1
00:00:00,160 --> 00:00:05,120
Speaker 1: Folks, Tonight's classic episode is one that is near and

2
00:00:05,160 --> 00:00:08,080
Speaker 1: dear and frightening to all of us. It's a question,

3
00:00:08,360 --> 00:00:10,520
Speaker 1: will Google start impersonated you?

4
00:00:11,160 --> 00:00:11,480
Speaker 2: Yes?

5
00:00:12,600 --> 00:00:16,560
Speaker 3: Wow, Oh the days of July twenty eighteen, we were

6
00:00:16,680 --> 00:00:17,360
Speaker 3: so naive.

7
00:00:17,720 --> 00:00:21,080
Speaker 2: Oh the salad days of twenty eighteen. Well, we were so.

8
00:00:21,200 --> 00:00:26,920
Speaker 1: Young, right, We had no idea what was going to happen,

9
00:00:27,040 --> 00:00:31,240
Speaker 1: because even DARPA can't fully predict the future yet. But

10
00:00:31,360 --> 00:00:34,760
Speaker 1: you know, I remember going back to because I lived

11
00:00:35,600 --> 00:00:37,360
Speaker 1: really close to our old office for a while. I

12
00:00:37,360 --> 00:00:41,000
Speaker 1: remember going back in during the pandemic and there was

13
00:00:41,040 --> 00:00:44,440
Speaker 1: this scary moment where I saw somewhat one of our

14
00:00:44,479 --> 00:00:50,120
Speaker 1: coworker's calendars left on like March thirteenth, or like the

15
00:00:50,280 --> 00:00:51,720
Speaker 1: day everything shut down.

16
00:00:52,200 --> 00:00:55,080
Speaker 2: That's like zombie movie stuff. That's eerie. I think.

17
00:00:55,440 --> 00:00:58,880
Speaker 3: Was it on Robert's desk, I don't. I just remember

18
00:00:58,920 --> 00:01:01,160
Speaker 3: seeing that too, Ben like, oh wow.

19
00:01:01,680 --> 00:01:04,880
Speaker 1: It close sped. It was our buddy Seth, who yeah,

20
00:01:04,920 --> 00:01:07,360
Speaker 1: for a long time worked on stuff to blow your mind.

21
00:01:08,319 --> 00:01:12,040
Speaker 1: But yeah, it's the dystopian. It's the very last of us.

22
00:01:13,040 --> 00:01:14,759
Speaker 2: And you know, what we're talking about in this episode

23
00:01:14,840 --> 00:01:17,400
Speaker 2: is a tech that I don't know if it made

24
00:01:17,440 --> 00:01:19,760
Speaker 2: it all the way through to fruition. It certainly not

25
00:01:19,760 --> 00:01:21,639
Speaker 2: what we associate with this kind of stuff, But Google

26
00:01:22,040 --> 00:01:25,160
Speaker 2: was developing something called duplex that was a tech for

27
00:01:25,480 --> 00:01:29,560
Speaker 2: conducting natural conversations to carry out real world talks over

28
00:01:29,640 --> 00:01:31,800
Speaker 2: the phone. Now, of course the name of the game

29
00:01:31,840 --> 00:01:36,120
Speaker 2: for that is chat GPT and all of the many

30
00:01:36,360 --> 00:01:39,880
Speaker 2: worms that come in that bag, that worm bag, badger bag.

31
00:01:40,400 --> 00:01:42,160
Speaker 2: But this is kind of like almost like a peak,

32
00:01:42,319 --> 00:01:44,480
Speaker 2: a sneak peek into like maybe what could be a

33
00:01:44,520 --> 00:01:46,360
Speaker 2: precursor to that, right, guys.

34
00:01:46,480 --> 00:01:52,800
Speaker 1: One hundred percent yeah, Because this this idea of automating

35
00:01:53,080 --> 00:01:58,560
Speaker 1: ones day to day routine interactions. It feels cool. But

36
00:01:58,800 --> 00:02:02,200
Speaker 1: I am certain that in episodes like this one of

37
00:02:02,280 --> 00:02:07,480
Speaker 1: us will will all inevitably reference that scary scene with

38
00:02:07,840 --> 00:02:11,040
Speaker 1: Mickey as a wizard in Fantasia. You know what I mean.

39
00:02:11,240 --> 00:02:13,919
Speaker 1: You get a mop and at first it's easy, it's

40
00:02:13,919 --> 00:02:14,560
Speaker 1: a lot of fun.

41
00:02:14,680 --> 00:02:16,400
Speaker 2: Just brooms were psychotic.

42
00:02:17,600 --> 00:02:20,400
Speaker 3: So come with us back to those happy days of

43
00:02:20,440 --> 00:02:23,760
Speaker 3: twenty eighteen, and let's discuss what AI might become.

44
00:02:24,280 --> 00:02:26,080
Speaker 2: Knowing what it has become.

45
00:02:26,200 --> 00:02:30,680
Speaker 1: From UFOs to psychic powers and government conspiracies, history is

46
00:02:30,760 --> 00:02:35,080
Speaker 1: riddled with unexplained events. You can turn back now or

47
00:02:35,160 --> 00:02:37,120
Speaker 1: learn this stuff they don't want you to know.

48
00:02:50,360 --> 00:02:54,320
Speaker 2: Hello, and welcome back to the show. I'm Noel standing

49
00:02:54,320 --> 00:02:57,200
Speaker 2: in Formatt with that line, because whenever Matt's not here,

50
00:02:57,919 --> 00:02:59,240
Speaker 2: Ben and I always look at each other for a

51
00:02:59,280 --> 00:03:01,800
Speaker 2: few minutes and realize that we're not quite sure how

52
00:03:01,800 --> 00:03:02,560
Speaker 2: to start the show.

53
00:03:03,800 --> 00:03:06,919
Speaker 1: However, Never fear Met is off on a very special

54
00:03:06,960 --> 00:03:09,799
Speaker 1: secret project that we cannot wait to tell you about.

55
00:03:10,080 --> 00:03:12,799
Speaker 1: We will have to tell you sometime soon, so look

56
00:03:12,840 --> 00:03:15,720
Speaker 1: forward to that. No spoilers. In the meantime, they call

57
00:03:15,800 --> 00:03:19,280
Speaker 1: me Ben. We are joined with our super producer Paul

58
00:03:19,639 --> 00:03:24,760
Speaker 1: the Personal Digital Assistant decade. Yeah, PDA, but most importantly,

59
00:03:24,960 --> 00:03:28,000
Speaker 1: you are you. You are here, and that makes this

60
00:03:28,400 --> 00:03:31,680
Speaker 1: stuff they don't want you to know. We're pretty excited

61
00:03:31,720 --> 00:03:32,720
Speaker 1: about today's episode.

62
00:03:32,760 --> 00:03:34,560
Speaker 2: What did PDA stand for when it was like a

63
00:03:34,600 --> 00:03:37,440
Speaker 2: pom pilot? Was it personal digital assistant?

64
00:03:37,760 --> 00:03:39,320
Speaker 3: Uh?

65
00:03:39,400 --> 00:03:42,320
Speaker 1: Yes, I believe or was it Okay, so there was

66
00:03:42,360 --> 00:03:43,560
Speaker 1: public display of affections?

67
00:03:43,560 --> 00:03:45,640
Speaker 2: Well there's that, But they used to call like BlackBerrys

68
00:03:45,720 --> 00:03:49,680
Speaker 2: PDAs back before the advent of the iPhone and the

69
00:03:49,720 --> 00:03:52,240
Speaker 2: more tablet like device.

70
00:03:52,080 --> 00:03:54,480
Speaker 1: Personal digital assistant. That's what it has to be.

71
00:03:54,560 --> 00:03:56,760
Speaker 2: Isn't that funny though, because that's kind of become a

72
00:03:56,800 --> 00:03:59,320
Speaker 2: new thing, because today we're talking about the kinds of

73
00:03:59,360 --> 00:04:03,200
Speaker 2: personal didit assistance that you can talk to and can

74
00:04:03,240 --> 00:04:04,840
Speaker 2: potentially talk back.

75
00:04:05,280 --> 00:04:07,920
Speaker 1: Yes, So let's look at some of the context. It's

76
00:04:07,960 --> 00:04:13,240
Speaker 1: a very common trope in science fiction robots impersonating human

77
00:04:13,280 --> 00:04:16,599
Speaker 1: beings with increasing levels of fidelity, and we see it

78
00:04:16,600 --> 00:04:18,839
Speaker 1: in pop culture all the time, and some stories, like

79
00:04:18,920 --> 00:04:22,760
Speaker 1: in every episode of The Terminator everything related to the

80
00:04:22,839 --> 00:04:28,560
Speaker 1: Terminator franchise, Machine Consciousness tries to mimic humanity exclusively as

81
00:04:28,560 --> 00:04:31,479
Speaker 1: a means of waging war, and in other places or

82
00:04:31,520 --> 00:04:35,119
Speaker 1: other series such as The Matrix, Artificial Intelligence or AI

83
00:04:35,400 --> 00:04:39,840
Speaker 1: attempts to surround us meat bags with an impersonation of reality,

84
00:04:40,120 --> 00:04:44,280
Speaker 1: complete with individual machine minds that can pass for humans.

