 00:09:45,120
Speaker 1: I think I would watch television more often if there

166
00:09:45,280 --> 00:09:49,360
Speaker 1: were more if there was a larger degree of variants

167
00:09:49,640 --> 00:09:52,520
Speaker 1: in the narratives and stories presented in the West at least,

168
00:09:52,960 --> 00:09:55,720
Speaker 1: so people like me naturally end up watching things on

169
00:09:55,760 --> 00:09:59,400
Speaker 1: the Internet because that's the easiest way to get opposing viewpoints. Right,

170
00:09:59,679 --> 00:10:03,360
Speaker 1: you want to see Tinhua or Al Jazeera or RT,

171
00:10:04,240 --> 00:10:07,559
Speaker 1: all of which are imperfect, then you you probably are

172
00:10:07,600 --> 00:10:10,400
Speaker 1: not going to see that in your basic Comcast package, right,

173
00:10:11,240 --> 00:10:15,000
Speaker 1: So the majority of Americans don't prefer online video just yet,

174
00:10:15,320 --> 00:10:17,960
Speaker 1: even though you know, I think a lot of us

175
00:10:18,000 --> 00:10:21,840
Speaker 1: listening because we listen to podcasts, are already all up

176
00:10:21,840 --> 00:10:26,080
Speaker 1: on the internet for video content for news, right and.

177
00:10:26,000 --> 00:10:29,160
Speaker 2: For honestly audio podcasts about the news, like The Daily

178
00:10:29,400 --> 00:10:33,600
Speaker 2: or just Get Daily Zeitgeist. There's all kinds of shows

179
00:10:33,920 --> 00:10:36,680
Speaker 2: out there that will give you what you need in

180
00:10:36,720 --> 00:10:37,520
Speaker 2: an audio way.

181
00:10:37,760 --> 00:10:43,160
Speaker 1: The economists BBC, Global News, et cetera. And this this

182
00:10:43,280 --> 00:10:46,560
Speaker 1: makes sense for us. But even if the majority of

183
00:10:46,880 --> 00:10:52,480
Speaker 1: Americans or the largest swath of Americans don't prefer online

184
00:10:52,559 --> 00:10:55,480
Speaker 1: video just yet, at least for the news, there's no

185
00:10:55,600 --> 00:10:57,840
Speaker 1: arguing that the Internet has made a massive impact on

186
00:10:57,960 --> 00:11:01,520
Speaker 1: video technology and filming in jet general. And we are

187
00:11:01,640 --> 00:11:05,360
Speaker 1: in the midst of this evolution as we speak. It

188
00:11:05,440 --> 00:11:10,240
Speaker 1: is an evolution at a very fast cadence. The earlier,

189
00:11:10,559 --> 00:11:13,800
Speaker 1: the earliest, and the earlier film technology. And you guys

190
00:11:14,080 --> 00:11:17,199
Speaker 1: and Paul know this very very well. Like the earliest

191
00:11:17,200 --> 00:11:21,960
Speaker 1: stuff was super expensive, highly specialized, it was cantankerous, and

192
00:11:22,000 --> 00:11:25,520
Speaker 1: it broke a lot. Add to that, distribution channels for

193
00:11:25,720 --> 00:11:29,640
Speaker 1: anything filmed were owned by a fairly small number of

194
00:11:29,720 --> 00:11:33,000
Speaker 1: corporate or state interests, and this meant that whether your

195
00:11:33,040 --> 00:11:36,160
Speaker 1: film was a sci fi blockbuster or whether it was

196
00:11:36,400 --> 00:11:39,640
Speaker 1: just some shameless World War II propaganda, it would go

197
00:11:39,760 --> 00:11:43,400
Speaker 1: through predictable channels. The same people would shake the same

198
00:11:43,600 --> 00:11:46,040
Speaker 1: hands to get that to a theater or a screen

199
00:11:46,280 --> 00:11:50,760
Speaker 1: near you. But the technology continued to evolve. Soon people

200
00:11:50,800 --> 00:11:54,839
Speaker 1: were able to purchase televisions and instead of relying on radios,

201
00:11:55,240 --> 00:11:59,160
Speaker 1: they would be able to put an image with a sound.

202
00:11:59,480 --> 00:12:02,800
Speaker 1: This played huge role like the first televised presidential debate.

203
00:12:03,080 --> 00:12:05,120
Speaker 1: You know about this one, right, Nixon and Kennedy.

204
00:12:05,240 --> 00:12:08,560
Speaker 2: Oh yeah, just the difference in their appearances, and how

205
00:12:08,640 --> 00:12:11,360
Speaker 2: much of a difference that made outside of the words

206
00:12:11,400 --> 00:12:12,080
Speaker 2: they were saying.

207
00:12:12,160 --> 00:12:15,600
Speaker 1: Exactly. So, people listening to the radio, regardless of political party,

208
00:12:15,640 --> 00:12:18,560
Speaker 1: thought that Nixon won the debate, and people watching television,

209
00:12:18,559 --> 00:12:22,800
Speaker 1: regardless of the political party, thought that Kennedy won the debate.

210
00:12:23,360 --> 00:12:25,400
Speaker 1: So we were able to bring a small version of

211
00:12:25,440 --> 00:12:27,880
Speaker 1: the big screen to living rooms around the world, and

212
00:12:28,040 --> 00:12:32,000
Speaker 1: happily ever after, right, yes, yeah, well, so then we

213
00:12:32,080 --> 00:12:35,560
Speaker 1: had the rise of home projectors vhs, like you had

214
00:12:35,559 --> 00:12:38,120
Speaker 1: mentioned earlier in old DVD and so on. These allowed

215
00:12:38,200 --> 00:12:40,640
Speaker 1: viewers more agency. You didn't just have to stick to

216
00:12:40,679 --> 00:12:44,000
Speaker 1: the programming dictated by your TV channels. Your CBS is

217
00:12:44,040 --> 00:12:45,120
Speaker 1: your NBCs. Yeah.

218
00:12:45,120 --> 00:12:47,480
Speaker 4: It's like a rudimentary early version of what we now

219
00:12:47,520 --> 00:12:51,839
Speaker 4: think of as on demand consuming of media, which has

220
00:12:51,880 --> 00:12:53,600
Speaker 4: really changed the game entirely.

221
00:12:53,679 --> 00:12:57,880
Speaker 1: Right, right, And we're still segments of the industry are

222
00:12:57,880 --> 00:13:01,400
Speaker 1: still attempting to keep up with that change. As film

223
00:13:01,440 --> 00:13:07,760
Speaker 1: technology improved cost for equipment began to creator. Nowadays, filmmakers

224
00:13:07,840 --> 00:13:10,360
Speaker 1: don't always have to go to a studio. You don't

225
00:13:10,400 --> 00:13:13,839
Speaker 1: have to shake the same hands at Warner or whatever, and.

226
00:13:13,800 --> 00:13:17,920
Speaker 2: You could shoot a low budget movie with technology available

227
00:13:18,160 --> 00:13:21,680
Speaker 2: and you know, make VHS copies let's say, of it,

228
00:13:21,720 --> 00:13:25,000
Speaker 2: and distribute it locally or maybe two stores in a

229
00:13:25,320 --> 00:13:27,720
Speaker 2: local area. I'm just saying, like, in that time, even

230
00:13:27,800 --> 00:13:33,520
Speaker 2: before you know nowadays, it was possible to escape the studio.

231
00:13:33,520 --> 00:13:37,479
Speaker 1: Right right, And this escape from the studio, as exodus

232
00:13:37,520 --> 00:13:40,680
Speaker 1: from the studio, has accelerated the rise of home and

233
00:13:40,720 --> 00:13:45,320
Speaker 1: internet video commingled and it led us to the current world,

234
00:13:45,400 --> 00:13:47,400
Speaker 1: a world in which anyone with a little bit of

235
00:13:47,480 --> 00:13:51,240
Speaker 1: scratch can purchase tools to make their own video. Right.

236
00:13:51,280 --> 00:13:53,880
Speaker 4: And not only is it the democratization of creating, it's

237
00:13:53,880 --> 00:13:57,800
Speaker 4: the democratization of consuming, because you know, anyone can upload

238
00:13:57,800 --> 00:14:00,760
Speaker 4: a video to the internet for free on YouTube and

239
00:14:00,960 --> 00:14:05,320
Speaker 4: you can command your own audience depending on the quality

240
00:14:05,400 --> 00:14:08,120
Speaker 4: of your workhound quality is sort of a loose term there,

241
00:14:08,160 --> 00:14:11,599
Speaker 4: I guess, but at least in terms of how salacious

242
00:14:11,679 --> 00:14:14,160
Speaker 4: or how kind of hooky it is and how much

243
00:14:14,160 --> 00:14:17,880
Speaker 4: it grabs people's attention you can actually command an audience

244
00:14:17,960 --> 00:14:21,840
Speaker 4: of scale as a creator with very little overhead.

245
00:14:22,040 --> 00:14:26,440
Speaker 2: Yeah, with worldwide virtually instant distribution sick.

246
00:14:27,400 --> 00:14:30,520
Speaker 1: Yes, So if these filmmakers have an Internet connection, they

247
00:14:30,520 --> 00:14:34,320
Speaker 1: can bypass those antiquated channels of distribution send their work

248
00:14:34,320 --> 00:14:38,240
Speaker 1: across the planet. Anyone else with web access can, in theory,

249
00:14:38,800 --> 00:14:40,920
Speaker 1: watch it to their hearts content as much or as

250
00:14:40,920 --> 00:14:43,560
Speaker 1: little as they wish. But don't feel bad for the

251
00:14:43,640 --> 00:14:47,760
Speaker 1: old guard. The democratization of av technology did not lead

252
00:14:47,800 --> 00:14:50,640
Speaker 1: them to extinction. They evolved as well.

253
00:14:50,720 --> 00:14:53,880
Speaker 4: Yeah they had to. I mean, a journalist in say

254
00:14:53,920 --> 00:14:58,080
Speaker 4: Hong Kong now during the protests that we're experiencing, can

255
00:14:58,120 --> 00:15:02,320
Speaker 4: immediately post video and help out those news providers, shedding

256
00:15:02,400 --> 00:15:05,800
Speaker 4: light on events that might have been otherwise relegated to

257
00:15:05,800 --> 00:15:09,800
Speaker 4: the shadows. Reporters in Belgium or Bolivia can record political

258
00:15:09,840 --> 00:15:13,720
Speaker 4: announcements that the capitol live and stream it to millions

259
00:15:13,760 --> 00:15:15,119
Speaker 4: of viewers or followers.

260
00:15:15,320 --> 00:15:19,560
Speaker 2: The president can, you know, before having important meetings with

261
00:15:19,760 --> 00:15:23,080
Speaker 2: world leaders decide to talk for forty five minutes because

262
00:15:23,080 --> 00:15:24,000
Speaker 2: the cameras are live.

263
00:15:24,320 --> 00:15:28,200
Speaker 1: Mm and other world leaders can be caught doing their

264
00:15:28,200 --> 00:15:32,120
Speaker 1: own rendition of mean girls who are significant geopolitical events.

265
00:15:34,040 --> 00:15:37,320
Speaker 1: All in all, this is impressive stuff. Right live video

266
00:15:37,480 --> 00:15:42,520
Speaker 1: has the potential to bring our incredibly litigious and bellicose

267
00:15:42,600 --> 00:15:46,200
Speaker 1: species a little bit closer together. If everyone can see

268
00:15:46,200 --> 00:15:49,320
Speaker 1: something happening plain as day at the same time while

269
00:15:49,320 --> 00:15:52,840
Speaker 1: it's happening, what on earth is there left to argue about?

270
00:15:53,240 --> 00:15:54,680
Speaker 1: Scene is believing?

271
00:15:55,880 --> 00:15:56,120
Speaker 2: Right?

272
00:15:57,800 --> 00:15:58,760
Speaker 1: No? What?

273
00:15:59,200 --> 00:16:01,760
Speaker 2: But we'll talk about out right after word from our splatzer.

274
00:16:02,720 --> 00:16:03,200
Speaker 4: Wow.

275
00:16:05,240 --> 00:16:07,320
Speaker 1: This episode of stuff they don't want you to know

276
00:16:07,480 --> 00:16:09,320
Speaker 1: is brought to you by Express VPN.

277
00:16:09,640 --> 00:16:13,600
Speaker 2: Recently, over one hundred million people had their personal information

278
00:16:13,720 --> 00:16:17,600
Speaker 2: stolen in a major data reach. We're talking social security numbers,

279
00:16:17,640 --> 00:16:21,160
Speaker 2: contact details, credit scores and more, all taken from Capital

280
00:16:21,280 --> 00:16:23,200
Speaker 2: One customers. Oh that's me, and.

281
00:16:23,160 --> 00:16:26,840
Speaker 1: That means there's a good chance you were personally affected. Folks.

282
00:16:26,880 --> 00:16:30,320
Speaker 1: These kinds of attacks are getting more frequent and more severe.

283
00:16:30,320 --> 00:16:35,080
Speaker 1: It's not just Capital One, Zequifax, Facebook, eBay, Uber, PlayStation,

284
00:16:35,360 --> 00:16:39,000
Speaker 1: and Yahoo. They have all leaked passwords, credit card info,

285
00:16:39,320 --> 00:16:42,720
Speaker 1: and bank numbers belonging to billions of users.

286
00:16:42,880 --> 00:16:47,479
Speaker 2: That's why we use ExpressVPN to safeguard our personal data online.

287
00:16:47,520 --> 00:16:50,400
Speaker 1: According to recent reports, hackers can make up to one

288
00:16:50,440 --> 00:16:53,320
Speaker 1: thousand dollars when they sell someone's personal information on the

289
00:16:53,400 --> 00:16:57,680
Speaker 1: dark web, making people like us easy lucrative targets.

290
00:16:57,840 --> 00:17:00,800
Speaker 2: ExpressVPN is an app in the app connects with just

291
00:17:00,840 --> 00:17:03,800
Speaker 2: one click. It's lightning fast, and the best part is

292
00:17:03,960 --> 00:17:07,520
Speaker 2: ExpressVPN costs less than seven bucks a month to use.