85
00:04:44,640 --> 00:04:47,799
Speaker 1: And then in other cases, humans work together to build

86
00:04:47,800 --> 00:04:51,320
Speaker 1: technologies that can impersonate other human beings in any number

87
00:04:51,360 --> 00:04:51,680
Speaker 1: of ways.

88
00:04:51,760 --> 00:04:55,880
Speaker 2: Yeah, sometimes for sexy times reasons, sure, sometimes for just companionship,

89
00:04:56,200 --> 00:04:59,080
Speaker 2: and we've seen that go awry. I mean the Terminator.

90
00:04:59,120 --> 00:05:01,000
Speaker 2: I think the whole point of that series was that

91
00:05:01,279 --> 00:05:05,040
Speaker 2: humans created Skynet or whatever for their own to serve

92
00:05:05,080 --> 00:05:08,880
Speaker 2: their own purposes. And then Skynet said, what up, humans,

93
00:05:08,920 --> 00:05:10,440
Speaker 2: We're done with you or going to do our own thing,

94
00:05:10,640 --> 00:05:12,200
Speaker 2: or like in West World m hm.

95
00:05:12,800 --> 00:05:15,559
Speaker 1: And this goes, like you said, in any number of ways,

96
00:05:15,560 --> 00:05:20,440
Speaker 1: from increasingly lifelike androids to entities that exist purely in

97
00:05:20,480 --> 00:05:25,680
Speaker 1: the digital sphere, able to hold genuine, seeming conversations and

98
00:05:25,720 --> 00:05:29,200
Speaker 1: functioning at or above what we would consider the average

99
00:05:29,279 --> 00:05:33,919
Speaker 1: level of human intelligence. As history has proven, science fiction

100
00:05:34,120 --> 00:05:37,680
Speaker 1: is often prescient, and it's not uncommon for authors to

101
00:05:37,760 --> 00:05:43,200
Speaker 1: spin fantastic tales, only for those tales to move years

102
00:05:43,480 --> 00:05:46,240
Speaker 1: decades later from the realm of science fiction to the

103
00:05:46,240 --> 00:05:49,479
Speaker 1: world of science fact. And the quest for this human

104
00:05:49,640 --> 00:05:53,000
Speaker 1: like technology in real life is no different.

105
00:05:53,320 --> 00:05:55,680
Speaker 2: Yeah, I mean, I actually saw a YouTube video a

106
00:05:55,760 --> 00:05:58,640
Speaker 2: bunch of clips of times that science fiction films have

107
00:05:58,720 --> 00:06:02,080
Speaker 2: predicted technology that it totally a thing right now. Like

108
00:06:02,800 --> 00:06:06,440
Speaker 2: Star Trek, for example, there's little communicators, they're basically iPhones.

109
00:06:06,680 --> 00:06:10,040
Speaker 2: And remember that scene in Total Recall where they're walking

110
00:06:10,080 --> 00:06:12,360
Speaker 2: through a full body scanner at the airport and you

111
00:06:12,400 --> 00:06:15,080
Speaker 2: see like their skeletons and you can see the weapons

112
00:06:15,160 --> 00:06:17,039
Speaker 2: hiding and stuff. That's pretty much what we do now

113
00:06:17,080 --> 00:06:18,760
Speaker 2: at the airport. We got to put our hands over

114
00:06:18,800 --> 00:06:21,800
Speaker 2: our heads and stand in those full body scanners where

115
00:06:22,040 --> 00:06:24,000
Speaker 2: hopefully they just draw a little cartoon of us with

116
00:06:24,000 --> 00:06:26,760
Speaker 2: the naughty bits scrubbed out. But in theory they can

117
00:06:26,800 --> 00:06:30,839
Speaker 2: see everything. So too is AI and the kinds of

118
00:06:30,880 --> 00:06:32,000
Speaker 2: things we're talking about today.

119
00:06:32,279 --> 00:06:35,880
Speaker 1: Yeah, and the concept of what we call artificial intelligence.

120
00:06:36,240 --> 00:06:40,400
Speaker 1: Longtime listeners, you may recall, as we use this phrase,

121
00:06:40,400 --> 00:06:43,240
Speaker 1: you may recall that some former guests of ours have

122
00:06:43,320 --> 00:06:47,040
Speaker 1: objected to the term artificial intelligence with a very good question,

123
00:06:47,440 --> 00:06:49,880
Speaker 1: what makes it artificial? What makes it if we're talking

124
00:06:49,880 --> 00:06:53,120
Speaker 1: about consciousness, what makes it any less of a consciousness

125
00:06:53,240 --> 00:06:54,119
Speaker 1: than our own?

126
00:06:54,320 --> 00:06:54,520
Speaker 2: Right?

127
00:06:54,960 --> 00:06:57,120
Speaker 1: And we think that's a good We think that's a

128
00:06:57,200 --> 00:07:01,960
Speaker 1: very good and valid point. We tend to agree with it.

129
00:07:02,000 --> 00:07:04,440
Speaker 1: But for the sake of brevity, we're just going to

130
00:07:04,520 --> 00:07:07,880
Speaker 1: go with AI as a non pejorative thing. It's just

131
00:07:08,000 --> 00:07:10,559
Speaker 1: easier to say it that way. The concept of AI

132
00:07:10,960 --> 00:07:13,960
Speaker 1: has surprisingly old roots in our culture, especially if we

133
00:07:14,040 --> 00:07:19,960
Speaker 1: consider those ancient tales of non human entities impersonating human beings,

134
00:07:20,080 --> 00:07:23,320
Speaker 1: the fairy stories of changelings switched out at birth, or

135
00:07:23,920 --> 00:07:28,120
Speaker 1: God's changing shape to breed with animals or people, or

136
00:07:28,200 --> 00:07:32,640
Speaker 1: shape shifters. In the twentieth century, the concept of this artificial,

137
00:07:32,840 --> 00:07:37,520
Speaker 1: inorganic thinking life form was popularized, just like the point

138
00:07:37,560 --> 00:07:41,000
Speaker 1: we made with science fiction through culture and fiction. If

139
00:07:41,040 --> 00:07:44,800
Speaker 1: you think of one of the earliest artificial intelligences that

140
00:07:45,360 --> 00:07:48,119
Speaker 1: blew up in the Western world, it's the tin man

141
00:07:48,280 --> 00:07:49,200
Speaker 1: in The Wizard of Oz.

142
00:07:49,280 --> 00:07:49,440
Speaker 3: Yeah.

143
00:07:49,480 --> 00:07:50,920
Speaker 2: I never thought of him like that, but I guess

144
00:07:50,960 --> 00:07:52,720
Speaker 2: it's true. It's a good stand in for that because

145
00:07:52,720 --> 00:07:54,480
Speaker 2: even the heart that he gets at the end spoiler

146
00:07:54,520 --> 00:07:56,720
Speaker 2: alert for it, the Wizard of Oz sure is like

147
00:07:56,760 --> 00:07:59,360
Speaker 2: a clockwork hert, you know, it's like something to add

148
00:07:59,440 --> 00:08:02,520
Speaker 2: to his mechanations. It's not actually a physical heart. So

149
00:08:02,560 --> 00:08:05,480
Speaker 2: he's absolutely meant to be like an automaton, which is

150
00:08:06,040 --> 00:08:08,400
Speaker 2: kind of the earliest form of robotics that go back

151
00:08:08,480 --> 00:08:12,040
Speaker 2: to ancient times, even where you have these incredibly intricate

152
00:08:12,800 --> 00:08:16,880
Speaker 2: creations that move through a series of gears and pulleys

153
00:08:16,920 --> 00:08:19,360
Speaker 2: and what have you, but aren't necessarily imbued with any

154
00:08:19,440 --> 00:08:21,320
Speaker 2: kind of like ability to make choices. But there are

155
00:08:21,360 --> 00:08:23,200
Speaker 2: some that can even like play games. I think there's

156
00:08:23,200 --> 00:08:24,920
Speaker 2: one that's like a writer.

157
00:08:25,120 --> 00:08:29,160
Speaker 1: The mechanical Turk Well yeah, yeah, yeah, And they were

158
00:08:30,040 --> 00:08:34,600
Speaker 1: lauded for their ability to appear to do human esque

159
00:08:34,679 --> 00:08:38,240
Speaker 1: things right, but people generally did not think they had

160
00:08:38,280 --> 00:08:39,520
Speaker 1: a soul for sustenance.

161
00:08:39,679 --> 00:08:41,319
Speaker 2: And it kind of goes back to you were mentioning

162
00:08:41,360 --> 00:08:43,839
Speaker 2: we'd had some folks say you had to take issue

163
00:08:43,840 --> 00:08:46,200
Speaker 2: with the term of artificial intelligence. We've also had some

164
00:08:46,240 --> 00:08:48,360
Speaker 2: folks right and taking issue with the term the idea

165
00:08:48,360 --> 00:08:52,559
Speaker 2: of machine learning, because you know, it is a matter

166
00:08:52,720 --> 00:08:55,360
Speaker 2: of we're still at a place where we have to

167
00:08:55,400 --> 00:08:58,800
Speaker 2: program machines to do what we want. We certainly have

168
00:08:58,800 --> 00:09:01,160
Speaker 2: have yet to fully experience, and it's this idea of

169
00:09:01,200 --> 00:09:04,920
Speaker 2: a singularity where the machine takes what we've imbued it

170
00:09:04,960 --> 00:09:08,800
Speaker 2: with and develops its ability to go outside of that

171
00:09:08,960 --> 00:09:11,160
Speaker 2: or like make decisions outside of the parameters that we've

172
00:09:11,440 --> 00:09:13,920
Speaker 2: read programmed. Very rarely have we seen that, and when

173
00:09:13,920 --> 00:09:16,120
Speaker 2: we do see it, it typically gets shut down.

174
00:09:16,640 --> 00:09:20,840
Speaker 1: It's yeah, metacognition, right, so thinking about thinking rather so.

175
00:09:20,920 --> 00:09:25,720
Speaker 1: By the nineteen fifties, scientists and mathematicians and philosophers were

176
00:09:25,840 --> 00:09:30,439
Speaker 1: familiar with this concept of quote unquote artificial intelligence, and

177
00:09:30,520 --> 00:09:34,520
Speaker 1: our species began to cognitively migrate from this world of

178
00:09:34,559 --> 00:09:38,760
Speaker 1: fantasy toward a world increasingly grounded in fact and we

179
00:09:39,320 --> 00:09:41,959
Speaker 1: can hit some of the high points of artificial intelligence here.

180
00:09:42,120 --> 00:09:46,360
Speaker 1: In nineteen fifty, the famous codebreaker Alan Turing made one

181
00:09:46,400 --> 00:09:49,319
Speaker 1: of the most significant early steps in real life AI

182
00:09:49,640 --> 00:09:53,040
Speaker 1: when he and some of his colleagues created what we

183
00:09:53,120 --> 00:09:57,120
Speaker 1: now commonly call the Touring Test. It's named after him,

184
00:09:57,160 --> 00:10:00,960
Speaker 1: of course. He wrote this paper called Computing Machine Intelligence

185
00:10:01,120 --> 00:10:04,760
Speaker 1: that laid the groundwork both for the means of constructing

186
00:10:04,760 --> 00:10:08,400
Speaker 1: AI and for the ways in which we could measure

187
00:10:08,720 --> 00:10:12,120
Speaker 1: the intelligence of that AI or our success building it,

188
00:10:12,160 --> 00:10:17,000
Speaker 1: which might kind of be two different things. And sadly, well,

189
00:10:17,120 --> 00:10:20,600
Speaker 1: amazingly and sadly this line of thought was ahead of

190
00:10:20,640 --> 00:10:24,560
Speaker 1: the curve. Turing could not get right to work building

191
00:10:24,559 --> 00:10:29,000
Speaker 1: these human like minds, or more specifically, he couldn't get

192
00:10:29,040 --> 00:10:32,679
Speaker 1: to work building minds that could fool humans into thinking

193
00:10:32,840 --> 00:10:36,120
Speaker 1: they were also human minds, because the technology at the

194
00:10:36,160 --> 00:10:38,880
Speaker 1: time had hard limits, like to the point you make

195
00:10:38,960 --> 00:10:43,400
Speaker 1: know about the mechanical turk. Up until nineteen forty nine

196
00:10:43,520 --> 00:10:47,839
Speaker 1: or so, computers couldn't really store commands we can say

197
00:10:47,880 --> 00:10:51,200
Speaker 1: commands are decisions here. They could only execute them. And

198
00:10:51,240 --> 00:10:53,920
Speaker 1: this meant that the computers we could build at that

199
00:10:54,040 --> 00:10:58,600
Speaker 1: time were unable to satisfy a key prerequisitive intelligence. They

200
00:10:58,600 --> 00:11:03,320
Speaker 1: couldn't remember past events, past information, and therefore they could

201
00:11:03,320 --> 00:11:07,359
Speaker 1: not use this memory to inform present commands or decisions.

202
00:11:07,960 --> 00:11:12,559
Speaker 1: And in this way, computers began at that tabula rasa state,

203
00:11:12,600 --> 00:11:16,320
Speaker 1: that blank slate state that so many mystic, spiritualist and

204
00:11:16,360 --> 00:11:23,280
Speaker 1: philosophers spend their lives attempting to attain. I think that's fascinating. Computers,

205
00:11:24,080 --> 00:11:28,840
Speaker 1: like people who practice hardcore meditation, existed mentally only in

206
00:11:28,880 --> 00:11:29,320
Speaker 1: the present.

207
00:11:29,920 --> 00:11:31,960
Speaker 2: Yeah, And we'll get back to that concept in a

208
00:11:31,960 --> 00:11:35,640
Speaker 2: little bit in terms of some of the newer computers

209
00:11:35,640 --> 00:11:38,680
Speaker 2: and how they are able to quote unquote figure things

210
00:11:38,760 --> 00:11:41,880
Speaker 2: out some better than others. But let's just go back

211
00:11:41,880 --> 00:11:45,880
Speaker 2: to the fifties for a second, when computers were insanely expensive.

212
00:11:45,880 --> 00:11:48,720
Speaker 2: I think this is no secret they ran. You would

213
00:11:48,840 --> 00:11:50,560
Speaker 2: least one, you couldn't even own it, and that wasn't

214
00:11:50,559 --> 00:11:53,000
Speaker 2: a thing for about two hundred K a month, which

215
00:11:53,080 --> 00:11:56,280
Speaker 2: I believe been in today's standards. That would be about

216
00:11:56,280 --> 00:11:59,480
Speaker 2: two million dollars for one month of computer use. Yeah,

217
00:12:00,200 --> 00:12:03,560
Speaker 2: eighteen dollars a lot on Netflix subscriptions, right, my friend,

218
00:12:04,559 --> 00:12:08,440
Speaker 2: And only, of course, the most prestigious universities and huge

219
00:12:08,640 --> 00:12:13,320
Speaker 2: technology corporations could even you know, afford the cost of entry.

220
00:12:14,120 --> 00:12:19,480
Speaker 2: So for Turing and his ILK to actually be able

221
00:12:19,480 --> 00:12:24,599
Speaker 2: to succeed in building something resembling anywhere close to artificial intelligence,

222
00:12:24,880 --> 00:12:27,800
Speaker 2: they would have to be part of some network of

223
00:12:28,000 --> 00:12:35,320
Speaker 2: high profile, very wealthy and influential funders of research. Yeah,

224
00:12:35,360 --> 00:12:37,959
Speaker 2: that would they would be able to receive just an

225
00:12:38,040 --> 00:12:40,520
Speaker 2: absolute boatload of money from to get this kind of

226
00:12:40,520 --> 00:12:41,120
Speaker 2: work done.

227
00:12:41,360 --> 00:12:43,920
Speaker 1: And imagine, you know, that's a hard sell. It seems really

228
00:12:43,960 --> 00:12:47,319
Speaker 1: interesting now, but back on what we know, Yeah, yeah,

229
00:12:47,360 --> 00:12:49,960
Speaker 1: but back then it was literally walking up to people

230
00:12:50,000 --> 00:12:52,280
Speaker 1: and saying, hey, we're good at math, and do you

231
00:12:52,320 --> 00:12:55,520
Speaker 1: remember the tin man from Wizard of Oz. We sort

232
00:12:55,520 --> 00:12:57,760
Speaker 1: of want to make that. Can we have all of

233
00:12:57,800 --> 00:13:00,280
Speaker 1: the money. So that's that's tough, but the soldiers on.

234
00:13:00,440 --> 00:13:03,760
Speaker 1: In nineteen fifty five, another groundbreaking event occurred. There was

235
00:13:03,800 --> 00:13:06,880
Speaker 1: the premiere of a program called The Logic Theorist. It

236
00:13:06,920 --> 00:13:10,280
Speaker 1: was supposed to mimic the problem solving skills of a

237
00:13:10,360 --> 00:13:14,280
Speaker 1: human being, and it was funded by one of our

238
00:13:14,320 --> 00:13:18,559
Speaker 1: favorite shady boogeymen, the Research and Development Corporation known today

239
00:13:18,559 --> 00:13:19,400
Speaker 1: as RAND.

240
00:13:19,320 --> 00:13:21,320
Speaker 2: A RAND corporation that sounds like something out of a

241
00:13:21,320 --> 00:13:23,600
Speaker 2: sci fi movie in and of itself, and it's still around.

242
00:13:23,920 --> 00:13:27,280
Speaker 2: I'm doing pretty interesting and secretive work because I believe

243
00:13:27,320 --> 00:13:31,760
Speaker 2: they ended up having a relationship with the US. Oh yeah, government,

244
00:13:31,800 --> 00:13:32,240
Speaker 2: big time.