293
00:17:07,680 --> 00:17:11,120
Speaker 1: And Listen, Honestly, if a breach can happen to Capital One,

294
00:17:11,240 --> 00:17:14,960
Speaker 1: it can easily happen to anyone else. So protect yourself

295
00:17:15,000 --> 00:17:18,720
Speaker 1: with ExpressVPN, the number one VPN rated by tech Radar,

296
00:17:18,840 --> 00:17:21,000
Speaker 1: c Net, the Verge, and countless others.

297
00:17:21,119 --> 00:17:24,719
Speaker 2: You can use our special link Express vpn dot com

298
00:17:24,800 --> 00:17:28,399
Speaker 2: slash conspiracy right now to arm yourself with an extra

299
00:17:28,520 --> 00:17:31,360
Speaker 2: three months of ExpressVPN for free.

300
00:17:31,440 --> 00:17:34,880
Speaker 1: Support the show and keep your information safe. That's ExpressVPN

301
00:17:34,960 --> 00:17:45,440
Speaker 1: dot com slash conspiracy for an extra three months free.

302
00:17:46,440 --> 00:17:48,440
Speaker 1: Here's where it gets crazy.

303
00:17:49,200 --> 00:17:52,520
Speaker 2: Hey, Remember when we said that technology, that stuff that

304
00:17:52,600 --> 00:17:54,440
Speaker 2: lets us do all the things that we like to do.

305
00:17:55,320 --> 00:18:00,520
Speaker 2: It's still evolving. Remember we talked about that. Uh, well,

306
00:18:00,520 --> 00:18:03,680
Speaker 2: we're on the precipice of this other thing, this this

307
00:18:03,800 --> 00:18:09,040
Speaker 2: new shift in video technology, and it is not I

308
00:18:09,440 --> 00:18:12,560
Speaker 2: don't see good things happening from it. It's gonna be

309
00:18:13,920 --> 00:18:18,000
Speaker 2: like amazing for those dank memes, but for everything else.

310
00:18:18,560 --> 00:18:24,040
Speaker 1: Oh boy, Yeah, it's an inherently conspiratorial shift. Today we

311
00:18:24,080 --> 00:18:27,160
Speaker 1: are talking about the rise of the deep fake, which

312
00:18:27,160 --> 00:18:30,280
Speaker 1: sounds hyperbolic but very much is not. What is a

313
00:18:30,280 --> 00:18:33,160
Speaker 1: deep fake? Somebody might be saying, well, we're glad you asked.

314
00:18:33,240 --> 00:18:40,280
Speaker 1: Our story starts with a fellow named Ian J. Goodfellow.

315
00:18:40,560 --> 00:18:43,399
Speaker 1: I think that's really good. But while you may not

316
00:18:43,480 --> 00:18:46,000
Speaker 1: have heard of his name before, while you may be

317
00:18:46,119 --> 00:18:49,320
Speaker 1: unfamiliar with the concept of deep fakes, you have almost

318
00:18:49,600 --> 00:18:54,359
Speaker 1: certainly already encountered some version of Ian Goodfellow's work. He

319
00:18:54,800 --> 00:18:59,119
Speaker 1: works extensively in areas of what we call machine learning.

320
00:18:59,160 --> 00:19:04,240
Speaker 1: Anybody who remembers our earlier conversation about machine consciousness with

321
00:19:04,400 --> 00:19:07,399
Speaker 1: a friend of the show, Damian Patrick Williams is probably

322
00:19:07,400 --> 00:19:11,440
Speaker 1: familiar with these edges of science, the bleeding edge of

323
00:19:11,520 --> 00:19:15,760
Speaker 1: artificial intelligence. Here's what Ian did. Essentially, he taught algorithms

324
00:19:16,000 --> 00:19:18,600
Speaker 1: to play games with each other, specifically to kind of

325
00:19:18,600 --> 00:19:22,160
Speaker 1: play game theory, which is still incredibly strange and important.

326
00:19:22,760 --> 00:19:26,400
Speaker 2: Well, yeah, let's talk about what deep learning is really,

327
00:19:26,400 --> 00:19:31,800
Speaker 2: because we're talking about machine learning just essentially teaching versions

328
00:19:31,800 --> 00:19:35,919
Speaker 2: of artificial intelligence how to learn. And in this case,

329
00:19:36,480 --> 00:19:38,840
Speaker 2: this is a sub field of machine learning we're going

330
00:19:38,880 --> 00:19:42,560
Speaker 2: to talk about called deep learning. And it's fascinating stuff.

331
00:19:42,880 --> 00:19:45,520
Speaker 2: It is the it's also the stuff of nightmares. It's

332
00:19:45,560 --> 00:19:48,160
Speaker 2: the stuff of our eventual future, and there's no way

333
00:19:48,200 --> 00:19:52,240
Speaker 2: around it. But it's the concept of focusing algorithms that

334
00:19:52,320 --> 00:19:56,760
Speaker 2: are focusing on algorithms that are inspired by the way

335
00:19:56,880 --> 00:20:02,280
Speaker 2: the human brain functions. And they are called artificial neural networks.

336
00:20:02,359 --> 00:20:04,720
Speaker 2: And if you're watching the final season of Silicon Valley,

337
00:20:05,240 --> 00:20:07,280
Speaker 2: you're getting kind of a crash course in that right

338
00:20:07,320 --> 00:20:10,280
Speaker 2: now as the well, as we're recording this, the penultimate

339
00:20:10,320 --> 00:20:14,800
Speaker 2: episode just came out. But anyway, it's it's really fascinating stuff.

340
00:20:14,920 --> 00:20:18,520
Speaker 2: And Goodfellow, our friend Ian J. Goodfellow actually wrote a

341
00:20:18,520 --> 00:20:19,840
Speaker 2: book on this subject.

342
00:20:20,560 --> 00:20:23,960
Speaker 1: Yeah, a book called his book about deep learning is

343
00:20:23,960 --> 00:20:24,840
Speaker 1: called deep learning.

344
00:20:24,880 --> 00:20:26,520
Speaker 2: Hey come, yeah, I know it's a coining.

345
00:20:26,560 --> 00:20:30,560
Speaker 1: And he's a busy guy. So he explains deep learning

346
00:20:31,080 --> 00:20:32,720
Speaker 1: this way. He thinks of it in terms of a

347
00:20:32,800 --> 00:20:35,840
Speaker 1: hierarchy of concepts, and he says having a hierarchy of

348
00:20:35,880 --> 00:20:40,120
Speaker 1: concepts allows a computer to learn these complicated concepts by

349
00:20:40,160 --> 00:20:43,840
Speaker 1: building them out of simpler ones, which is what we

350
00:20:43,920 --> 00:20:45,920
Speaker 1: have spent a lot of time doing it. How stuff works?

351
00:20:46,000 --> 00:20:48,199
Speaker 2: Right, have this episode even we do?

352
00:20:48,840 --> 00:20:52,400
Speaker 1: Yeah, that's how, because that's how our brains often approach things.

353
00:20:52,440 --> 00:20:56,160
Speaker 1: We build toward that gestalt. So he says, if we

354
00:20:56,240 --> 00:20:59,760
Speaker 1: draw a graph showing how these concepts are built upon

355
00:21:00,040 --> 00:21:03,360
Speaker 1: each other, we see that the graph is just visually,

356
00:21:03,480 --> 00:21:05,880
Speaker 1: it's deep. It has a ton of layers. And he says,

357
00:21:05,880 --> 00:21:08,800
Speaker 1: for this reason, we call this approach to artificial intelligence

358
00:21:09,280 --> 00:21:13,720
Speaker 1: deep learning. In plain English. What that means is that

359
00:21:13,880 --> 00:21:18,520
Speaker 1: we as a species have programs that can work more

360
00:21:18,680 --> 00:21:22,520
Speaker 1: and more like an organic brain, and artificial neural network

361
00:21:22,680 --> 00:21:27,359
Speaker 1: is meant to function more and more like a brain.

362
00:21:28,000 --> 00:21:32,560
Speaker 1: And he has one very well known invention. This is

363
00:21:32,600 --> 00:21:35,399
Speaker 1: the engine behind his work that you have already seen.

364
00:21:35,560 --> 00:21:37,400
Speaker 1: Even if you've never heard of a deep fake, you've

365
00:21:37,440 --> 00:21:39,879
Speaker 1: never heard of being goodfellow, and you've never heard of

366
00:21:40,000 --> 00:21:41,880
Speaker 1: machine or deep learning.

367
00:21:41,720 --> 00:21:45,920
Speaker 4: That's right, and it's his most well known invention innovation.

368
00:21:46,480 --> 00:21:50,920
Speaker 4: It's something that's called generative adversarial network or a general

369
00:21:51,000 --> 00:21:56,520
Speaker 4: adversarial network or GAN and gans enable algorithms to move

370
00:21:56,600 --> 00:22:03,119
Speaker 4: beyond classifying data into actually generating or creating images.

371
00:22:04,000 --> 00:22:07,159
Speaker 2: Oh yes, Now maybe you're the gears are turning in

372
00:22:07,200 --> 00:22:10,160
Speaker 2: your mind. Maybe I have seen something like is it.

373
00:22:10,119 --> 00:22:14,280
Speaker 4: Like the deep dream kind of stuff really Google's deep Dream, right, yeah,

374
00:22:14,320 --> 00:22:17,080
Speaker 4: where it would uh sort of take a like a

375
00:22:17,160 --> 00:22:21,120
Speaker 4: face or an image and then it would pull things

376
00:22:21,400 --> 00:22:23,919
Speaker 4: from elsewhere on the Internet that it's sort of matched

377
00:22:24,000 --> 00:22:27,639
Speaker 4: up to those textures or you know, spaces like to

378
00:22:27,720 --> 00:22:32,240
Speaker 4: fill with other images of say like dogs or slugs

379
00:22:32,400 --> 00:22:34,919
Speaker 4: or what have you. And then people started animating them

380
00:22:34,960 --> 00:22:40,200
Speaker 4: and they became these like hellscape kind of psychedelic nightmare images. Yeah,

381
00:22:40,280 --> 00:22:45,159
Speaker 4: very dalias, very eschery, just like super trippy really for

382
00:22:45,240 --> 00:22:46,119
Speaker 4: lack of a better term.

383
00:22:46,200 --> 00:22:48,560
Speaker 1: Yeah, you're you're on the right track there, because they

384
00:22:48,600 --> 00:22:51,840
Speaker 1: are indeed related. In terms of the science, deep dream

385
00:22:52,240 --> 00:22:56,040
Speaker 1: is a makes use of rather something called a convolutional

386
00:22:56,160 --> 00:23:00,720
Speaker 1: neural network or a com net or CNN, which could

387
00:23:00,720 --> 00:23:04,680
Speaker 1: be a little confusing. So they're very similar approaches.

388
00:23:05,880 --> 00:23:06,560
Speaker 2: At basis.

389
00:23:06,680 --> 00:23:11,560
Speaker 1: So these generational adversarial networks are trying to trick each other.

390
00:23:11,600 --> 00:23:17,360
Speaker 1: They can move beyond classifying data into generating or creating data,

391
00:23:17,400 --> 00:23:23,000
Speaker 1: generating or creating images. So these two networks, these two

392
00:23:23,080 --> 00:23:27,600
Speaker 1: generative adversarial networks they attempt to fool each other into

393
00:23:27,640 --> 00:23:31,040
Speaker 1: thinking that a given image is real and using as

394
00:23:31,080 --> 00:23:33,600
Speaker 1: little as one image. From that back and forth between

395
00:23:33,640 --> 00:23:38,840
Speaker 1: what's called the generative and the discriminative sides of this thing.

396
00:23:40,520 --> 00:23:44,159
Speaker 1: Just using one image, they can create a video clip

397
00:23:44,400 --> 00:23:46,639
Speaker 1: of a person, so they can animate a picture, and

398
00:23:46,680 --> 00:23:50,040
Speaker 1: they also you can also take it a step further

399
00:23:50,320 --> 00:23:53,639
Speaker 1: and have that animation speaking and what sounds like that

400
00:23:53,680 --> 00:23:58,639
Speaker 1: person or that image's voice. Samsung's AI Center released a

401
00:23:58,720 --> 00:24:02,920
Speaker 1: report on the science behind DAN and they said such

402
00:24:02,920 --> 00:24:06,680
Speaker 1: an approach is able to learn highly realistic and personalized

403
00:24:06,720 --> 00:24:10,600
Speaker 1: talking head models of new people and even portrait paintings.

404
00:24:10,800 --> 00:24:12,720
Speaker 2: It's just people, new people.

405
00:24:12,520 --> 00:24:16,280
Speaker 1: Created people, generated humans, and they look great.

406
00:24:16,640 --> 00:24:17,400
Speaker 2: They really do.

407
00:24:17,480 --> 00:24:21,840
Speaker 1: I'll say it, some of them are attractive. If you

408
00:24:21,960 --> 00:24:25,960
Speaker 1: didn't know, and you just saw a picture of one

409
00:24:26,000 --> 00:24:30,280
Speaker 1: of these generated images on your dating app of choice, yeah,

410
00:24:30,760 --> 00:24:32,919
Speaker 1: there are a couple you would probably swipe right on.

411
00:24:33,160 --> 00:24:34,960
Speaker 2: I wouldn't know which way to swipe because I don't

412
00:24:35,000 --> 00:24:38,200
Speaker 2: understand those things, but I totally get what you're saying.

413
00:24:38,560 --> 00:24:41,440
Speaker 1: So and it's startling because now even now you can

414
00:24:41,480 --> 00:24:45,679
Speaker 1: take tests where you attempt to identify a real person

415
00:24:45,720 --> 00:24:50,120
Speaker 1: from a generated talking head or image, and it's tough. Yeah,

416
00:24:50,160 --> 00:24:53,560
Speaker 1: we're getting closer and closer to traversing the Uncanny Valley.

417
00:24:53,840 --> 00:24:59,200
Speaker 2: Oh yeah, well, and it's it really is frustrating that

418
00:24:59,280 --> 00:25:02,000
Speaker 2: it's not easy year because for so long there and

419
00:25:02,040 --> 00:25:05,159
Speaker 2: I think you, Noel, you've referenced Final Fantasy cut scenes

420
00:25:05,240 --> 00:25:08,040
Speaker 2: or something earlier on. And I remember when that Final

421
00:25:08,080 --> 00:25:10,879
Speaker 2: Fantasy movie came out a long time ago. There was

422
00:25:10,880 --> 00:25:13,040
Speaker 2: almost all too good at that point. Well it was, yeah,

423
00:25:13,040 --> 00:25:15,840
Speaker 2: it was all computer generated and it looked fantastic. It

424
00:25:15,880 --> 00:25:19,040
Speaker 2: was a feature film. I remember thinking how incredible it

425
00:25:19,280 --> 00:25:22,760
Speaker 2: was seeing that, and then seeing something like Avatar, where

426
00:25:22,800 --> 00:25:26,840
Speaker 2: you've got these I forget what they're called, the nav

427
00:25:27,480 --> 00:25:31,160
Speaker 2: the nave that don't look human necessarily, but they look

428
00:25:31,240 --> 00:25:34,560
Speaker 2: real enough, right, And then when you get to something

429
00:25:34,600 --> 00:25:37,760
Speaker 2: like this and you're looking at these just the portraits,

430
00:25:37,800 --> 00:25:43,320
Speaker 2: even it feels it feels pretty scary, not being able

431
00:25:43,359 --> 00:25:46,400
Speaker 2: to trust your eyes to know if something is real

432
00:25:46,480 --> 00:25:49,119
Speaker 2: or not, even though it's generated. Either way, if it

433
00:25:49,160 --> 00:25:51,399
Speaker 2: was an actual image of a person and it was

434
00:25:51,480 --> 00:25:54,400
Speaker 2: taken converted into data you know, ones and zeros, then

435
00:25:54,440 --> 00:25:58,320
Speaker 2: displayed on your screen, that's not a real person necessarily

436
00:25:58,400 --> 00:26:01,639
Speaker 2: you're looking at the representation, but just knowing that a

437
00:26:01,680 --> 00:26:05,640
Speaker 2: computer can fool you that hard is pretty.

438
00:26:05,440 --> 00:26:06,439
Speaker 1: Cruel and very easily.

439
00:26:06,640 --> 00:26:10,280
Speaker 4: Yeah, that's kind of why I wonder too, why you know,

440
00:26:10,480 --> 00:26:12,840
Speaker 4: to be fair, a lot of the de aging stuff

441
00:26:12,880 --> 00:26:15,920
Speaker 4: in movies is quite good, so much better than it's

442
00:26:15,920 --> 00:26:18,280
Speaker 4: ever been. But there were a couple spots in The

443
00:26:18,320 --> 00:26:20,840
Speaker 4: Irishman where I was like, really, like, I thought the

444
00:26:20,840 --> 00:26:24,040
Speaker 4: technology was better than this, and it is, But I

445
00:26:24,040 --> 00:26:27,119
Speaker 4: guess it's different when you're making a younger version that

446
00:26:27,200 --> 00:26:30,280
Speaker 4: has to then coincide with written lines and map up

447
00:26:30,320 --> 00:26:33,280
Speaker 4: to you know, an actor's face and you know, look

448
00:26:33,320 --> 00:26:35,280
Speaker 4: believable in terms of the way the mouth moves and

449
00:26:35,359 --> 00:26:38,040
Speaker 4: acting ticks and all of that stuff that's specific to

450
00:26:38,840 --> 00:26:42,840
Speaker 4: a scene rather than a pre existing video of some kind, right.

451
00:26:42,720 --> 00:26:46,000
Speaker 1: And it also goes down to how much existing footage

452
00:26:46,160 --> 00:26:49,520
Speaker 1: they may be able to obtain of when the person

453
00:26:49,720 --> 00:26:52,199
Speaker 1: was that age. So that's one of the reasons why

454
00:26:52,800 --> 00:26:56,440
Speaker 1: it probably works best with celebrities and political figures because

455
00:26:56,720 --> 00:27:00,200
Speaker 1: there's just you know, not everybody did taxi driver or

456
00:27:00,240 --> 00:27:04,720
Speaker 1: a kid, right. So the stuff with DeNiro specifically, I

457
00:27:04,720 --> 00:27:07,760
Speaker 1: would argue he just always looked vaguely in his late

458
00:27:08,119 --> 00:27:12,240
Speaker 1: thirties to early forties and somewhat perturbed, like, right, was

459
00:27:12,280 --> 00:27:15,200
Speaker 1: it a short or is he mad about a relationship?

460
00:27:15,320 --> 00:27:18,040
Speaker 1: You know what I mean? His face has a story.

461
00:27:18,080 --> 00:27:21,000
Speaker 2: It tells I want to bring something up here quickly.

462
00:27:21,040 --> 00:27:26,480
Speaker 2: With the the high priced effects that were going on

463
00:27:26,560 --> 00:27:28,879
Speaker 2: in a movie like The Irishman, with these aging effects

464
00:27:29,560 --> 00:27:32,960
Speaker 2: this these are designed to be displayed in the highest

465
00:27:33,040 --> 00:27:37,600
Speaker 2: resolution possible. So you're talking four K, ten ADP something

466
00:27:37,640 --> 00:27:40,040
Speaker 2: like that. It's designed to do that, right, it's on

467
00:27:40,080 --> 00:27:42,440
Speaker 2: a streaming service. They don't know what you're going to

468
00:27:42,520 --> 00:27:44,879
Speaker 2: play it on what screen, but it's got to be

469
00:27:45,040 --> 00:27:48,200
Speaker 2: high res, right, And what we're saying is it's fairly

470
00:27:48,280 --> 00:27:51,280
Speaker 2: easy to discern that something is going on here at

471
00:27:51,280 --> 00:27:55,159
Speaker 2: that high resolution. But what if it's a much lower resolution,

472
00:27:55,680 --> 00:27:58,919
Speaker 2: more grainy, like more are destroyed like a gift, like

473
00:27:58,960 --> 00:28:02,800
Speaker 2: a small YouTube video that isn't maybe ten ADP or something,

474
00:28:03,000 --> 00:28:07,560
Speaker 2: or a cell phone camera video that then gets uploaded integrated.

475
00:28:08,119 --> 00:28:12,560
Speaker 2: It changes our ability to discern some of these things.

476
00:28:12,640 --> 00:28:14,480
Speaker 4: And we'll get to this, but again, all of this

477
00:28:14,600 --> 00:28:17,320
Speaker 4: comes down to these algorithms, which, as it turns out,

478
00:28:17,359 --> 00:28:22,879
Speaker 4: require an insane amount of computing power. Yes, yeah, even

479
00:28:22,880 --> 00:28:26,040
Speaker 4: to do in these low res forms right for now,

480
00:28:26,160 --> 00:28:28,800
Speaker 4: at least exactly right. So think of it this way

481
00:28:29,400 --> 00:28:31,600
Speaker 4: now without too much of a hassle. As you are

482
00:28:31,640 --> 00:28:37,960
Speaker 4: listening to today's episode, you can get this technology online.

483
00:28:38,560 --> 00:28:43,000
Speaker 4: You can create videos that are nearly impossible to identify

484
00:28:43,120 --> 00:28:47,080
Speaker 4: as quote unquote fake. For a fun example, we just

485
00:28:47,120 --> 00:28:50,600
Speaker 4: want to keep it innocuous before we have to strap

486
00:28:50,640 --> 00:28:53,280
Speaker 4: in and go down this rabbit hole. For fun example,

487
00:28:53,320 --> 00:28:55,080
Speaker 4: Let's do two fun examples. Let's say you have a

488
00:28:55,120 --> 00:28:59,480
Speaker 4: friend who knows you love Marvel movies, and so for

489
00:29:00,080 --> 00:29:02,680
Speaker 4: you know, your birthday, your King Seniera or whatever, they

490
00:29:02,720 --> 00:29:05,520
Speaker 4: make a deep fake video where it looks like you're

491
00:29:05,520 --> 00:29:09,120
Speaker 4: in a Marvel film. The Avengers all assemble, and holy smokes,

492
00:29:09,160 --> 00:29:09,880
Speaker 4: there's Derek.

493
00:29:10,040 --> 00:29:11,520
Speaker 2: That's fun, all right.

494
00:29:11,440 --> 00:29:14,880
Speaker 1: That seems fun. What a thoughtful kind gift. Or to

495
00:29:14,920 --> 00:29:18,480
Speaker 1: make it a little more applicable to our show. One

496
00:29:18,520 --> 00:29:22,000
Speaker 1: thing that would be a fantastic deep fake present for

497
00:29:22,120 --> 00:29:24,400
Speaker 1: our very own super producer, Paul Michigan Control Deck, and

498
00:29:24,680 --> 00:29:27,320
Speaker 1: would be to take an Applebee's commercial and just put

499
00:29:27,400 --> 00:29:30,920
Speaker 1: him in it. Oh man, And so he's the person there,

500
00:29:31,040 --> 00:29:36,440
Speaker 1: you know, gesturing in amazement as the ribs and the

501
00:29:36,440 --> 00:29:41,400
Speaker 1: bottomless Jalapaniel poppers or whatever come out, and we even

502
00:29:41,480 --> 00:29:44,160
Speaker 1: have him say the tagline in his voice. That is

503
00:29:44,240 --> 00:29:47,080
Speaker 1: so much fun. But that is not the only use

504
00:29:47,240 --> 00:29:50,920
Speaker 1: of this technology. Make no mistake, the lid of Pandora's jar,

505
00:29:51,080 --> 00:29:54,360
Speaker 1: and it was a jar, is unscrewed. This technology is

506
00:29:54,400 --> 00:29:57,280
Speaker 1: no longer theoretical. It is very real, and it is

507
00:29:57,360 --> 00:30:01,960
Speaker 1: immensely dangerous. Why we'll tell you after a word from our.

508
00:30:01,840 --> 00:30:10,680
Speaker 2: Sponsor, psych We're not gonna tell you.

509
00:30:11,520 --> 00:30:12,520
Speaker 1: It's not even us.

510
00:30:14,240 --> 00:30:17,680
Speaker 2: Oh what you thought this was conspiracy stuff. We're just

511
00:30:17,760 --> 00:30:18,920
Speaker 2: your car talking to you.

512
00:30:19,200 --> 00:30:24,360
Speaker 1: Is my real voice, and Robert de Niro so terrible

513
00:30:24,440 --> 00:30:28,360
Speaker 1: denu impressions. Aside, there are applications of deep fakes which

514
00:30:28,400 --> 00:30:34,080
Speaker 1: should trouble every single person or bought listening to this show.

515
00:30:35,080 --> 00:30:38,880
Speaker 1: I did not initially think of the first application, which

516
00:30:39,040 --> 00:30:41,800
Speaker 1: was dumb and naive of me. Apparently one of the

517
00:30:41,840 --> 00:30:45,920
Speaker 1: first things people tried to do when gan technology got

518
00:30:46,000 --> 00:30:48,280
Speaker 1: out of its got out of its research R and

519
00:30:48,360 --> 00:30:52,040
Speaker 1: D Heidi Hole was to apply it to pornography, and

520
00:30:52,160 --> 00:30:56,200
Speaker 1: pornography drives a lot of technology. I mean, arguably, there's

521
00:30:56,240 --> 00:30:58,560
Speaker 1: a very good case to be made that the reason

522
00:30:58,680 --> 00:31:01,760
Speaker 1: VHS won out over Beta Max was because the porn

523
00:31:01,880 --> 00:31:03,200
Speaker 1: studios went with VHS.

524
00:31:03,800 --> 00:31:06,360
Speaker 2: Yeah. I don't want to be crass or get too

525
00:31:06,440 --> 00:31:10,040
Speaker 2: much away here, but I do remember far before this

526
00:31:10,200 --> 00:31:13,720
Speaker 2: technology was available, when photoshopping was really the only option,

527
00:31:14,120 --> 00:31:17,600
Speaker 2: there were sites, i want to say, early on in

528
00:31:17,680 --> 00:31:22,560
Speaker 2: the Internet where it was just sites dedicated to celebrity pornography,

529
00:31:22,680 --> 00:31:26,800
Speaker 2: where it was just photoshopped images on purpose. That would

530
00:31:26,840 --> 00:31:32,720
Speaker 2: be pretty crazy applying it to video with this new technology.

531
00:31:33,480 --> 00:31:36,880
Speaker 1: Yikes, black mirror esque, right, Like, imagine you are a

532
00:31:37,000 --> 00:31:39,640
Speaker 1: creep with a crush on someone. It could be a colleague,

533
00:31:39,720 --> 00:31:42,280
Speaker 1: a classmate, a celebrity, you know what, it could be

534
00:31:42,920 --> 00:31:46,680
Speaker 1: an historical figure. Maybe King Touch just really does it

535
00:31:46,760 --> 00:31:47,760
Speaker 1: for you for some reason.

536
00:31:49,080 --> 00:31:51,040
Speaker 2: Actual King Touch, Yeah, the actual one.

537
00:31:51,160 --> 00:31:52,680
Speaker 1: We know what the guy actually.

538
00:31:52,440 --> 00:31:54,560
Speaker 2: Looks like now, and not Steve Martin.

539
00:31:54,680 --> 00:31:57,800
Speaker 1: Not Steve Martin. He's one of the best banjo players

540
00:31:57,800 --> 00:31:57,960
Speaker 1: in the.

541
00:31:57,920 --> 00:31:59,800
Speaker 2: World, and he really does do it for me at least,

542
00:31:59,840 --> 00:32:02,440
Speaker 2: but in this case, okay.

543
00:32:02,280 --> 00:32:05,640
Speaker 1: So let's Steve Martin example. Then now there's nothing wrong

544
00:32:05,760 --> 00:32:10,040
Speaker 1: with having a crush. It becomes creepy. However, if you

545
00:32:10,240 --> 00:32:14,240
Speaker 1: use deep fake technology and put Steve Martin's face on

546
00:32:14,360 --> 00:32:18,200
Speaker 1: the face of someone in a pornographic video, especially when

547
00:32:18,240 --> 00:32:22,960
Speaker 1: the video will genuinely look real and it will sound

548
00:32:23,360 --> 00:32:27,400
Speaker 1: like them. This is not science fiction. This is the

549
00:32:27,520 --> 00:32:31,560
Speaker 1: idea behind a website with the immensely creative name Deep Nudes.