245
00:13:33,120 --> 00:13:38,520
Speaker 1: You're absolutely right. This Logic Theorist is considered by many

246
00:13:38,559 --> 00:13:44,280
Speaker 1: people to be the first technically the first artificial intelligence program.

247
00:13:44,360 --> 00:13:48,079
Speaker 1: And there was a big conference in nineteen fifty six

248
00:13:48,160 --> 00:13:51,600
Speaker 1: hosted by John McCarthy and a guy named Marvin Minsky,

249
00:13:52,040 --> 00:13:55,839
Speaker 1: and in this conference they presented the Logic Theorist as

250
00:13:55,880 --> 00:13:59,559
Speaker 1: a proof of concept. The conference is called the Dartmouth

251
00:13:59,640 --> 00:14:04,400
Speaker 1: Summer Research Project on Artificial Intelligence. And you know, Noel,

252
00:14:04,400 --> 00:14:06,440
Speaker 1: I thought of you with this because I know how

253
00:14:06,720 --> 00:14:08,200
Speaker 1: how we both love acronyms.

254
00:14:08,200 --> 00:14:09,640
Speaker 2: Well, I feel like in this one, the D would

255
00:14:09,640 --> 00:14:11,319
Speaker 2: be silence. So I'm just going to call it SURPI.

256
00:14:11,600 --> 00:14:14,240
Speaker 1: That's great. That sounds better because the DED, you.

257
00:14:14,280 --> 00:14:18,840
Speaker 2: Know, you can't really do a DS sound.

258
00:14:20,280 --> 00:14:25,800
Speaker 1: So the conference itself fell short of the original very

259
00:14:25,840 --> 00:14:28,760
Speaker 1: ambitious aims of the organizers. They wanted to bring together

260
00:14:28,800 --> 00:14:32,160
Speaker 1: the world's best and brightest subject matter experts and by god,

261
00:14:32,280 --> 00:14:36,600
Speaker 1: make an artificial mind of a weekend, right right, They

262
00:14:36,600 --> 00:14:39,120
Speaker 1: said it took god seven days, let's uh, what do

263
00:14:39,120 --> 00:14:41,200
Speaker 1: you say? We could do it in four something like that.

264
00:14:42,080 --> 00:14:45,840
Speaker 1: But the problem was pretty much everybody disagreed on how

265
00:14:45,960 --> 00:14:49,080
Speaker 1: exactly you would make a human like intelligence. And at

266
00:14:49,080 --> 00:14:52,800
Speaker 1: this point they're still thinking in terms of artificial intelligence

267
00:14:53,000 --> 00:14:56,400
Speaker 1: being like humans, which is a huge assumption. But they

268
00:14:56,520 --> 00:14:59,520
Speaker 1: unanimously agreed for the very first time on a single

269
00:14:59,600 --> 00:15:04,360
Speaker 1: crucial point that it was possible to make AI, and

270
00:15:04,400 --> 00:15:07,560
Speaker 1: this set the stage for the next two decades of research.

271
00:15:08,120 --> 00:15:13,000
Speaker 2: So, from nineteen fifty seven to nineteen seventy four, artificial

272
00:15:13,040 --> 00:15:16,160
Speaker 2: intelligence interest in it or research and it really flourished.

273
00:15:16,160 --> 00:15:19,720
Speaker 2: Computers they could stick, you know, they were improving by

274
00:15:19,840 --> 00:15:23,000
Speaker 2: leaps and bounds. They could store more information, which is crucial,

275
00:15:23,240 --> 00:15:26,240
Speaker 2: and of course they became faster and cheaper and more accessible.

276
00:15:27,360 --> 00:15:30,600
Speaker 2: Machine learning algorithms also began to improve, and people got

277
00:15:30,600 --> 00:15:34,920
Speaker 2: better at knowing which algorithm to use for their particular purposes,

278
00:15:34,960 --> 00:15:37,280
Speaker 2: which was also important because it wasn't you know that

279
00:15:37,360 --> 00:15:41,240
Speaker 2: there was sort of an established language of algorithms that

280
00:15:41,240 --> 00:15:44,200
Speaker 2: people could pick and choose from to suit their particular problems.

281
00:15:44,760 --> 00:15:48,880
Speaker 2: Early successes like the General problem solver Ben tell me

282
00:15:48,920 --> 00:15:50,800
Speaker 2: a little bit more about the general problem solver.

283
00:15:51,000 --> 00:15:53,480
Speaker 1: It does what it says under ten. You could give

284
00:15:53,520 --> 00:15:56,760
Speaker 1: it a variety of problems, okay, and generally speaking it

285
00:15:56,760 --> 00:15:58,120
Speaker 1: would attempt to solve them.

286
00:15:58,080 --> 00:16:00,200
Speaker 2: Not necessarily going to give you the most creative solution,

287
00:16:00,320 --> 00:16:01,479
Speaker 2: but that Yeah.

288
00:16:01,240 --> 00:16:04,600
Speaker 1: And again that's that's like an early day's thing. So

289
00:16:04,680 --> 00:16:08,880
Speaker 1: at the time it was very impressive, but of course

290
00:16:09,040 --> 00:16:12,240
Speaker 1: it was just the beginning. And you know, you had

291
00:16:12,280 --> 00:16:18,120
Speaker 1: another example I think of AI application in that time.

292
00:16:18,280 --> 00:16:20,400
Speaker 2: Yeah, this is when Alexa was first invented.

293
00:16:21,080 --> 00:16:22,680
Speaker 1: Oh yes, yes, I'm kidding.

294
00:16:22,760 --> 00:16:27,120
Speaker 2: It's called Eliza. But this is like a spoken language interpreter,

295
00:16:27,240 --> 00:16:29,400
Speaker 2: which I don't know. I wonder if Alexa is a

296
00:16:29,640 --> 00:16:31,680
Speaker 2: is a nod to Eliza. What do you think, Ben,

297
00:16:32,120 --> 00:16:34,400
Speaker 2: that would be pretty cool. Only a couple letters off.

298
00:16:34,560 --> 00:16:38,400
Speaker 2: It was a spoken language interpreter that helped convince the

299
00:16:38,440 --> 00:16:41,440
Speaker 2: government to really, Okay, this is something that we're into

300
00:16:41,520 --> 00:16:44,360
Speaker 2: and this was really important for our story today.

301
00:16:44,440 --> 00:16:49,360
Speaker 1: Yeah, it convinced the government, specifically DARPA here in the

302
00:16:49,600 --> 00:16:55,560
Speaker 1: United States to start funding artificial intelligence DARPA. As you

303
00:16:55,640 --> 00:16:58,600
Speaker 1: know if you are a longtime listener of this show,

304
00:16:58,920 --> 00:17:02,920
Speaker 1: DARPA is the the resident mad Science Department of the

305
00:17:03,160 --> 00:17:08,440
Speaker 1: United States government and it stands for Defense Advanced Research

306
00:17:08,520 --> 00:17:12,480
Speaker 1: Projects Agency. They're the ones who do all the X

307
00:17:12,560 --> 00:17:16,439
Speaker 1: files level or fringe level stuff you hear about in

308
00:17:16,480 --> 00:17:19,720
Speaker 1: the news or often it's going to be. They're a

309
00:17:19,760 --> 00:17:22,680
Speaker 1: forefront of it. And then there's also stuff like Matt's

310
00:17:22,680 --> 00:17:25,720
Speaker 1: favorite public private partnership, skunk Works.

311
00:17:25,800 --> 00:17:28,239
Speaker 2: It's just a good name, I mean, if you can

312
00:17:28,320 --> 00:17:29,840
Speaker 2: like it for nothing other than that, But they do

313
00:17:29,960 --> 00:17:34,040
Speaker 2: all the secret Air Force projects and you know, spyplanes

314
00:17:34,160 --> 00:17:36,679
Speaker 2: and all kinds of stuff. It's kind of technology you

315
00:17:36,720 --> 00:17:40,320
Speaker 2: think about that is probably years ahead of anything we've

316
00:17:40,359 --> 00:17:42,119
Speaker 2: seen out there in the world today.

317
00:17:42,440 --> 00:17:48,000
Speaker 1: It's mistaken for UFOs, right right. This made people very gassed,

318
00:17:48,119 --> 00:17:52,600
Speaker 1: very optimistic, because that's another very human thing that we

319
00:17:52,640 --> 00:17:57,360
Speaker 1: haven't quite learned how to program optimism. In nineteen seventy,

320
00:17:57,480 --> 00:18:00,840
Speaker 1: Marvin Minsky, that guy who co hosted this conference, he

321
00:18:00,960 --> 00:18:03,440
Speaker 1: told Life magazine from three d eight years, we will

322
00:18:03,480 --> 00:18:07,399
Speaker 1: have a machine with the general intelligence of an average

323
00:18:07,480 --> 00:18:10,960
Speaker 1: human being. He was wrong, But to your point about

324
00:18:11,000 --> 00:18:14,159
Speaker 1: suppressed technology novel, who's wrong? So far as we know.