550
00:32:31,760 --> 00:32:33,560
Speaker 2: Now, boy, Deep Nudes.

551
00:32:33,400 --> 00:32:38,400
Speaker 1: Did exactly what we just described here. Luckily, the founder

552
00:32:38,640 --> 00:32:43,480
Speaker 1: eventually canceled the site's launch and they had a public

553
00:32:43,560 --> 00:32:44,240
Speaker 1: statement about it.

554
00:32:44,760 --> 00:32:48,120
Speaker 4: Yeah, and the founder actually eventually canceled launching the site,

555
00:32:48,840 --> 00:32:52,440
Speaker 4: noting that quote the probability that people will misuse it

556
00:32:53,040 --> 00:32:57,160
Speaker 4: is too high. Oh. I never would never would have

557
00:32:57,160 --> 00:32:57,440
Speaker 4: thought that.

558
00:32:57,840 --> 00:33:02,440
Speaker 2: What somebody's gonna misuse this.

559
00:33:04,280 --> 00:33:06,840
Speaker 1: There's a great Mitchell and Web sketch about like an

560
00:33:06,880 --> 00:33:10,600
Speaker 1: evil scientist where it's like I built the Ultra violence

561
00:33:10,760 --> 00:33:14,840
Speaker 1: Lazer to save the world, not destroy it. Nice, I'll

562
00:33:14,880 --> 00:33:16,800
Speaker 1: send it to you guys. Maybe we could post it

563
00:33:17,080 --> 00:33:20,640
Speaker 1: on here's where it gets crazy. But that's just one use.

564
00:33:20,760 --> 00:33:23,040
Speaker 1: That's the immediate one. And again I don't know about you, guys,

565
00:33:23,080 --> 00:33:27,080
Speaker 1: by a felt naive for not immediately assuming that's what

566
00:33:27,160 --> 00:33:27,720
Speaker 1: would happen.

567
00:33:28,240 --> 00:33:32,000
Speaker 2: I yeah, I think it's pretty obvious. It's just hey, cheers,

568
00:33:32,040 --> 00:33:34,400
Speaker 2: guys for not always thinking about pornography.

569
00:33:34,760 --> 00:33:35,120
Speaker 1: I guess.

570
00:33:35,160 --> 00:33:35,200
Speaker 2: So.

571
00:33:35,480 --> 00:33:37,920
Speaker 1: Yeah, it just felt awkward when I was talking to

572
00:33:38,200 --> 00:33:41,320
Speaker 1: some contacts about this off air and they looked at

573
00:33:41,400 --> 00:33:45,320
Speaker 1: me like I was from you know, a different universe

574
00:33:45,360 --> 00:33:50,080
Speaker 1: or time period. When they said, they're like, yeah, porn, Ben,

575
00:33:50,160 --> 00:33:55,000
Speaker 1: it's that's why people have technology, is to get better porn,

576
00:33:55,760 --> 00:33:58,360
Speaker 1: which I don't know. I don't know whether that's completely true,

577
00:33:58,400 --> 00:34:01,480
Speaker 1: but it was a weird night. There's another use of

578
00:34:01,680 --> 00:34:06,760
Speaker 1: deep fakes that is more apparent and has the potential.

579
00:34:06,920 --> 00:34:13,120
Speaker 1: So like this fake pornography or this fake rendition of

580
00:34:13,160 --> 00:34:16,439
Speaker 1: people in these intimate times, it can ruin an individual's life.

581
00:34:16,719 --> 00:34:19,640
Speaker 1: It could be blackmail, and it could be blackmail. But

582
00:34:19,920 --> 00:34:23,799
Speaker 1: there's another version of a deep fake, a weaponized deep fake,

583
00:34:24,000 --> 00:34:27,080
Speaker 1: that could ruin the lives of hundreds of thousands or

584
00:34:27,200 --> 00:34:28,120
Speaker 1: millions of people.

585
00:34:28,239 --> 00:34:30,960
Speaker 4: Yeah, because we can't forget that a lot of countries,

586
00:34:31,000 --> 00:34:37,000
Speaker 4: including US, are very heavily entrenched in this notion of

587
00:34:37,040 --> 00:34:40,200
Speaker 4: asymmetrical warfare called cyber war.

588
00:34:41,080 --> 00:34:43,279
Speaker 2: Awkwardly, yeah, yeah, a little bit.

589
00:34:43,920 --> 00:34:47,960
Speaker 4: Many world leaders have extensive, like you said, Ben, video

590
00:34:48,040 --> 00:34:53,960
Speaker 4: footage out there of themselves at functions, giving speeches, events

591
00:34:54,040 --> 00:34:57,400
Speaker 4: and the like. There's more than enough for Gan to

592
00:34:58,080 --> 00:34:59,279
Speaker 4: work from here.

593
00:34:59,520 --> 00:35:04,040
Speaker 1: Yeah, so let's say, who can we use as an example?

594
00:35:04,160 --> 00:35:09,480
Speaker 1: All right, Matt? He did Steve Berner earlier. Yeah, okay, okay.

595
00:35:09,600 --> 00:35:12,719
Speaker 1: So let's say let's say, Matt, you and knowl are

596
00:35:13,080 --> 00:35:17,360
Speaker 1: the leaders of these different opposing countries. It's been a

597
00:35:17,400 --> 00:35:20,440
Speaker 1: lot of tension for a while. Okay, So if there

598
00:35:20,520 --> 00:35:23,080
Speaker 1: was someone either on one of your sides or a

599
00:35:23,160 --> 00:35:25,880
Speaker 1: third party, let's say I'm a country that just messes

600
00:35:25,880 --> 00:35:28,520
Speaker 1: with other countries for fun, right, I'm Russia, I'm the US,

601
00:35:28,680 --> 00:35:32,759
Speaker 1: I'm one of the hits, right, And I say, you

602
00:35:32,840 --> 00:35:35,600
Speaker 1: know what's going to be great. I can't take on

603
00:35:37,600 --> 00:35:43,520
Speaker 1: no Landia or the Republic of Frederick with conventional military might.

604
00:35:43,880 --> 00:35:46,839
Speaker 1: So I'm going to turn them, turn them against each other.

605
00:35:46,880 --> 00:35:49,799
Speaker 1: I'm going to foment instability, or I'm going to mess

606
00:35:49,840 --> 00:35:53,399
Speaker 1: with their election by just going on Facebook. I don't

607
00:35:53,440 --> 00:35:55,279
Speaker 1: need to launch an ICBM. I'm just going to go

608
00:35:55,360 --> 00:35:57,920
Speaker 1: on Facebook and Twitter. I'm going to make fake videos

609
00:35:58,320 --> 00:36:01,600
Speaker 1: of leaders of both countries Landy and the Republic of Frederick,

610
00:36:01,719 --> 00:36:03,239
Speaker 1: and I'm going to have him say things that they

611
00:36:03,280 --> 00:36:07,480
Speaker 1: would never actually say. This sounds like a we're building

612
00:36:07,520 --> 00:36:11,480
Speaker 1: toward a comedy bit, but there's a real, real fake video,

613
00:36:12,200 --> 00:36:15,759
Speaker 1: multiple levels of Nancy Pelosi, a politician here.

614
00:36:15,760 --> 00:36:16,319
Speaker 2: In the US.

615
00:36:16,480 --> 00:36:18,239
Speaker 1: And have you guys seen this video? I have not

616
00:36:18,360 --> 00:36:23,080
Speaker 1: seen this one, so it's it came out with a

617
00:36:23,239 --> 00:36:26,319
Speaker 1: nice side by side view, but the deep fake video

618
00:36:26,440 --> 00:36:29,160
Speaker 1: was also propagated on its own. It's very interesting to

619
00:36:29,280 --> 00:36:32,160
Speaker 1: watch the difference between the two. This gives us a

620
00:36:32,280 --> 00:36:36,439
Speaker 1: chance to rewrite history in a disturbingly or Wellian way.

621
00:36:36,600 --> 00:36:41,120
Speaker 1: What happens if, for example, all original let's see Noel

622
00:36:41,200 --> 00:36:46,000
Speaker 1: makes an historic speech in Nolandia that triggers a new

623
00:36:46,280 --> 00:36:51,320
Speaker 1: golden age for the country, and someone destroys the original

624
00:36:51,400 --> 00:36:55,880
Speaker 1: copies of this profound speech and replaces it with a

625
00:36:56,000 --> 00:37:00,960
Speaker 1: deep fake. And then the last living generation, the less

626
00:37:01,000 --> 00:37:04,120
Speaker 1: people who are there when the original speech was propagated,

627
00:37:05,000 --> 00:37:07,759
Speaker 1: they can keep its memory alive, but when they die,

628
00:37:08,800 --> 00:37:11,600
Speaker 1: history has in a very real way changed. Now.

629
00:37:11,640 --> 00:37:13,600
Speaker 4: I don't know if this counts or not, but uh,

630
00:37:13,800 --> 00:37:16,120
Speaker 4: there was a brief, brief period recently where there was

631
00:37:16,160 --> 00:37:18,680
Speaker 4: some what I would consider deep fake gifts that were

632
00:37:18,719 --> 00:37:22,000
Speaker 4: making the rounds. There's like Obama on a skateboard, there's

633
00:37:22,080 --> 00:37:25,120
Speaker 4: one of the Pope doing a magic trick where he

634
00:37:25,320 --> 00:37:29,440
Speaker 4: like pulls the the tablecloth out from under some kind

635
00:37:29,480 --> 00:37:32,000
Speaker 4: of votives like at a like on an event like

636
00:37:32,040 --> 00:37:34,040
Speaker 4: a live CNN stream.

637
00:37:33,880 --> 00:37:36,239
Speaker 1: Which I choose to I know it's fake, but I'm

638
00:37:36,320 --> 00:37:37,120
Speaker 1: just gonna believe in it.

639
00:37:37,320 --> 00:37:38,239
Speaker 2: But they're so good.

640
00:37:39,239 --> 00:37:41,920
Speaker 4: Yeah, I wanted to believe in it as well, especially

641
00:37:42,000 --> 00:37:43,239
Speaker 4: the Obama on a skateboard one.

642
00:37:43,400 --> 00:37:45,680
Speaker 2: But is that kind of fall is that sort of

643
00:37:45,719 --> 00:37:46,960
Speaker 2: fall under these this category.

644
00:37:47,080 --> 00:37:50,279
Speaker 1: Yeah, yeah, that's a that's a less dangerous version, you know,

645
00:37:50,400 --> 00:37:53,719
Speaker 1: because that's fun. See seeing the Pope pull off a

646
00:37:53,840 --> 00:37:58,880
Speaker 1: dope magic trick is not gonna like foment instability in

647
00:37:59,000 --> 00:38:03,319
Speaker 1: South America. But going back to our example, let's let's assume, uh,

648
00:38:03,440 --> 00:38:06,799
Speaker 1: the Republic of Frederick and Nolandia are kind of are

649
00:38:06,880 --> 00:38:09,320
Speaker 1: beefed up to the level of like Pakistan and India

650
00:38:09,800 --> 00:38:14,239
Speaker 1: or Israel and Palestine, and all of a sudden, in

651
00:38:14,400 --> 00:38:18,319
Speaker 1: no traceable way, the public of both countries gets hit

652
00:38:18,440 --> 00:38:22,480
Speaker 1: with this barrage of videos that seem to say, seems

653
00:38:22,560 --> 00:38:27,600
Speaker 1: seem to have you know, the benevolent dictator Matt Frederick

654
00:38:27,960 --> 00:38:31,480
Speaker 1: and the Prime Minister of no Landia and Noel Brown

655
00:38:31,600 --> 00:38:31,960
Speaker 1: seems to.

656
00:38:31,960 --> 00:38:34,800
Speaker 2: He's also a dictator by the ways, officially not in

657
00:38:34,880 --> 00:38:35,400
Speaker 2: his title.

658
00:38:35,680 --> 00:38:39,520
Speaker 1: Okay, so so the uh, let's these these guys have

659
00:38:39,640 --> 00:38:43,520
Speaker 1: these videos wherein each of them are announcing their their

660
00:38:43,640 --> 00:38:48,440
Speaker 1: intention to deploy nuclear weapons and mop the mop the

661
00:38:48,520 --> 00:38:52,759
Speaker 1: adjacent country clean. Let no stone stand on another How

662
00:38:52,800 --> 00:38:55,280
Speaker 1: would you know if you're the audience of this other country,

663
00:38:55,400 --> 00:38:57,959
Speaker 1: how would you know whether these were real or fake?

664
00:38:58,239 --> 00:39:00,440
Speaker 1: How would you react? How much time would you have

665
00:39:00,680 --> 00:39:04,040
Speaker 1: if you think that you literally saw the leader of

666
00:39:04,160 --> 00:39:07,040
Speaker 1: the country that you previously went to war with saying

667
00:39:07,320 --> 00:39:09,719
Speaker 1: that's it, We're launching in five Well, it's.

668
00:39:09,920 --> 00:39:14,880
Speaker 2: It's really scary because the way that would function, it

669
00:39:15,000 --> 00:39:17,880
Speaker 2: would be posted somewhere and it would become viral on

670
00:39:18,000 --> 00:39:20,719
Speaker 2: social media so greatly that's the only way that it

671
00:39:20,760 --> 00:39:25,480
Speaker 2: would propagate, But it would propagate likely. And I'm just

672
00:39:25,520 --> 00:39:27,760
Speaker 2: gonna say personally, what I would do in that situation

673
00:39:28,080 --> 00:39:31,120
Speaker 2: is go to the standby that we discussed at the top.

674
00:39:31,160 --> 00:39:34,080
Speaker 2: I would probably turn a TV on somewhere just to

675
00:39:34,239 --> 00:39:37,799
Speaker 2: check and see if somebody is talking about it seriously,

676
00:39:38,280 --> 00:39:41,480
Speaker 2: you know, on one of the major outlets. The problem is,

677
00:39:41,600 --> 00:39:45,200
Speaker 2: what if you fool them. This happens all the time,

678
00:39:45,840 --> 00:39:47,480
Speaker 2: even with quote unquote fake news.