325
00:18:15,040 --> 00:18:17,480
Speaker 2: It's interesting though too not to get too far ahead

326
00:18:17,480 --> 00:18:20,520
Speaker 2: of ourselves. But with everything that's going on with big

327
00:18:20,840 --> 00:18:25,240
Speaker 2: public facing companies like Google and Amazon and Apple and stuff,

328
00:18:25,640 --> 00:18:28,320
Speaker 2: you start to get a sense that maybe the secret

329
00:18:28,400 --> 00:18:32,160
Speaker 2: government stuff isn't quite as far ahead as it once was, right.

330
00:18:32,240 --> 00:18:33,880
Speaker 2: I mean, that's just my opinion. I don't know.

331
00:18:34,920 --> 00:18:37,240
Speaker 1: I think it's a good opinion. It's something we want

332
00:18:37,240 --> 00:18:40,520
Speaker 1: to hear from you about, ladies and gentlemen, because it's

333
00:18:40,560 --> 00:18:46,280
Speaker 1: proven that in terms of physical hardware, material of war,

334
00:18:47,680 --> 00:18:51,159
Speaker 1: weapons and aviation and stuff, it is proven that the

335
00:18:51,280 --> 00:18:55,240
Speaker 1: US and most other governments want to keep that stuff

336
00:18:55,400 --> 00:18:56,040
Speaker 1: under wraps.

337
00:18:56,240 --> 00:18:58,760
Speaker 2: Absolutely. I guess what I'm saying is when you're looking

338
00:18:58,760 --> 00:19:00,879
Speaker 2: at a company that's trying to sell you something and

339
00:19:00,920 --> 00:19:04,840
Speaker 2: you see how far they advance with each update every year, right,

340
00:19:04,920 --> 00:19:06,679
Speaker 2: you kind of get a sense that maybe this is

341
00:19:06,720 --> 00:19:08,240
Speaker 2: about where they're actually at.

342
00:19:08,440 --> 00:19:11,160
Speaker 1: That's that's the point. So I bring up the aviation

343
00:19:11,280 --> 00:19:14,680
Speaker 1: stuff and the weapons of war stuff to contrast it here,

344
00:19:14,760 --> 00:19:20,960
Speaker 1: because what we're seeing also is that the governments of

345
00:19:20,960 --> 00:19:24,639
Speaker 1: the world. This is true, the governments of the world

346
00:19:24,880 --> 00:19:28,800
Speaker 1: usually can't pay the best and brightest as much as

347
00:19:29,040 --> 00:19:32,720
Speaker 1: the private entity skin, so they're grabbing some of the

348
00:19:32,760 --> 00:19:37,320
Speaker 1: best workers unless those people are severely ideological and then

349
00:19:38,160 --> 00:19:40,800
Speaker 1: they're making most of the progress which they would later

350
00:19:40,880 --> 00:19:43,040
Speaker 1: sell to a government. So I think maybe there is

351
00:19:43,040 --> 00:19:44,320
Speaker 1: a little more transparency.

352
00:19:44,480 --> 00:19:46,359
Speaker 2: I think so, which is a which is an interestingly

353
00:19:46,480 --> 00:19:47,080
Speaker 2: positive thing.

354
00:19:47,160 --> 00:19:48,480
Speaker 1: But let's let's get share.

355
00:19:48,600 --> 00:19:50,400
Speaker 2: Yes, keep, let's keep, let's keep humming right along.

356
00:19:50,640 --> 00:19:55,600
Speaker 1: So the quest for human like artificial intelligence soldiered on,

357
00:19:55,760 --> 00:19:59,200
Speaker 1: but it still had tremendous obstacles. Although computers could now

358
00:19:59,280 --> 00:20:02,520
Speaker 1: store information, they couldn't store enough of it, and they

359
00:20:02,560 --> 00:20:05,520
Speaker 1: could not process it at a fast enough pace. So

360
00:20:05,600 --> 00:20:11,000
Speaker 1: funding dwindled until the eighties. And that's when AI experienced

361
00:20:11,040 --> 00:20:14,400
Speaker 1: a renaissance, which we'll get to after a word from

362
00:20:14,440 --> 00:20:15,200
Speaker 1: our sponsor.

363
00:20:21,720 --> 00:20:26,880
Speaker 2: So in the nineteen eighties, new tools and methods gave AI,

364
00:20:27,200 --> 00:20:30,919
Speaker 2: the field of AI, that renaissance that Ben mentioned, that

365
00:20:31,040 --> 00:20:34,400
Speaker 2: kick in the digital pant and a guy named John

366
00:20:34,440 --> 00:20:39,560
Speaker 2: Hopfield and another fellow named David Rommelhart popularized this concept

367
00:20:39,600 --> 00:20:43,080
Speaker 2: of deep learning, which is a technique that allowed computers

368
00:20:43,080 --> 00:20:51,760
Speaker 2: to learn using experiences, using paying attention to surroundings. For example,

369
00:20:52,119 --> 00:20:55,240
Speaker 2: the idea of the fact that our phones are serving

370
00:20:55,320 --> 00:20:57,840
Speaker 2: us ads because we're talking about stuff, and it's taking

371
00:20:57,840 --> 00:21:00,960
Speaker 2: that information in and using it to do something and

372
00:21:01,040 --> 00:21:04,600
Speaker 2: learning our habits right, and this is a really important

373
00:21:04,640 --> 00:21:08,199
Speaker 2: part of things like these personal digital assistance that we

374
00:21:08,280 --> 00:21:09,760
Speaker 2: talked about at the top of.

375
00:21:09,720 --> 00:21:13,600
Speaker 1: The show, right m and on the other side of

376
00:21:13,640 --> 00:21:16,320
Speaker 1: the brain here too, or I guess. In a parallel approach,

377
00:21:16,840 --> 00:21:22,280
Speaker 1: a guy named Edward Figenbaum introduced expert systems that mimicked

378
00:21:22,520 --> 00:21:26,000
Speaker 1: the decision making process of a human expert. So what

379
00:21:26,119 --> 00:21:29,920
Speaker 1: happens is the program will ask an expert in the

380
00:21:29,960 --> 00:21:34,159
Speaker 1: field like it would ask super producer Paul or Richard

381
00:21:34,200 --> 00:21:37,760
Speaker 1: Feinman a question about production or physics, and then they

382
00:21:37,760 --> 00:21:40,440
Speaker 1: would say, how do you respond in this given situation?

383
00:21:40,840 --> 00:21:44,119
Speaker 1: And it could and it would take this for every

384
00:21:44,160 --> 00:21:47,160
Speaker 1: situation it could soak up, and then non experts could

385
00:21:47,200 --> 00:21:51,119
Speaker 1: later receive advice from that program about that information. It

386
00:21:51,160 --> 00:21:54,359
Speaker 1: sounds basic now. It sounds like how a search engine

387
00:21:54,480 --> 00:21:58,399
Speaker 1: can know, with increasingly accurate levels of fidelity, what question

388
00:21:58,520 --> 00:22:00,840
Speaker 1: you mean to ask when you ask question? Remember when

389
00:22:00,840 --> 00:22:03,680
Speaker 1: Google used to be super passive aggressive yeah, and say

390
00:22:04,400 --> 00:22:05,240
Speaker 1: did you mean?

391
00:22:05,960 --> 00:22:07,560
Speaker 2: Well, Now it just fills it in for you, yeah,

392
00:22:07,600 --> 00:22:10,359
Speaker 2: Google instantly. And the reason for that is that it

393
00:22:10,440 --> 00:22:13,960
Speaker 2: has access to every entry that anyone has ever put

394
00:22:13,960 --> 00:22:16,800
Speaker 2: into Google, and so it combines all that information and

395
00:22:16,880 --> 00:22:19,520
Speaker 2: makes the best guess as to what it thinks you're

396
00:22:19,560 --> 00:22:22,480
Speaker 2: probably searching for based on what everyone else that has

397
00:22:22,480 --> 00:22:24,880
Speaker 2: started searching for that kind of thing has also done,

398
00:22:24,880 --> 00:22:27,480
Speaker 2: which is that same deep learning stuff that is coming

399
00:22:27,520 --> 00:22:28,320
Speaker 2: into play.

400
00:22:28,160 --> 00:22:30,760
Speaker 1: Which makes predictive text so funny.

401
00:22:30,560 --> 00:22:34,160
Speaker 2: It can be. I guess it depends on the platform

402
00:22:34,160 --> 00:22:35,840
Speaker 2: you're on, and we'll give that a little bit too,

403
00:22:35,960 --> 00:22:39,040
Speaker 2: because sometimes it can be just really boneheaded. And haven't

404
00:22:39,080 --> 00:22:43,439
Speaker 2: you figured this out yet, Siri spoiler alert, No he hasn't.