679
00:39:47,520 --> 00:39:50,880
Speaker 4: Everyone's so into getting the scoop with these the fast

680
00:39:51,040 --> 00:39:54,400
Speaker 4: cycle the turnaround of news, that everyone wants to be

681
00:39:54,480 --> 00:39:56,520
Speaker 4: the first, so they tend to not vet things like

682
00:39:56,640 --> 00:39:59,439
Speaker 4: they used to, and that's how you often get these

683
00:40:00,080 --> 00:40:03,520
Speaker 4: misreporting of election results, et cetera. You know, and this

684
00:40:03,680 --> 00:40:06,280
Speaker 4: is just perfect example. If someone's like, here's a good example.

685
00:40:06,520 --> 00:40:09,399
Speaker 4: There is a deep fake Donald Trump pe tape that's

686
00:40:09,440 --> 00:40:14,120
Speaker 4: out there. What if that had been pushed out, you know,

687
00:40:14,840 --> 00:40:18,520
Speaker 4: by let's say, you know, some a network opposing that

688
00:40:18,600 --> 00:40:22,840
Speaker 4: the Trump administration, that didn't happen. Now that we know

689
00:40:23,160 --> 00:40:25,200
Speaker 4: a little bit more about deep fakes being a thing,

690
00:40:25,320 --> 00:40:27,160
Speaker 4: maybe there's some caution. No one wants to be the

691
00:40:27,239 --> 00:40:32,120
Speaker 4: news agency that does that. But with an inflammatory enough thing,

692
00:40:32,920 --> 00:40:34,719
Speaker 4: maybe you don't have time to think about it.

693
00:40:35,040 --> 00:40:37,359
Speaker 2: Well see, yeah, here's here's what I wanted to bring

694
00:40:37,480 --> 00:40:42,400
Speaker 2: up in the examples we're talking about with you know,

695
00:40:42,560 --> 00:40:46,200
Speaker 2: a speech or something that exists. So you so, in theory,

696
00:40:46,239 --> 00:40:49,840
Speaker 2: you would take the base level video of the speech

697
00:40:50,280 --> 00:40:53,600
Speaker 2: and maybe maybe that audio. Then you'd manipulate the audio

698
00:40:53,719 --> 00:40:56,279
Speaker 2: and then the face right or something to where the

699
00:40:57,239 --> 00:41:01,320
Speaker 2: the words being spoken are different on that video. I

700
00:41:01,440 --> 00:41:06,120
Speaker 2: think the scariest ones, the scariest versions of this are

701
00:41:06,480 --> 00:41:09,200
Speaker 2: where it's supposed to be a hidden camera or something,

702
00:41:09,840 --> 00:41:12,600
Speaker 2: where it's, like I was saying before, it's so degraded

703
00:41:13,040 --> 00:41:15,840
Speaker 2: it's difficult to truly make out what's going on. But

704
00:41:15,920 --> 00:41:19,440
Speaker 2: you can tell that, oh, that's definitely Billy Aikner. And

705
00:41:19,960 --> 00:41:21,640
Speaker 2: where you can, you know your brain is at least

706
00:41:21,680 --> 00:41:24,040
Speaker 2: telling you that, and it sounds like Billy Aikner's voice

707
00:41:24,080 --> 00:41:24,720
Speaker 2: saying things.

708
00:41:24,840 --> 00:41:25,680
Speaker 1: Who's Billy Aikner?

709
00:41:26,880 --> 00:41:29,560
Speaker 2: Billy on the street. Billy on the street, I mean,

710
00:41:29,600 --> 00:41:29,799
Speaker 2: and he.

711
00:41:29,840 --> 00:41:32,240
Speaker 4: Was in Parks and Rack. He's the really like wound

712
00:41:32,360 --> 00:41:35,680
Speaker 4: up guy from Eagleton that ends up working in the

713
00:41:35,760 --> 00:41:37,560
Speaker 4: office when they combine.

714
00:41:37,840 --> 00:41:40,359
Speaker 1: I'm so glad you guys are here. Like, I recognize

715
00:41:40,560 --> 00:41:43,000
Speaker 1: maybe three out of five Billy Aker.

716
00:41:43,000 --> 00:41:44,680
Speaker 4: He's the guys that cost people on the street with

717
00:41:44,760 --> 00:41:47,560
Speaker 4: a microphone and just yells celebrity names at them and stuff.

718
00:41:47,880 --> 00:41:52,120
Speaker 1: I thought that was Will Ferrell doing Harry Carey. Yeah,

719
00:41:52,280 --> 00:41:55,399
Speaker 1: sort of, okay, but I'm sorry, I'm derailing this. Okay.

720
00:41:55,719 --> 00:41:57,960
Speaker 2: The point is there's a there's a human being that

721
00:41:58,080 --> 00:41:59,880
Speaker 2: you know that is famous for one reason, and of

722
00:42:00,560 --> 00:42:04,880
Speaker 2: possibly powerful and their voice is being manipulated. But the

723
00:42:05,000 --> 00:42:09,239
Speaker 2: video itself isn't something that you can reference to like

724
00:42:09,320 --> 00:42:12,080
Speaker 2: a speech or to a movie or something that you

725
00:42:12,200 --> 00:42:15,360
Speaker 2: remember you can verify with. It looks like a brand

726
00:42:15,400 --> 00:42:18,760
Speaker 2: new video, but you can still tell that it's that person.

727
00:42:18,880 --> 00:42:21,160
Speaker 2: Their mouth is moving and the words coming out are

728
00:42:21,360 --> 00:42:22,640
Speaker 2: something awful.

729
00:42:23,200 --> 00:42:26,520
Speaker 1: I feel like we're putting this in an accurate, humorous way.

730
00:42:27,080 --> 00:42:29,839
Speaker 1: We do have to emphasize this is a very scary thing. Matt.

731
00:42:29,840 --> 00:42:32,400
Speaker 1: You've built a beautiful example that leads us to another

732
00:42:32,880 --> 00:42:38,600
Speaker 1: nefarious use of deep fakes, which will happen. It absolutely

733
00:42:38,640 --> 00:42:41,760
Speaker 1: will happen, false accusations.

734
00:42:42,080 --> 00:42:44,040
Speaker 2: Ah, Yes, exactly so.

735
00:42:44,280 --> 00:42:48,160
Speaker 1: According to Andrea Hickerson, who is the director of the

736
00:42:48,200 --> 00:42:51,320
Speaker 1: School of Journalism and Mass Communications at the University of

737
00:42:51,400 --> 00:42:54,719
Speaker 1: South Carolina, this is a problem because at the most

738
00:42:54,760 --> 00:42:58,440
Speaker 1: basic level, deep fakes are lies disguised to look like truth.

739
00:42:59,320 --> 00:43:01,520
Speaker 1: In the Hickerson says, if we take them as truth

740
00:43:01,680 --> 00:43:05,279
Speaker 1: or evidence, we can easily make false conclusions with potentially

741
00:43:05,560 --> 00:43:12,120
Speaker 1: disastrous consequences. So if you want to ruin someone's life,

742
00:43:12,280 --> 00:43:15,840
Speaker 1: you want to smear a political opponent, or you just

743
00:43:16,840 --> 00:43:20,480
Speaker 1: don't really like your neighbor, you don't like their their vibe.

744
00:43:20,840 --> 00:43:24,640
Speaker 1: It just irks you. Then you could with this, with

745
00:43:24,800 --> 00:43:27,480
Speaker 1: this capacity in the future, you would be able to

746
00:43:28,400 --> 00:43:32,000
Speaker 1: make something where it appears that they're saying something terrible,

747
00:43:32,239 --> 00:43:35,480
Speaker 1: where they're like, yeah, I don't kick and puppies, burning buildings,

748
00:43:35,880 --> 00:43:38,800
Speaker 1: just whatever to feel something, you know what I mean. Also,

749
00:43:39,160 --> 00:43:42,160
Speaker 1: I Le, I rent scooters and I leave them in

750
00:43:42,200 --> 00:43:44,640
Speaker 1: the middle of the sidewalk because I'm that guy. I'm

751
00:43:44,680 --> 00:43:48,440
Speaker 1: the one, and then it would look real. But if

752
00:43:48,480 --> 00:43:51,360
Speaker 1: we go back to orwell, this becomes even more dangerous

753
00:43:51,480 --> 00:43:56,080
Speaker 1: because we have to consider activism and heavy handed state actors.

754
00:43:56,600 --> 00:44:00,000
Speaker 1: It's already dangerous to be an activist. To Knowle's example

755
00:44:00,120 --> 00:44:03,800
Speaker 1: about Hong Kong. Right, people are in danger, people are dying,

756
00:44:04,040 --> 00:44:10,480
Speaker 1: They're fighting an authoritarian massive state. So currently, state corporating

757
00:44:10,520 --> 00:44:14,719
Speaker 1: criminal actors seeking to silence dissidents all use the ordinary

758
00:44:14,960 --> 00:44:18,319
Speaker 1: tool kit of suppression, all the time tested stuff, all

759
00:44:18,360 --> 00:44:24,080
Speaker 1: the smooth jazz, all the hits, threats, violence, kidnapping, smear campaigns, incarceration,

760
00:44:24,320 --> 00:44:29,719
Speaker 1: disappearing and of course assassination the breakout single of suppression. Right,

761
00:44:30,600 --> 00:44:33,640
Speaker 1: But soon even as we record this, state actors will

762
00:44:33,640 --> 00:44:37,000
Speaker 1: have a new and powerful tool. If, for instance, you

763
00:44:37,160 --> 00:44:40,640
Speaker 1: are protesting something and you are the leader, the face

764
00:44:41,000 --> 00:44:43,120
Speaker 1: of activism, and say Hong Kong or one of the

765
00:44:43,239 --> 00:44:45,880
Speaker 1: many other places around the world where protests are active,

766
00:44:46,600 --> 00:44:48,799
Speaker 1: the authorities or the opposition would be able to make

767
00:44:48,920 --> 00:44:52,640
Speaker 1: videos in which you are on camera disagreeing with the

768
00:44:52,680 --> 00:44:56,440
Speaker 1: status quo, saying, hey, they were right all along. I

769
00:44:56,560 --> 00:44:58,680
Speaker 1: had a change of heart, and I want to confess

770
00:44:59,040 --> 00:45:02,279
Speaker 1: that my motives we're not pure. I was paid by

771
00:45:02,480 --> 00:45:06,200
Speaker 1: someone else to do this, so I apologize. I'm turning

772
00:45:06,280 --> 00:45:10,160
Speaker 1: myself in. And then the actual you finds out about

773
00:45:10,200 --> 00:45:13,360
Speaker 1: this when your confession is aired on the news and

774
00:45:13,520 --> 00:45:14,960
Speaker 1: people start contacting you.

775
00:45:15,640 --> 00:45:20,960
Speaker 2: Yeah, that is an intense hypothetical currently situation.

776
00:45:21,400 --> 00:45:22,240
Speaker 1: It's gonna happen.

777
00:45:22,480 --> 00:45:26,120
Speaker 2: I know, I know right now. It's just it is possible.

778
00:45:26,719 --> 00:45:28,560
Speaker 4: Well, that leads us to a lot of things that

779
00:45:28,960 --> 00:45:31,600
Speaker 4: need to happen or potentially are going to happen. And

780
00:45:31,600 --> 00:45:34,360
Speaker 4: I want to lead with this. It's really interesting in

781
00:45:35,040 --> 00:45:38,239
Speaker 4: the law now, and lawyers out there correct me if

782
00:45:38,239 --> 00:45:42,560
Speaker 4: I'm oversimplifying this, But video evidence isn't like the end

783
00:45:42,640 --> 00:45:45,520
Speaker 4: all be all already right, it has to fit a

784
00:45:45,640 --> 00:45:48,840
Speaker 4: couple of requirements in order to be admissible in the

785
00:45:48,880 --> 00:45:52,200
Speaker 4: first place, which is relevance and authenticity.

786
00:45:52,640 --> 00:45:54,560
Speaker 2: Yeah, with that chain of custody.

787
00:45:54,680 --> 00:45:56,080
Speaker 4: Chain of custody is a lot of things that go

788
00:45:56,160 --> 00:45:57,440
Speaker 4: into that, and we're going to get into that in

789
00:45:57,520 --> 00:45:59,840
Speaker 4: a minute, and also potentially what might have to have

790
00:45:59,840 --> 00:46:03,240
Speaker 4: happen as this technology gets better and better. But video

791
00:46:03,320 --> 00:46:07,480
Speaker 4: evidence can be considered hearsay if there isn't someone to

792
00:46:07,640 --> 00:46:11,520
Speaker 4: corroborate it. If it's the only evidence, that's not a

793
00:46:11,600 --> 00:46:14,759
Speaker 4: great case that you know, an eyewitness. You know, I

794
00:46:14,840 --> 00:46:17,040
Speaker 4: was just saying the good crime shows an eyeball witness

795
00:46:17,400 --> 00:46:20,000
Speaker 4: is really your best bet to getting a conviction. Video

796
00:46:20,120 --> 00:46:22,319
Speaker 4: and video alone, if that's all you got, especially if

797
00:46:22,320 --> 00:46:25,120
Speaker 4: it's like grainy security cam footage, a lawyer could say

798
00:46:25,520 --> 00:46:30,319
Speaker 4: that is not a reasonable representation of the subject being

799
00:46:30,400 --> 00:46:33,080
Speaker 4: displayed or in question, et cetera. But what this comes

800
00:46:33,120 --> 00:46:35,440
Speaker 4: down to is like, when is the log going to

801
00:46:35,520 --> 00:46:38,200
Speaker 4: catch up to this new technology in terms of that

802
00:46:38,360 --> 00:46:41,319
Speaker 4: chain of custody, Because that's the thing. If we can't

803
00:46:41,440 --> 00:46:43,239
Speaker 4: believe our eyes, like we said at the top of

804
00:46:43,280 --> 00:46:46,280
Speaker 4: the show, you said, Ben seeing isn't really believing.

805
00:46:46,280 --> 00:46:48,600
Speaker 2: How do we authenticate this stuff?

806
00:46:49,000 --> 00:46:51,920
Speaker 4: It's going to have to be that chain of custody

807
00:46:51,960 --> 00:46:54,360
Speaker 4: that's kept under lock and key, so we know that

808
00:46:54,480 --> 00:46:59,080
Speaker 4: the footage we're seeing was captured and then disseminated with

809
00:46:59,320 --> 00:47:00,440
Speaker 4: no between.

810
00:47:00,800 --> 00:47:03,440
Speaker 1: Right, Just video is nine enough. That's why Bigfoot will

811
00:47:03,520 --> 00:47:06,960
Speaker 1: never see a day in jail ever. And he needs

812
00:47:07,000 --> 00:47:07,200
Speaker 1: to be.

813
00:47:07,280 --> 00:47:10,239
Speaker 2: There, he really does. But you know, here's the other thing.

814
00:47:10,280 --> 00:47:12,879
Speaker 2: We can't trust our memories either. We're talking about in court.

815
00:47:13,320 --> 00:47:16,920
Speaker 2: Video without corroboration from a witness doesn't work, and the

816
00:47:17,000 --> 00:47:20,480
Speaker 2: witness is the best way, the eyewitness. But the eyewitness

817
00:47:21,080 --> 00:47:24,360
Speaker 2: is very unreliable. It's perhaps the most unreliable thing that

818
00:47:24,520 --> 00:47:27,520
Speaker 2: exists as far as evidence goes. So what the heck

819
00:47:27,680 --> 00:47:29,880
Speaker 2: do we trust when it comes to evidence of something

820
00:47:29,960 --> 00:47:30,960
Speaker 2: that actually happened.

821
00:47:31,120 --> 00:47:35,279
Speaker 1: The approach is at what of accretion of aggregation? Right,

822
00:47:35,320 --> 00:47:37,279
Speaker 1: you have to have multiple like nol said, you got

823
00:47:37,360 --> 00:47:39,839
Speaker 1: to have multiple things so you can say you can

824
00:47:39,920 --> 00:47:43,640
Speaker 1: triangulate a little bit, say, yeah, memories imperfect. We can't

825
00:47:43,680 --> 00:47:45,640
Speaker 1: trust video alone. But we have a video, and we

826
00:47:45,760 --> 00:47:49,319
Speaker 1: have a witness, and then maybe we have some other

827
00:47:49,440 --> 00:47:53,840
Speaker 1: forensic evidence like a gun case, scenes of you know,

828
00:47:54,040 --> 00:47:57,399
Speaker 1: spent shells were found and they match the gun.

829
00:47:57,640 --> 00:48:01,320
Speaker 2: I completely agree, and you're right. What worries me is

830
00:48:01,440 --> 00:48:05,280
Speaker 2: that what is stopping let's say, world leaders and powerful

831
00:48:05,320 --> 00:48:09,520
Speaker 2: people who do wrong things where a video is an

832
00:48:09,680 --> 00:48:13,239
Speaker 2: actual video is taken, maybe a surveillance cam video is

833
00:48:13,360 --> 00:48:17,440
Speaker 2: taken of some wrongdoing occurring. Let's says as high as

834
00:48:17,480 --> 00:48:21,400
Speaker 2: an assassination is something as dire as that occurs on camera,

835
00:48:21,520 --> 00:48:24,440
Speaker 2: but there are no witnesses. But for sure, on camera

836
00:48:24,680 --> 00:48:27,640
Speaker 2: there is a world leader shooting somebody in the head.

837
00:48:30,200 --> 00:48:32,680
Speaker 2: What is stopping them from saying, oh, that's a deep

838
00:48:32,719 --> 00:48:33,239
Speaker 2: fake video?

839
00:48:33,360 --> 00:48:35,840
Speaker 1: Obviously, I'm so glad you said that, because that's the

840
00:48:35,920 --> 00:48:38,279
Speaker 1: other side of the coin, right, If you are caught

841
00:48:38,360 --> 00:48:42,360
Speaker 1: on video doing something despicable, doing something illegal, whatever, you

842
00:48:42,520 --> 00:48:45,680
Speaker 1: can just say I have enemies, this was a deep fake.

843
00:48:46,040 --> 00:48:51,239
Speaker 1: You know, my ex hates me, or I am I

844
00:48:51,360 --> 00:48:54,320
Speaker 1: am doing work for I don't know whatever, a union

845
00:48:54,560 --> 00:48:58,600
Speaker 1: or something or an NGO. So there's not a very

846
00:48:58,719 --> 00:49:02,120
Speaker 1: good way to refute that other than having to resort

847
00:49:02,200 --> 00:49:06,360
Speaker 1: to as much other corroborating evidence as you can. And

848
00:49:06,480 --> 00:49:08,439
Speaker 1: then again you have to rely on, like you said,

849
00:49:08,440 --> 00:49:11,839
Speaker 1: those eyeball witnesses, whose memories at times can be very

850
00:49:11,880 --> 00:49:13,960
Speaker 1: financially motivated. Wow.

851
00:49:14,520 --> 00:49:17,279
Speaker 2: So so in the end, we're relying on law, which

852
00:49:17,320 --> 00:49:19,040
Speaker 2: is such an old concept, right.

853
00:49:19,200 --> 00:49:21,720
Speaker 1: I think this is I think this is where we're going, Nolan.

854
00:49:21,800 --> 00:49:24,360
Speaker 1: I know this is something we've talked about on this

855
00:49:24,480 --> 00:49:29,400
Speaker 1: show so often because it's so important. Technology will always

856
00:49:29,440 --> 00:49:33,200
Speaker 1: outpace legislation. If we finally get around to making a

857
00:49:33,280 --> 00:49:37,120
Speaker 1: law about something, the horse has already left the barn,

858
00:49:38,040 --> 00:49:40,759
Speaker 1: the you know, the detainee has already pulled the black

859
00:49:40,800 --> 00:49:43,720
Speaker 1: hood off and as well on their way to international waters.

860
00:49:44,760 --> 00:49:48,200
Speaker 1: In the summer of twenty nineteen, the US House Representatives

861
00:49:48,239 --> 00:49:51,400
Speaker 1: Intelligence Committee sent a letter. They didn't pass anything. They

862
00:49:51,440 --> 00:49:55,160
Speaker 1: sent a letter to the big socials Twitter, Facebook, Google,

863
00:49:55,600 --> 00:50:00,080
Speaker 1: asking them how these sites planned to fight again and

864
00:50:00,200 --> 00:50:05,719
Speaker 1: steep fakes, particularly in the twenty twenty election. And this

865
00:50:05,920 --> 00:50:09,640
Speaker 1: all came about because remember we mentioned that deep fake

866
00:50:09,760 --> 00:50:12,919
Speaker 1: video of Nancy Pelosi, House speaker here in the US.

867
00:50:14,160 --> 00:50:18,560
Speaker 1: The current president, President Donald Trump, didn't just co sign

868
00:50:18,640 --> 00:50:20,360
Speaker 1: that deep fake video. He retweeted it.

869
00:50:20,400 --> 00:50:22,279
Speaker 4: Wait a minute, when you say he wasn't just like saying, hey,

870
00:50:22,400 --> 00:50:26,640
Speaker 4: watch out for this fake video of my friend Nancy Pelosi. No, no, no,

871
00:50:26,760 --> 00:50:29,880
Speaker 4: he put it out there like he was implying that

872
00:50:30,000 --> 00:50:32,520
Speaker 4: it was real. Yeah, he's a savage on Twitter too,

873
00:50:32,640 --> 00:50:33,680
Speaker 4: as anybody knows.

874
00:50:34,280 --> 00:50:35,879
Speaker 1: On Twitter. As a matter of fact, if you look

875
00:50:35,880 --> 00:50:38,640
Speaker 1: at his record, he disagrees with himself extensively.

876
00:50:38,880 --> 00:50:42,680
Speaker 2: Well, it's just it's one of those things where it

877
00:50:43,360 --> 00:50:48,520
Speaker 2: really shows how convincing these things can be. I think

878
00:50:48,640 --> 00:50:51,320
Speaker 2: back to I think we've mentioned it on this show before,

879
00:50:51,440 --> 00:50:54,840
Speaker 2: but the Jordan Peel deep fake video of President Obama.

880
00:50:54,960 --> 00:51:00,359
Speaker 2: Oh yeah, yeah, yeah, where it looks like President BarackObama

881
00:51:00,880 --> 00:51:03,680
Speaker 2: sitting at a chair in the Oval office somewhere and

882
00:51:03,840 --> 00:51:07,680
Speaker 2: he's just saying weird things and it sounds pretty close

883
00:51:07,719 --> 00:51:11,680
Speaker 2: to him, but it's actually just Jordan Peele doing a

884
00:51:11,800 --> 00:51:14,560
Speaker 2: voice like a voiceover in the character that he's played

885
00:51:14,640 --> 00:51:19,839
Speaker 2: before on Key and Peel, and it is it's pretty disturbing,

886
00:51:20,120 --> 00:51:24,880
Speaker 2: and I think it speaks to how well Jordan's impression is,

887
00:51:25,200 --> 00:51:27,520
Speaker 2: or how good his impression is, but it also speaks

888
00:51:27,560 --> 00:51:31,480
Speaker 2: to the ability, this ability of matching that face, turning

889
00:51:31,520 --> 00:51:35,239
Speaker 2: it into or or changing the way the president's mouth

890
00:51:35,360 --> 00:51:38,360
Speaker 2: is moving to match the words. And if you imagine

891
00:51:38,600 --> 00:51:41,560
Speaker 2: the technology that is coming out right now where you

892
00:51:41,680 --> 00:51:46,799
Speaker 2: can take five seconds of any voice recording and then

893
00:51:47,000 --> 00:51:51,279
Speaker 2: you can recreate that voice saying anything you want it to. Again,

894
00:51:51,360 --> 00:51:53,680
Speaker 2: this is a small research project coming out of a

895
00:51:53,760 --> 00:51:58,920
Speaker 2: couple universities, but you could if you combined that voice

896
00:51:59,160 --> 00:52:03,759
Speaker 2: changing technology with the face changing and manipulating technology, you

897
00:52:04,400 --> 00:52:08,600
Speaker 2: get to a point where it will truly be we

898
00:52:08,719 --> 00:52:10,280
Speaker 2: will be unable to tell.

899
00:52:11,320 --> 00:52:14,640
Speaker 1: That's right, and we know that therefore there's a ticking

900
00:52:14,760 --> 00:52:22,480
Speaker 1: time bomb on this, right. So the presidential retweet of

901
00:52:22,600 --> 00:52:25,600
Speaker 1: that deep fake video is what inspired the House representatives

902
00:52:25,680 --> 00:52:29,439
Speaker 1: to send that letter. But this followed a request from

903
00:52:29,560 --> 00:52:32,960
Speaker 1: Congress that occurred earlier this year in January, where they

904
00:52:33,000 --> 00:52:35,800
Speaker 1: asked the d and I Director of National Intelligence to

905
00:52:36,040 --> 00:52:39,840
Speaker 1: give a formal report on deep fake technology. It was basically,

906
00:52:40,600 --> 00:52:43,200
Speaker 1: explain it to us so that we sound like we

907
00:52:43,280 --> 00:52:46,120
Speaker 1: know what's going on with our constituents. And also, of

908
00:52:46,239 --> 00:52:49,600
Speaker 1: course there's more than a little self preservation involved because

909
00:52:49,640 --> 00:52:53,960
Speaker 1: these are members of Congress, right, so we know that

910
00:52:54,080 --> 00:52:58,000
Speaker 1: we have to have legislation involved. But we also know

911
00:52:58,239 --> 00:53:00,360
Speaker 1: there's a big chance that it's just not going to

912
00:53:00,400 --> 00:53:03,320
Speaker 1: be enough. It's too easy to do this, it's too convenient,

913
00:53:03,360 --> 00:53:05,040
Speaker 1: it's too powerful, right.

914
00:53:07,280 --> 00:53:11,480
Speaker 4: Really quick in our industry, as podcasters. There's a there's

915
00:53:11,480 --> 00:53:13,840
Speaker 4: an audio equivalent of this that's kind of on the

916
00:53:14,360 --> 00:53:15,680
Speaker 4: on the horizon.

917
00:53:15,440 --> 00:53:18,200
Speaker 1: Frank and Biting's next evolution very much so.

918
00:53:19,280 --> 00:53:21,920
Speaker 4: I found out about this through a third party. I

919
00:53:22,000 --> 00:53:24,480
Speaker 4: can't name names. I don't think it's technology that's really

920
00:53:24,600 --> 00:53:28,439
Speaker 4: on the market yet. But literally, an algorithm that could

921
00:53:28,719 --> 00:53:33,759
Speaker 4: sweep through our catalog. Me you, Ben and Matt run

922
00:53:33,800 --> 00:53:37,080
Speaker 4: an algorithm on our catalog, and then you could feed

923
00:53:37,200 --> 00:53:41,120
Speaker 4: it lines and it'll it'll approximate our voices. And I've

924
00:53:41,160 --> 00:53:44,160
Speaker 4: heard it in action and it does a very convincing job.

925
00:53:44,440 --> 00:53:46,000
Speaker 4: And on the one hand, we could say, oh cool,

926
00:53:46,040 --> 00:53:47,960
Speaker 4: we don't have to read ads anymore. But on the

927
00:53:48,040 --> 00:53:51,160
Speaker 4: other hand, we could say, oh no, we don't have

928
00:53:51,320 --> 00:53:54,640
Speaker 4: jobs anymore, we don't need podcasting. I'm just saying, like

929
00:53:55,080 --> 00:53:59,520
Speaker 4: the the the implications of stuff like this, it's always

930
00:53:59,640 --> 00:54:01,799
Speaker 4: more far reaching than you would originally think.

931
00:54:01,960 --> 00:54:04,560
Speaker 2: Well, you know, that's why Dan is ending Harmontown is

932
00:54:04,600 --> 00:54:07,520
Speaker 2: because he's just gonna take all the voices and bottom

933
00:54:07,680 --> 00:54:10,759
Speaker 2: together and make a whole new podcast that he can

934
00:54:10,880 --> 00:54:11,239
Speaker 2: just write.

935
00:54:11,560 --> 00:54:13,480
Speaker 1: Did I tell you I watched the last episode?