405
00:22:44,040 --> 00:22:45,560
Speaker 2: But this was all really put to the test in

406
00:22:45,600 --> 00:22:48,960
Speaker 2: the nineties and two thousands when we really hit some

407
00:22:49,119 --> 00:22:53,959
Speaker 2: big landmark moments and high water mark moments in artificial intelligence,

408
00:22:54,040 --> 00:22:57,080
Speaker 2: like with Gary Kasprov, who I know that you are

409
00:22:57,119 --> 00:23:00,000
Speaker 2: fascinated with, although he is a problematic figure at times.

410
00:23:00,200 --> 00:23:06,280
Speaker 1: Yeah, yeah, raging anti semitism aside. Yeah, chess grandmasters. The

411
00:23:06,400 --> 00:23:13,800
Speaker 1: human ones at least right are immensely, even fractally compelling,

412
00:23:13,880 --> 00:23:17,399
Speaker 1: because there's always some other layer to their personality and

413
00:23:17,440 --> 00:23:23,000
Speaker 1: you have to wonder about the purported correlation between mental

414
00:23:23,000 --> 00:23:28,800
Speaker 1: instability and high thresholds of intelligence, because well, that's a

415
00:23:28,960 --> 00:23:31,840
Speaker 1: story for another day. We should do an episode. Yeah,

416
00:23:32,760 --> 00:23:34,560
Speaker 1: might get a little close to home for us, but

417
00:23:34,640 --> 00:23:35,280
Speaker 1: we can do it.

418
00:23:35,840 --> 00:23:35,959
Speaker 3: So.

419
00:23:36,320 --> 00:23:39,399
Speaker 1: Yeah, as you're seeing, he was defeated by Deep Blue,

420
00:23:39,440 --> 00:23:42,679
Speaker 1: built by IBM, a computer just built to play chess.

421
00:23:43,000 --> 00:23:46,240
Speaker 1: This was a John Henry moment for the human race.

422
00:23:46,560 --> 00:23:50,840
Speaker 1: And then a scientist named Cynthia Brazel created Kismet, a

423
00:23:50,920 --> 00:23:56,440
Speaker 1: program of recognizing and displaying emotion. Now, for all our

424
00:23:56,880 --> 00:24:03,680
Speaker 1: techno futurists philosophers in the crowd, does Kismet recognizing and

425
00:24:03,800 --> 00:24:09,119
Speaker 1: displaying emotion automatically mean that Kismet experiences emotion? Story for

426
00:24:09,200 --> 00:24:15,600
Speaker 1: another day. But these person versus program one on one

427
00:24:15,760 --> 00:24:18,600
Speaker 1: matches didn't stop with the game of chess. There was

428
00:24:18,640 --> 00:24:23,160
Speaker 1: also Jeopardy, right, And then there was one other one

429
00:24:23,280 --> 00:24:27,600
Speaker 1: that had tremendous implications for the world of programming.

430
00:24:28,080 --> 00:24:30,959
Speaker 2: Yeah, and that was much more recently with Google's Alpha Go.

431
00:24:32,119 --> 00:24:36,200
Speaker 2: That's Go, as in the ancient Chinese strategy game of Go,

432
00:24:36,800 --> 00:24:42,399
Speaker 2: and it successfully defeated a Go champion Kaig and Go

433
00:24:42,840 --> 00:24:48,800
Speaker 2: is a notoriously complex game, very difficult to predict, and

434
00:24:49,200 --> 00:24:51,880
Speaker 2: always have to be many many moves ahead of your opponent.

435
00:24:52,200 --> 00:24:55,359
Speaker 2: So that's pretty cool. That's almost a step further. Wouldn't

436
00:24:55,359 --> 00:24:59,119
Speaker 2: you say, Ben that Go would be more challenging for

437
00:24:59,200 --> 00:25:03,000
Speaker 2: a computer too, who beat a human opponent? Then chess?

438
00:25:03,040 --> 00:25:06,720
Speaker 2: Even Like, Yes, this is like a real serious leap forward.

439
00:25:06,960 --> 00:25:11,040
Speaker 1: I would absolutely agree, And a lot of people were

440
00:25:11,560 --> 00:25:16,320
Speaker 1: even more skeptical watching that game. People understood Go. We

441
00:25:16,400 --> 00:25:18,320
Speaker 1: can also go out there and say, Paul, I don't

442
00:25:18,320 --> 00:25:20,240
Speaker 1: know if you are familiar with this, but are you

443
00:25:20,960 --> 00:25:25,000
Speaker 1: Are you a Go enthusiast? Do you play this game? Not?

444
00:25:25,080 --> 00:25:28,800
Speaker 2: Really, Noel, I've never played it. I remember it from

445
00:25:28,840 --> 00:25:33,160
Speaker 2: the movie Pie. Remember Pie, Yeah, the one of Darren

446
00:25:33,160 --> 00:25:35,919
Speaker 2: Aronofski's first movies, which people kind of crap on. But

447
00:25:35,960 --> 00:25:37,800
Speaker 2: I enjoyed it a lot when I was young anyway.

448
00:25:37,840 --> 00:25:41,720
Speaker 2: But that movie is about a kind of a mad

449
00:25:42,040 --> 00:25:48,400
Speaker 2: computer scientist who realizes that the Kabbala contains some kind

450
00:25:48,400 --> 00:25:53,200
Speaker 2: of code involving the number for pie and gets goes

451
00:25:53,240 --> 00:25:56,000
Speaker 2: down a crazy rabbit hole of insanity in paranoia. But

452
00:25:56,080 --> 00:26:00,320
Speaker 2: Go as a recurring game in that movie, implying that

453
00:26:00,560 --> 00:26:03,480
Speaker 2: it's all about high level, very very high level thinking.

454
00:26:03,800 --> 00:26:06,080
Speaker 1: Yeah, I think that's I think that's excellent. You know,

455
00:26:06,200 --> 00:26:11,479
Speaker 1: Pie is a fantastic but jarring film, you know, and

456
00:26:11,560 --> 00:26:15,720
Speaker 1: so we see a common trend in these again, we'll

457
00:26:15,720 --> 00:26:19,600
Speaker 1: call them John Henry moments. And people generally know the

458
00:26:19,640 --> 00:26:21,879
Speaker 1: story of John Henry, right, nol you think they do.

459
00:26:22,000 --> 00:26:24,320
Speaker 2: I know he was a steel driving man, right, give

460
00:26:24,359 --> 00:26:25,040
Speaker 2: me some more.

461
00:26:25,080 --> 00:26:28,440
Speaker 1: Well John Henry for anyone who is outside of the US.

462
00:26:28,480 --> 00:26:31,040
Speaker 1: And I don't think we know for sure how common

463
00:26:31,080 --> 00:26:36,760
Speaker 1: this story is in the States anymore. He was in folklore.

464
00:26:37,200 --> 00:26:41,600
Speaker 1: He was an African American steel driving man, as Noel said,

465
00:26:41,680 --> 00:26:44,439
Speaker 1: that's absolutely true. And his job was to hammer a

466
00:26:44,640 --> 00:26:48,600
Speaker 1: steel drill into rock to make holes for explosives, right,

467
00:26:48,720 --> 00:26:52,520
Speaker 1: And they would blast rock constructing a railroad tunnel. And

468
00:26:53,200 --> 00:26:58,480
Speaker 1: in the legend he was pitted against a new fangled

469
00:26:58,800 --> 00:27:02,840
Speaker 1: steam powered hammer. It was a man against machine race.

470
00:27:03,680 --> 00:27:08,320
Speaker 1: And the historical aspect of whether this did or didn't

471
00:27:08,359 --> 00:27:10,920
Speaker 1: happen is did people go back and forth on it?

472
00:27:11,640 --> 00:27:14,440
Speaker 1: But what we're experiencing now with AI is a series

473
00:27:14,480 --> 00:27:19,840
Speaker 1: of increasingly high stakes John Henry moments. Key G you said,

474
00:27:20,080 --> 00:27:24,520
Speaker 1: Gary kasprov was it was it Ken Jennings who played

475
00:27:24,560 --> 00:27:27,280
Speaker 1: Deep Blue? Yeah, it was our own buddy.

476
00:27:27,880 --> 00:27:32,800
Speaker 2: Colleague, Ken Jennings of Omnibus fame, and then other things. Yeah,

477
00:27:32,840 --> 00:27:33,440
Speaker 2: he does.

478
00:27:35,520 --> 00:27:35,720
Speaker 3: Yeah.

479
00:27:35,800 --> 00:27:39,639
Speaker 1: Yeah, in addition to you know, his big tent item,

480
00:27:39,680 --> 00:27:42,680
Speaker 1: which is hanging hanging out with us at how stuff works.

481
00:27:42,840 --> 00:27:43,080
Speaker 2: Right.

482
00:27:43,760 --> 00:27:46,919
Speaker 1: Uh So, now we are on the cusp of a

483
00:27:46,960 --> 00:27:52,440
Speaker 1: world that is both brave to quote Alta Watson, Watson, Watson,

484
00:27:53,520 --> 00:27:55,480
Speaker 1: I said, deep Blue, and that's the chess.

485
00:27:55,200 --> 00:27:57,520
Speaker 2: On the chess one. Watson was the quiz the quiz

486
00:27:57,560 --> 00:27:59,959
Speaker 2: bot also by IBM though, right, that's correct.

487
00:28:00,080 --> 00:28:05,400
Speaker 1: Okay, yes, so Ken Jennings versus Watson, caspar Off versus

488
00:28:05,440 --> 00:28:09,640
Speaker 1: Deep Blue, and now we are on the cusp of

489
00:28:10,480 --> 00:28:14,320
Speaker 1: a new world that is both brave to quote Aldus Huxley,

490
00:28:14,680 --> 00:28:19,640
Speaker 1: and strange. The average citizen in a developed country interacts

491
00:28:19,720 --> 00:28:25,000
Speaker 1: with some form of artificial intelligence on an increasingly frequent basis,

492
00:28:25,080 --> 00:28:27,640
Speaker 1: even if it's indirect. I mean, think about it. When

493
00:28:27,720 --> 00:28:30,359
Speaker 1: is the last time you called a large company and

494
00:28:30,400 --> 00:28:34,159
Speaker 1: didn't first go through an automated line, a rough impersonation

495
00:28:34,320 --> 00:28:35,920
Speaker 1: of a conversation with a computer.

496
00:28:36,920 --> 00:28:40,280
Speaker 2: I just always start mashing zero right away. Just need

497
00:28:40,320 --> 00:28:44,200
Speaker 2: to pass because they it's so annoying, because it's like,

498
00:28:44,320 --> 00:28:48,480
Speaker 2: if I thought that a computer simulation could help me

499
00:28:48,520 --> 00:28:50,400
Speaker 2: solve the problem, I probably would have just done this

500
00:28:50,520 --> 00:28:52,880
Speaker 2: online right in the fact that I'm calling the company

501
00:28:52,920 --> 00:28:54,360
Speaker 2: in the first place, not to derail it with my

502
00:28:54,520 --> 00:28:57,240
Speaker 2: sure might get off my lawn moment. No, please do seriously,

503
00:28:57,280 --> 00:28:59,560
Speaker 2: It's like they're never that helpful. You always have to

504
00:28:59,600 --> 00:29:02,280
Speaker 2: repeat yourself over and over again, and you usually do

505
00:29:02,400 --> 00:29:06,680
Speaker 2: not get the result you're looking for, which could change

506
00:29:07,080 --> 00:29:10,360
Speaker 2: with some of the new technology we're talking about today,

507
00:29:10,400 --> 00:29:12,280
Speaker 2: which I think we can get into right now.

508
00:29:12,320 --> 00:29:16,000
Speaker 1: Possibly maybe maybe it'll change because one of the big

509
00:29:16,040 --> 00:29:20,280
Speaker 1: tasks for artificial intelligence right now, what we use these

510
00:29:20,320 --> 00:29:24,800
Speaker 1: applications for now is big data. We have built so

511
00:29:24,920 --> 00:29:28,880
Speaker 1: many ways to scoop up information that are poor. Old

512
00:29:29,040 --> 00:29:32,120
Speaker 1: primate brains, which are built for foraging and living in

513
00:29:32,160 --> 00:29:35,800
Speaker 1: small bands and forests, cannot analyze and process all this

514
00:29:35,840 --> 00:29:39,800
Speaker 1: stuff ourselves, so we built machines and codes. In this case,

515
00:29:39,960 --> 00:29:42,800
Speaker 1: codes could just be standing for lines of thought to

516
00:29:42,920 --> 00:29:46,200
Speaker 1: decipher all this data for us. And the application of

517
00:29:46,320 --> 00:29:49,880
Speaker 1: artificial intelligence in this regard has already done amazing things

518
00:29:49,880 --> 00:29:53,680
Speaker 1: in industries like technology, banking, marketing, entertainment.

519
00:29:53,920 --> 00:29:56,960
Speaker 2: Well, it's automation of like sweeping through giant sets of

520
00:29:57,040 --> 00:29:59,160
Speaker 2: data sure so that humans don't have to do it,

521
00:29:59,240 --> 00:30:03,360
Speaker 2: and using particular what's the word algorithms, I guess is

522
00:30:03,360 --> 00:30:06,080
Speaker 2: one way of putting it, or just rules saying, look

523
00:30:06,120 --> 00:30:08,880
Speaker 2: for this thing in this set of data, parse out

524
00:30:08,880 --> 00:30:11,320
Speaker 2: this file. If then whatever. So that's kind of been

525
00:30:11,320 --> 00:30:13,640
Speaker 2: the primary use of that up to this point. In

526
00:30:13,680 --> 00:30:17,400
Speaker 2: all of those things. You said, Yeah, so what's next?

527
00:30:17,760 --> 00:30:18,360
Speaker 2: Right right?

528
00:30:18,400 --> 00:30:22,880
Speaker 1: Because right now it doesn't matter if whether the algorithms

529
00:30:23,000 --> 00:30:26,640
Speaker 1: improve much, it doesn't matter whether there is a c change.

530
00:30:26,680 --> 00:30:30,600
Speaker 1: The basic concept is there. And this massive computing approach,

531
00:30:30,680 --> 00:30:34,040
Speaker 1: this vacuum cleaner approach, for instance, that the NSA uses,

532
00:30:34,840 --> 00:30:39,360
Speaker 1: just allows artificial intelligence to learn through brute force. In

533
00:30:39,400 --> 00:30:42,880
Speaker 1: the future, we're going to see more variation in AI.

534
00:30:43,000 --> 00:30:46,960
Speaker 1: We'll see autonomous vehicles, we'll see predictive ordering services, which

535
00:30:47,000 --> 00:30:50,240
Speaker 1: I know most of us will hate, and soon it

536
00:30:50,280 --> 00:30:53,760
Speaker 1: will seem strange not to have some sort of artificial

537
00:30:53,800 --> 00:30:58,480
Speaker 1: intelligence existing in some aspect of your everyday life. But

538
00:30:58,800 --> 00:31:02,200
Speaker 1: today's question, and a spooky one, is how close to

539
00:31:02,320 --> 00:31:05,480
Speaker 1: human can these programs actually get?

540
00:31:06,200 --> 00:31:09,480
Speaker 2: And I think we'll get to that after just one

541
00:31:09,480 --> 00:31:17,760
Speaker 2: more quick little sponsor break.

542
00:31:17,000 --> 00:31:19,320
Speaker 1: Excuse where it gets crazy.

543
00:31:22,600 --> 00:31:25,760
Speaker 2: See how I hear you? Hi, I'd like to reserve

544
00:31:25,760 --> 00:31:26,720
Speaker 2: a table for Wednesday.

545
00:31:26,840 --> 00:31:27,920
Speaker 1: The seventh.

546
00:31:29,440 --> 00:31:35,840
Speaker 2: For seven people. It's for four people, four people when

547
00:31:38,800 --> 00:31:43,640
Speaker 2: Wednesday at six pm? Oh, actually we leave her for

548
00:31:43,920 --> 00:31:47,720
Speaker 2: like apra, like a bye people for before you can come.

549
00:31:49,400 --> 00:31:52,600
Speaker 2: How long is the wait usually to be seated for

550
00:31:53,320 --> 00:31:59,040
Speaker 2: when tomorrow or weekey or for next Wednesday the seventh?

551
00:32:00,040 --> 00:32:01,360
Speaker 3: Oh no, it's not too easy.

552
00:32:02,360 --> 00:32:10,160
Speaker 2: You can com okay, Oh I gotcha. Thanks. But then

553
00:32:10,200 --> 00:32:12,560
Speaker 2: that that didn't sound like anything at all. That was

554
00:32:12,600 --> 00:32:15,520
Speaker 2: just a conversation between a young man and the proprietor

555
00:32:15,520 --> 00:32:17,320
Speaker 2: of a Chinese restaurant.

556
00:32:17,240 --> 00:32:22,840
Speaker 1: Right, and who was who is? Whoever has a great

557
00:32:22,880 --> 00:32:24,520
Speaker 1: time making restaurant reservation?

558
00:32:24,720 --> 00:32:26,880
Speaker 2: Now you know, no, because then we have open table

559
00:32:26,960 --> 00:32:28,000
Speaker 2: for that, am I? Right? Ben?

560
00:32:28,160 --> 00:32:31,840
Speaker 1: Right? I, like many people in our generation, hate talking

561
00:32:31,920 --> 00:32:32,480
Speaker 1: on the phone.

562
00:32:32,520 --> 00:32:34,800
Speaker 2: It's true. There's even a term for that. I think

563
00:32:34,920 --> 00:32:37,680
Speaker 2: was a telephoneophobia. Yes, yes, that's a good one.

564
00:32:37,720 --> 00:32:39,720
Speaker 1: But no I found Yeah, when I found that one,

565
00:32:39,800 --> 00:32:41,920
Speaker 1: I was like, how can you make a can you

566
00:32:42,000 --> 00:32:44,080
Speaker 1: just make anything a phobia? And if you can, I

567
00:32:44,120 --> 00:32:45,760
Speaker 1: think it's very American English.