936
00:54:14,400 --> 00:54:16,759
Speaker 2: I yeah, it was No, you didn't tell me, But

937
00:54:17,320 --> 00:54:18,200
Speaker 2: I'm glad that you.

938
00:54:18,239 --> 00:54:22,680
Speaker 1: Did touch it too. That was cool. Yeah, So I

939
00:54:22,760 --> 00:54:25,040
Speaker 1: mean end of an era for sure. But then now

940
00:54:25,120 --> 00:54:27,400
Speaker 1: we have the scary proposition, and I think i'm I

941
00:54:27,480 --> 00:54:30,040
Speaker 1: think I'm familiar with what you're talking about, NOL. We

942
00:54:30,160 --> 00:54:33,120
Speaker 1: have the scary proposition where if someone has enough audio

943
00:54:34,239 --> 00:54:38,560
Speaker 1: footage to pull from, then Harmontown would never have to end.

944
00:54:38,680 --> 00:54:41,040
Speaker 1: It would just get really weird because the technology you

945
00:54:41,040 --> 00:54:42,080
Speaker 1: would still need to evolve.

946
00:54:42,160 --> 00:54:44,279
Speaker 2: I'm telling you that's what he's doing, guys, That's what

947
00:54:44,360 --> 00:54:46,600
Speaker 2: I'm saying. Like he's gonna make Rob Shop say whatever

948
00:54:46,640 --> 00:54:48,040
Speaker 2: he wants. It's gonna be amazing.

949
00:54:49,719 --> 00:54:51,719
Speaker 1: Well, I'll tune in. I'll tune in. If you guys

950
00:54:51,760 --> 00:54:54,520
Speaker 1: are listening, let us know we would. We would love

951
00:54:54,600 --> 00:54:57,239
Speaker 1: to hear it, and we applaud you for pioneering into

952
00:54:57,280 --> 00:55:02,840
Speaker 1: this brave, new strange world. Government institutions like our favorite

953
00:55:02,880 --> 00:55:06,319
Speaker 1: math scientists DARPA, and researchers at colleges like you had

954
00:55:06,360 --> 00:55:10,239
Speaker 1: mentioned Matt Carnegie, Mellon, University of Washington, Stanford, so on,

955
00:55:10,920 --> 00:55:15,520
Speaker 1: are also experimenting with deep fake technology. They're experimenting in

956
00:55:15,600 --> 00:55:19,960
Speaker 1: two different but very related paths. One they want to

957
00:55:20,000 --> 00:55:21,839
Speaker 1: figure out how to use it, how to make it better,

958
00:55:22,120 --> 00:55:24,239
Speaker 1: and two while doing that, they want to figure out

959
00:55:24,239 --> 00:55:27,000
Speaker 1: how to stop it. So think these goals are kind

960
00:55:27,040 --> 00:55:30,239
Speaker 1: of contradictory unless they build in some kind of equivalent

961
00:55:30,360 --> 00:55:34,640
Speaker 1: of a governor switch, you know, like for anyone unfamiliar

962
00:55:34,680 --> 00:55:39,719
Speaker 1: with automobiles, some engines have a specific switch in there

963
00:55:39,800 --> 00:55:42,560
Speaker 1: that limits the performance.

964
00:55:41,960 --> 00:55:45,800
Speaker 4: Of the engine governor, right, Like in long distance trucks especially,

965
00:55:45,920 --> 00:55:47,680
Speaker 4: there's a company. I only know this because I knew

966
00:55:47,719 --> 00:55:49,719
Speaker 4: a guy that that was a long haul truck driver.

967
00:55:50,120 --> 00:55:53,480
Speaker 4: And there's a company called Swift that everyone jokes in

968
00:55:53,560 --> 00:55:55,799
Speaker 4: the industry stands for sure, wish I had a fast

969
00:55:55,880 --> 00:55:59,319
Speaker 4: truck because they are notorious for putting governors on there

970
00:55:59,520 --> 00:56:02,320
Speaker 4: that I guess it's a calculation you make as a

971
00:56:02,400 --> 00:56:04,719
Speaker 4: business in terms of the risk, you know, versus like

972
00:56:04,760 --> 00:56:07,239
Speaker 4: how fast can we get there versus how likely our

973
00:56:07,360 --> 00:56:09,399
Speaker 4: drivers are going to drive too fast and potentially put

974
00:56:09,640 --> 00:56:12,400
Speaker 4: themselves at risk and others, and you know, open up

975
00:56:12,400 --> 00:56:15,080
Speaker 4: for liability and lawsuits. But yeah, that's absolutely a thing.

976
00:56:15,360 --> 00:56:19,400
Speaker 1: So Bluebird buses would be another great example, and they

977
00:56:19,440 --> 00:56:21,000
Speaker 1: have their own very interesting story.

978
00:56:21,080 --> 00:56:25,160
Speaker 4: Bird scooters have digitally triggered governors, where like we have

979
00:56:25,200 --> 00:56:27,480
Speaker 4: an area in town called the belt line where it

980
00:56:27,640 --> 00:56:30,239
Speaker 4: GEO targets that area, and when you ride them on

981
00:56:30,320 --> 00:56:32,239
Speaker 4: our belt line, which is like almost like the high

982
00:56:32,280 --> 00:56:33,960
Speaker 4: Line in New York, it's like a walking trail, but

983
00:56:34,040 --> 00:56:35,920
Speaker 4: you can ride the scooters. It does not allow you

984
00:56:36,040 --> 00:56:38,120
Speaker 4: to go past a certain speed that you could go

985
00:56:38,280 --> 00:56:39,640
Speaker 4: past elsewhere in the city.

986
00:56:39,960 --> 00:56:42,480
Speaker 2: Man, I remember I had one on my ninety five

987
00:56:42,880 --> 00:56:45,919
Speaker 2: Dodge Caravan. Talked it out right around.

988
00:56:47,760 --> 00:56:52,239
Speaker 1: So we're talking about the technological analog. That's why I'm

989
00:56:52,239 --> 00:56:56,279
Speaker 1: bringing up governors switches because it seems like researching the

990
00:56:56,400 --> 00:56:59,640
Speaker 1: improvement of the given technology while also researching a way

991
00:56:59,719 --> 00:57:03,400
Speaker 1: to stemy that technology means that you would ultimately build

992
00:57:03,480 --> 00:57:06,680
Speaker 1: in something like that, and ideally you would want it

993
00:57:06,760 --> 00:57:09,680
Speaker 1: to be proprietary such that you would control it. So

994
00:57:09,800 --> 00:57:14,440
Speaker 1: in a way, these institutions are competing. Who will control

995
00:57:14,880 --> 00:57:17,520
Speaker 1: the nature of the truth and reality. That's it. That's

996
00:57:17,560 --> 00:57:19,760
Speaker 1: a big question, but it's it's there.

997
00:57:19,920 --> 00:57:22,440
Speaker 2: I think I think we need some kind of tech

998
00:57:22,720 --> 00:57:26,600
Speaker 2: ethics board or you know, or advisory committee like techics,

999
00:57:26,720 --> 00:57:30,120
Speaker 2: like like ethics, yeah, ethics, yeah, ethics. We should do that.

1000
00:57:30,240 --> 00:57:32,760
Speaker 2: We should get somebody to come through and like create

1001
00:57:33,120 --> 00:57:34,360
Speaker 2: something that everybody has to sign.

1002
00:57:34,560 --> 00:57:37,560
Speaker 4: Doesn't that have to be self imposed by like these

1003
00:57:37,640 --> 00:57:39,640
Speaker 4: tech companies, you know what I mean?

1004
00:57:39,800 --> 00:57:41,480
Speaker 2: Like, I don't know, if you get like a Gavin

1005
00:57:41,480 --> 00:57:43,960
Speaker 2: Belson or somebody like that, you could probably get everybody

1006
00:57:44,000 --> 00:57:45,960
Speaker 2: else on board, Like you know, a big name.

1007
00:57:46,200 --> 00:57:49,000
Speaker 1: You'd have to, uh, you have to ruin some people

1008
00:57:49,880 --> 00:57:52,320
Speaker 1: for an example. You'd have to ruin some people in

1009
00:57:52,360 --> 00:57:53,960
Speaker 1: the beginning for an example.

1010
00:57:53,720 --> 00:57:56,160
Speaker 2: Probably, But but I think you could. You could push

1011
00:57:56,200 --> 00:57:56,520
Speaker 2: it through.

1012
00:57:56,760 --> 00:57:59,240
Speaker 4: You just name dropped a fictional character. I just want

1013
00:57:59,280 --> 00:58:01,480
Speaker 4: to point that out to Wait what yeah, just huh,

1014
00:58:01,720 --> 00:58:04,760
Speaker 4: just putting that out there. Who any Silicon Valley fans

1015
00:58:04,800 --> 00:58:06,960
Speaker 4: out there trying to I'm trying to put something in

1016
00:58:07,040 --> 00:58:08,920
Speaker 4: there that you literally made me do a double take.

1017
00:58:09,000 --> 00:58:11,320
Speaker 4: Wait a minute, like is he who is the How

1018
00:58:11,360 --> 00:58:12,240
Speaker 4: come I haven't heard of him?

1019
00:58:12,240 --> 00:58:12,360
Speaker 1: Oh?

1020
00:58:12,440 --> 00:58:13,120
Speaker 4: Yeah, fiction.

1021
00:58:13,240 --> 00:58:14,320
Speaker 2: I'm trying to easter egg it.

1022
00:58:14,520 --> 00:58:16,680
Speaker 4: Sorry, dude, I shouldn't have said anything.

1023
00:58:16,720 --> 00:58:19,440
Speaker 2: He played along, but you played along perfectly. Oh thanks, man,

1024
00:58:19,560 --> 00:58:20,200
Speaker 2: that was beautiful.

1025
00:58:20,200 --> 00:58:23,280
Speaker 4: I really screwed it all. Sorry, Well, what what's next?

1026
00:58:23,880 --> 00:58:27,720
Speaker 1: Really? I mean? I think I think right now, regardless,

1027
00:58:27,920 --> 00:58:31,600
Speaker 1: there are people who are pro deep fake technology. Uh,

1028
00:58:31,680 --> 00:58:34,760
Speaker 1: there's there's a very convincing argument, all right, I think

1029
00:58:34,760 --> 00:58:37,960
Speaker 1: a very exciting argument that this can fundamentally change what

1030
00:58:38,080 --> 00:58:42,120
Speaker 1: we think of as film as entertainment because imagine you

1031
00:58:42,240 --> 00:58:45,840
Speaker 1: have the perfect role for an actor who has passed

1032
00:58:46,680 --> 00:58:49,920
Speaker 1: and you want them to be in your film. With

1033
00:58:50,080 --> 00:58:54,600
Speaker 1: this kind of technology, you can do it plausibly. And

1034
00:58:54,720 --> 00:58:58,680
Speaker 1: that means that, coupled also with machine learning writing of fiction,

1035
00:58:59,200 --> 00:59:02,480
Speaker 1: we could arrive I've had a time in our individual

1036
00:59:02,560 --> 00:59:08,440
Speaker 1: or collective lives where a film is made without human

1037
00:59:08,560 --> 00:59:12,920
Speaker 1: involvement on the creative ends. How insane is that? Forward

1038
00:59:12,960 --> 00:59:14,960
Speaker 1: to the future, I say, no, reduce.

1039
00:59:15,200 --> 00:59:17,320
Speaker 2: We know that other researchers are attempting to do that,

1040
00:59:17,640 --> 00:59:21,920
Speaker 2: like combining the different versions and types of neural networks

1041
00:59:21,960 --> 00:59:26,320
Speaker 2: to write a screenplay, to do some voice over at work,

1042
00:59:26,800 --> 00:59:32,240
Speaker 2: you know, to actually shoot video and edit video. It's

1043
00:59:32,280 --> 00:59:34,440
Speaker 2: all happening. It's just a matter of time. I think

1044
00:59:34,680 --> 00:59:36,080
Speaker 2: you're right on the money there, Ben, So.

1045
00:59:36,120 --> 00:59:38,760
Speaker 1: Where does this leave us for twenty twenty and beyond

1046
00:59:39,760 --> 00:59:44,040
Speaker 1: at a where to dilemma? Because it's really a matter

1047
00:59:44,080 --> 00:59:48,880
Speaker 1: of free expression versus true deception. According to Sharon Bradford Franklin,

1048
00:59:49,080 --> 00:59:53,240
Speaker 1: as a policy director for the Open Technology Institute out

1049
00:59:53,240 --> 00:59:56,400
Speaker 1: of New America. Deep fake videos threat in our civic

1050
00:59:56,480 --> 00:59:59,760
Speaker 1: discourse and can cause serious reputational and psychic harmed in

1051
01:00:00,040 --> 01:00:03,960
Speaker 1: vials right. They also make it more challenging for platforms

1052
01:00:04,000 --> 01:00:07,880
Speaker 1: to engage in responsible moderation of content, which is already

1053
01:00:07,960 --> 01:00:10,240
Speaker 1: a huge problem for anyone who's been paying attention to

1054
01:00:10,280 --> 01:00:15,480
Speaker 1: the latest news about Facebook. While the public is understandably

1055
01:00:15,600 --> 01:00:19,120
Speaker 1: calling for social media companies to develop techniques to detect

1056
01:00:19,200 --> 01:00:22,520
Speaker 1: and prevent the spread of deep fakes, we must also

1057
01:00:22,800 --> 01:00:26,400
Speaker 1: avoid establishing legal rules that push too far in the

1058
01:00:26,480 --> 01:00:30,720
Speaker 1: opposite direction and pressure platforms to engage in censorship of

1059
01:00:30,840 --> 01:00:34,640
Speaker 1: free expression online. So, to take your earlier argument, which

1060
01:00:34,640 --> 01:00:37,080
Speaker 1: again Matt, I love where someone says, that's not me,

1061
01:00:37,280 --> 01:00:39,840
Speaker 1: this is a deep fake. What happens if you were

1062
01:00:39,920 --> 01:00:43,800
Speaker 1: posting something, let's say you're an activist or you've a

1063
01:00:43,800 --> 01:00:47,080
Speaker 1: whistle blower or a whistleblower yes even better, and you

1064
01:00:47,440 --> 01:00:51,920
Speaker 1: propagate this film that is indisputable proof of the shenanigans

1065
01:00:51,960 --> 01:00:54,440
Speaker 1: you said we're occurring all along, and then the people

1066
01:00:54,520 --> 01:00:57,960
Speaker 1: who have their hands on the switches, their fingers on

1067
01:00:58,040 --> 01:01:00,320
Speaker 1: the faucet, they just say, oh, that's a deep fake.

1068
01:01:00,440 --> 01:01:02,440
Speaker 1: And they turn it off. We were always at war

1069
01:01:02,560 --> 01:01:04,080
Speaker 1: with a seysha. You know what I mean.

1070
01:01:08,120 --> 01:01:10,640
Speaker 2: It's harroring when you put that at the endpen We

1071
01:01:10,760 --> 01:01:12,200
Speaker 2: were always with these things.

1072
01:01:12,120 --> 01:01:13,880
Speaker 1: Right, I mean, but where do you where do you

1073
01:01:13,960 --> 01:01:17,120
Speaker 1: think this is going? I would say one person's opinion

1074
01:01:17,480 --> 01:01:21,120
Speaker 1: that it's it's going to continue. It will not disappear.

1075
01:01:21,360 --> 01:01:24,320
Speaker 4: I will say, if you want to see a really

1076
01:01:24,640 --> 01:01:29,600
Speaker 4: really jarringly good example of what this can look like,

1077
01:01:30,280 --> 01:01:31,480
Speaker 4: it's a video of.

1078
01:01:33,280 --> 01:01:33,840
Speaker 2: Bill Hayter.

1079
01:01:34,680 --> 01:01:37,680
Speaker 4: On is it the Letterman Show? No, yeah, it was

1080
01:01:37,720 --> 01:01:40,680
Speaker 4: an older a show, an old clip. It was an

1081
01:01:40,720 --> 01:01:42,640
Speaker 4: old clip. But it's a Bill Hayter. I believe it's

1082
01:01:42,680 --> 01:01:45,280
Speaker 4: The Letterman Show. It might be Conan and he's telling

1083
01:01:45,360 --> 01:01:50,040
Speaker 4: the story about meeting Tom Cruise at a party of

1084
01:01:50,080 --> 01:01:52,240
Speaker 4: some kind of it had to do with No, he

1085
01:01:52,360 --> 01:01:53,640
Speaker 4: was not what it was. He was in that movie

1086
01:01:53,680 --> 01:01:55,600
Speaker 4: Tropic Thunder with with Tom Cruise, and he sees he

1087
01:01:55,640 --> 01:01:59,000
Speaker 4: meets Tom Cruise at the premiere and Bill Hayter this

1088
01:01:59,040 --> 01:02:01,680
Speaker 4: point isn't like huge. He's an SNL. He's known for

1089
01:02:01,760 --> 01:02:04,240
Speaker 4: doing impressions, et cetera. So of course when he's telling

1090
01:02:04,400 --> 01:02:07,080
Speaker 4: the story, he's doing the Tom Cruise impression when he's

1091
01:02:07,120 --> 01:02:08,960
Speaker 4: doing Tom Cruise's parts of the story, and he's a

1092
01:02:09,000 --> 01:02:10,960
Speaker 4: great impression, and he's a great impressions and in this

1093
01:02:11,120 --> 01:02:13,760
Speaker 4: deep in his video, every time he starts to lapse

1094
01:02:13,800 --> 01:02:17,000
Speaker 4: into the Tom Cruise voice, his face turns into Tom

1095
01:02:17,080 --> 01:02:20,800
Speaker 4: Cruise and it is it is jarringly good and it's

1096
01:02:21,080 --> 01:02:23,880
Speaker 4: almos it's borderline like it makes your brain kind of

1097
01:02:24,040 --> 01:02:25,400
Speaker 4: like spasm.

1098
01:02:25,000 --> 01:02:26,720
Speaker 2: A little bit because it's just so good.

1099
01:02:27,080 --> 01:02:30,600
Speaker 1: He also does a great impression of Arnold Schwarzenegger and

1100
01:02:30,600 --> 01:02:33,280
Speaker 1: al Pacino, I want to say, and you can see

1101
01:02:33,320 --> 01:02:36,040
Speaker 1: the same technology of play. The Tom Cruise one, I

1102
01:02:36,080 --> 01:02:38,480
Speaker 1: would say is the best example because their faces are

1103
01:02:38,520 --> 01:02:40,160
Speaker 1: already a little more similar this.

1104
01:02:40,240 --> 01:02:41,960
Speaker 4: One, even though it goes a little further where he

1105
01:02:42,360 --> 01:02:45,200
Speaker 4: starts to do a Seth Rogan impression later and then

1106
01:02:45,280 --> 01:02:47,760
Speaker 4: his face becomes Seth Rogen. But the way it does

1107
01:02:47,840 --> 01:02:50,640
Speaker 4: it morphs where it just like for a split second

1108
01:02:50,800 --> 01:02:53,280
Speaker 4: it'll be Seth Rogen and then it's back to Bill Hayter.

1109
01:02:53,680 --> 01:02:57,360
Speaker 4: But the Tom Cruise parts are shockingly good. It doesn't

1110
01:02:57,400 --> 01:03:00,240
Speaker 4: look like mapping, it doesn't look like projection mapping, or

1111
01:03:00,480 --> 01:03:02,920
Speaker 4: you know, it really very much is like he becomes

1112
01:03:03,000 --> 01:03:06,360
Speaker 4: him and everything seems you can tell. It's like pulling

1113
01:03:06,520 --> 01:03:09,360
Speaker 4: from something that Tom Cruise did where he was acting,

1114
01:03:09,520 --> 01:03:12,480
Speaker 4: you know. Like, but like you said, Ben, this video

1115
01:03:12,520 --> 01:03:15,600
Speaker 4: that we're seeing doesn't really exist in the wild. It's

1116
01:03:15,720 --> 01:03:18,600
Speaker 4: like a you know, composite of like all of this

1117
01:03:18,680 --> 01:03:20,760
Speaker 4: stuff that's out there, right. Are you seeing it, Matt, Oh, yeah,

1118
01:03:20,760 --> 01:03:21,160
Speaker 4: I've seen it.

1119
01:03:21,200 --> 01:03:23,640
Speaker 2: And there's the there's one that's called the Deep Fake

1120
01:03:23,680 --> 01:03:27,400
Speaker 2: Impressionist or Deep There's there's another impersonator.

1121
01:03:27,680 --> 01:03:30,440
Speaker 4: It's a Instagram account that I follow as well that

1122
01:03:30,680 --> 01:03:31,640
Speaker 4: it does a lot of these.

1123
01:03:31,600 --> 01:03:34,080
Speaker 2: Oh okay, yeah, all I know is this one guy

1124
01:03:34,120 --> 01:03:36,160
Speaker 2: that I've seen several videos of. I couldn't even tell

1125
01:03:36,200 --> 01:03:38,680
Speaker 2: you his name right now, but he got together with somebody.

1126
01:03:38,760 --> 01:03:41,760
Speaker 2: They made an entire video of non stop versions of

1127
01:03:41,840 --> 01:03:45,480
Speaker 2: this and it really is convincing crazy.

1128
01:03:45,800 --> 01:03:48,880
Speaker 1: And this is just the beginning. People will look back

1129
01:03:49,040 --> 01:03:56,680
Speaker 1: on this era as the halcyon days of discernible fakes,

1130
01:03:57,400 --> 01:04:00,160
Speaker 1: at least if this continues. Yeah, just the time of

1131
01:04:00,280 --> 01:04:02,800
Speaker 1: us knowing what was real and what was not right.

1132
01:04:03,680 --> 01:04:06,240
Speaker 1: Where do you see this technology going? Folks who want

1133
01:04:06,280 --> 01:04:08,440
Speaker 1: to pass the torch to you, let us know and

1134
01:04:08,680 --> 01:04:11,760
Speaker 1: on the way, send us your favorite deep fakes. You

1135
01:04:11,840 --> 01:04:14,320
Speaker 1: can post them on our community page Here's where it

1136
01:04:14,320 --> 01:04:17,040
Speaker 1: gets crazy. On Facebook. You can find us on Instagram,

1137
01:04:17,160 --> 01:04:19,800
Speaker 1: you can find us on Twitter. You can also find

1138
01:04:19,880 --> 01:04:22,800
Speaker 1: every episode we have ever done on our website. Stuff

1139
01:04:22,800 --> 01:04:24,600
Speaker 1: they don't want you to know dot com And.

1140
01:04:24,640 --> 01:04:26,840
Speaker 2: We don't put this out there enough. But as the

1141
01:04:26,920 --> 01:04:30,400
Speaker 2: holidays are coming around, remember we've got a te public store.

1142
01:04:30,640 --> 01:04:33,240
Speaker 2: So if you want to get some some awesome stuff

1143
01:04:33,240 --> 01:04:35,880
Speaker 2: they don't want you to know, gifts, whatever it is,

1144
01:04:36,080 --> 01:04:40,920
Speaker 2: mousetads do it, and you know that's look candidly and upfront.

1145
01:04:41,040 --> 01:04:44,120
Speaker 2: We get a tiny, tiny, witty bitty percentage, but you

1146
01:04:44,240 --> 01:04:47,680
Speaker 2: are supporting our show bye by doing this. The designs

1147
01:04:47,720 --> 01:04:51,080
Speaker 2: are cool, they're really cool. I genuinely wear my new

1148
01:04:51,480 --> 01:04:53,800
Speaker 2: red stuff they don't want you to know shirt all

1149
01:04:53,840 --> 01:04:54,160
Speaker 2: this time.

1150
01:04:54,240 --> 01:04:56,080
Speaker 1: Ben Ben's old school.

1151
01:04:56,440 --> 01:04:58,640
Speaker 4: Yeah, that's awes the text version.

1152
01:04:58,720 --> 01:04:59,960
Speaker 1: They got Superman move right.

1153
01:05:00,200 --> 01:05:02,600
Speaker 2: Look, we feel weird wearing our own things, so a

1154
01:05:02,640 --> 01:05:04,320
Speaker 2: lot of times we do what Ben's doing. You put

1155
01:05:04,400 --> 01:05:06,960
Speaker 2: something over it, but when you wear it, it's almost

1156
01:05:07,120 --> 01:05:09,920
Speaker 2: like for me like having the Superman's thing on. You know,

1157
01:05:10,040 --> 01:05:11,600
Speaker 2: I wear the stuff they don't want you to know.

1158
01:05:11,840 --> 01:05:13,680
Speaker 2: Signature Tidy whities.

1159
01:05:13,640 --> 01:05:16,480
Speaker 1: What I wear these stuff? They don't want you to know. Snuggie,

1160
01:05:16,880 --> 01:05:17,080
Speaker 1: do you.

1161
01:05:17,160 --> 01:05:19,240
Speaker 4: Actually have custom made yes?

1162
01:05:19,640 --> 01:05:20,400
Speaker 2: Oh my god.

1163
01:05:20,640 --> 01:05:22,360
Speaker 4: I don't think t public offers it and I had

1164
01:05:22,400 --> 01:05:23,080
Speaker 4: to outsource it.

1165
01:05:23,400 --> 01:05:27,360
Speaker 1: And more importantly, we have a good authority. I don't

1166
01:05:27,360 --> 01:05:29,000
Speaker 1: know if you guys get these texts too, but every

1167
01:05:29,040 --> 01:05:31,800
Speaker 1: time somebody buys something from the store, I get a

1168
01:05:31,920 --> 01:05:34,600
Speaker 1: text from Connell that just says one more day.

1169
01:05:36,960 --> 01:05:39,960
Speaker 2: Yeah, I'm not on that chain, but I'd like to

1170
01:05:40,040 --> 01:05:41,959
Speaker 2: get on there. That would be fun. Really, it would

1171
01:05:41,960 --> 01:05:47,120
Speaker 2: really just fuel my neurosis.

1172
01:05:51,240 --> 01:05:55,800
Speaker 1: And that's our classic episode for this evening. We can't

1173
01:05:55,800 --> 01:05:56,680
Speaker 1: wait to hear your thoughts.

1174
01:05:56,760 --> 01:05:57,840
Speaker 4: It's right let us know what you think.

1175
01:05:57,840 --> 01:05:59,720
Speaker 2: You can reach. You to the handle Conspiracy Stuff.

1176
01:06:00,040 --> 01:06:04,120
Speaker 4: We exist on Facebook X and YouTube on Instagram and

1177
01:06:04,120 --> 01:06:05,600
Speaker 4: TikTok or Conspiracy Stuff Show.

1178
01:06:05,680 --> 01:06:07,880
Speaker 2: If you want to call us dial one eight three

1179
01:06:08,040 --> 01:06:13,160
Speaker 2: three STDWYTK. That's our voicemail system. You've got three minutes.

1180
01:06:13,240 --> 01:06:15,400
Speaker 2: Give yourself a cool nickname and let us know if

1181
01:06:15,440 --> 01:06:17,480
Speaker 2: we can use your name and message on the air.

1182
01:06:17,840 --> 01:06:19,360
Speaker 2: If you got more to say than can fit in

1183
01:06:19,440 --> 01:06:22,200
Speaker 2: that voicemail, why not instead send us a good old

1184
01:06:22,240 --> 01:06:23,000
Speaker 2: fashioned email.

1185
01:06:23,280 --> 01:06:26,120
Speaker 1: We are the entities to read every single piece of

1186
01:06:26,200 --> 01:06:30,640
Speaker 1: correspondence we receive, be aware, yet not afraid. Sometimes the

1187
01:06:30,720 --> 01:06:34,160
Speaker 1: void writes back conspiracy at iHeartRadio dot com.

1188
01:06:53,440 --> 01:06:55,480
Speaker 2: Stuff they don't want you to know is a production

1189
01:06:55,640 --> 01:06:59,960
Speaker 2: of iHeartRadio. For more podcasts from iHeartRadio, visit the iHeartRadio app,

1190
01:07:00,240 --> 01:07:03,080
Speaker 2: Apple Podcasts, or wherever you listen to your favorite shows.