568
00:32:45,840 --> 00:32:48,760
Speaker 2: Now I have microphoneophobia, which is weird considering that I

569
00:32:48,840 --> 00:32:51,239
Speaker 2: sit in front of one all the time. Right now,

570
00:32:51,400 --> 00:32:53,880
Speaker 2: I'm in utter terror. But no, I was being coy.

571
00:32:54,160 --> 00:32:57,360
Speaker 2: That clip was not between two humans. I think you

572
00:32:57,400 --> 00:33:00,240
Speaker 2: can probably guess which side of the conversation as the

573
00:33:00,320 --> 00:33:03,560
Speaker 2: non human yep, because if you listen back to it,

574
00:33:03,720 --> 00:33:07,840
Speaker 2: I think you'll notice a couple of tells in that

575
00:33:07,880 --> 00:33:10,000
Speaker 2: there was a few There were a few kind of

576
00:33:10,000 --> 00:33:14,680
Speaker 2: wrenches thrown into that exchange where the person was trying

577
00:33:14,720 --> 00:33:17,640
Speaker 2: to call to make a reservation, and there was some

578
00:33:17,680 --> 00:33:20,720
Speaker 2: confusion from the person on the other end about how many,

579
00:33:20,760 --> 00:33:23,320
Speaker 2: about what day they wanted the reservation to be, how

580
00:33:23,360 --> 00:33:26,880
Speaker 2: many people were in the party, and it ultimately ended

581
00:33:26,960 --> 00:33:29,360
Speaker 2: up with, well, you don't really need a reservation. So

582
00:33:29,400 --> 00:33:31,280
Speaker 2: the person on the other of the phone was not equipped.

583
00:33:31,400 --> 00:33:33,800
Speaker 2: It was not something they would typically do. Reservations seemed

584
00:33:33,800 --> 00:33:35,760
Speaker 2: like not really a thing in this restaurant, and so

585
00:33:35,840 --> 00:33:41,400
Speaker 2: the caller kind of repeated himself.

586
00:33:41,200 --> 00:33:44,800
Speaker 1: Specific things like the specific time and day, and then

587
00:33:44,840 --> 00:33:47,960
Speaker 1: there was a momentary pause when they said we we

588
00:33:48,000 --> 00:33:52,040
Speaker 1: only do reservations for groups of five those ah, gotcha,

589
00:33:53,120 --> 00:33:54,560
Speaker 1: which still sounded very human.

590
00:33:54,640 --> 00:33:55,000
Speaker 2: It did.

591
00:33:55,200 --> 00:33:57,480
Speaker 1: It was not, though. It was a conversation between an

592
00:33:57,560 --> 00:34:01,440
Speaker 1: unsuspected restaurant employee, as you mentioned, and a computer program.

593
00:34:01,840 --> 00:34:07,880
Speaker 1: This is an example of Google's Duplex system, a personal

594
00:34:07,920 --> 00:34:12,719
Speaker 1: assistant designed to make users' lives easier by handling standard

595
00:34:12,760 --> 00:34:16,839
Speaker 1: phone calls, so like doctor's appointments, reservations, and so on,

596
00:34:17,760 --> 00:34:21,200
Speaker 1: breaking up with a loved one without troubling you or

597
00:34:21,239 --> 00:34:23,960
Speaker 1: possibly even letting the person on the other end of

598
00:34:23,960 --> 00:34:26,759
Speaker 1: the phone know that it's not actually you making the call.

599
00:34:27,040 --> 00:34:29,320
Speaker 2: Yeah, and we get into all kinds of ethical quandaries

600
00:34:29,320 --> 00:34:31,799
Speaker 2: with this that we'll get into first, first and foremost, though,

601
00:34:31,840 --> 00:34:36,000
Speaker 2: I think it's interesting that the idea, it's inherently tricksy

602
00:34:36,200 --> 00:34:41,600
Speaker 2: the whole affair, right, The idea is to not inconvenience

603
00:34:41,680 --> 00:34:44,680
Speaker 2: either party, but you or you are genuinely kind of

604
00:34:44,719 --> 00:34:46,680
Speaker 2: tricking someone into believing they're talking to a person.

605
00:34:47,080 --> 00:34:49,080
Speaker 1: Yes, and you're giving up a lot of stuff. So

606
00:34:49,760 --> 00:34:55,719
Speaker 1: at first this seems amazing because, like we established earlier, telephoneophobia,

607
00:34:55,760 --> 00:34:58,839
Speaker 1: in addition to being really fun to say, is a

608
00:34:58,920 --> 00:35:03,200
Speaker 1: genuine thing. I would argue, increasingly people are resorting to text.

609
00:35:03,440 --> 00:35:05,120
Speaker 2: Well, I mean, how often do you talk on the

610
00:35:05,120 --> 00:35:07,000
Speaker 2: phone if you're not talking to like your mom or dad,

611
00:35:07,120 --> 00:35:08,919
Speaker 2: like on a preappointed time.

612
00:35:09,719 --> 00:35:12,040
Speaker 1: I mean, if I'm doing interviews for a podcast.

613
00:35:12,080 --> 00:35:15,759
Speaker 2: Well that's true, but that's a very specific task. Not

614
00:35:15,800 --> 00:35:18,759
Speaker 2: for fun, never for fun, and unless I just feel

615
00:35:18,800 --> 00:35:20,719
Speaker 2: like I could suss something out a lot quicker on

616
00:35:20,760 --> 00:35:23,560
Speaker 2: the phone, real quick than I could usually for business though,

617
00:35:24,040 --> 00:35:27,400
Speaker 2: for work stuff sometimes or conference calls, the dreaded conference

618
00:35:27,440 --> 00:35:28,280
Speaker 2: call that's.

619
00:35:28,280 --> 00:35:30,120
Speaker 1: Also works, the thing that still happens.

620
00:35:30,160 --> 00:35:34,640
Speaker 2: But yeah, for fun, not very often, you know, But

621
00:35:34,800 --> 00:35:38,040
Speaker 2: for other people in the mix. There's totally a dark

622
00:35:38,080 --> 00:35:42,400
Speaker 2: side to this idea of this super convenient, amazing idea

623
00:35:42,760 --> 00:35:45,080
Speaker 2: of my phone being able to make calls for me

624
00:35:45,440 --> 00:35:49,040
Speaker 2: and set up my you know, my waxing appointment or whatever,

625
00:35:50,239 --> 00:35:53,880
Speaker 2: because the implications are incredibly far reaching, aren't they been?

626
00:35:54,120 --> 00:35:58,040
Speaker 1: Yeah, they are. Noal it's not just a it's not

627
00:35:58,440 --> 00:36:03,799
Speaker 1: just an automated phone online saying press one if you

628
00:36:04,320 --> 00:36:06,600
Speaker 1: like to make a payment, press two if you'd like

629
00:36:06,719 --> 00:36:10,359
Speaker 1: to hear a mailing address. It's not just impersoning a

630
00:36:10,400 --> 00:36:15,840
Speaker 1: generic human. Now, it's a program capable of impersonating specific humans,

631
00:36:16,000 --> 00:36:20,799
Speaker 1: namely you, capital y oh, you specifically you listening to this,

632
00:36:20,880 --> 00:36:24,080
Speaker 1: and it's got a it's got a lot of opponents already.

633
00:36:24,520 --> 00:36:28,399
Speaker 2: Now, we're not saying that it mimics your voice. That's

634
00:36:28,440 --> 00:36:31,799
Speaker 2: not what this is. Implying not yet. Not yet, But

635
00:36:31,880 --> 00:36:34,120
Speaker 2: the idea is that it is. It has access to

636
00:36:34,120 --> 00:36:37,919
Speaker 2: all your information right It has access to your date

637
00:36:37,920 --> 00:36:41,360
Speaker 2: books and your phone numbers, and knows everything about you

638
00:36:41,400 --> 00:36:44,600
Speaker 2: because it's tied into this little thing you carry around

639
00:36:44,680 --> 00:36:47,400
Speaker 2: with you that is basically your life in a wallet

640
00:36:47,440 --> 00:36:48,000
Speaker 2: form you.

641
00:36:47,960 --> 00:36:53,719
Speaker 1: Know more specifically with Google. This this initial fear increases,

642
00:36:53,880 --> 00:36:57,359
Speaker 1: or the potential for misuse increases. This journalist, who wrote

643
00:36:57,360 --> 00:37:00,759
Speaker 1: an excellent article in her name's Alex Kranz. To her,

644
00:37:00,880 --> 00:37:04,960
Speaker 1: the dangers are threefold. First, similar to what you're proposing,

645
00:37:05,719 --> 00:37:10,080
Speaker 1: this is a program from Google from Alphabet, and Google

646
00:37:10,239 --> 00:37:14,040
Speaker 1: already knows a ton of personal information about you, whether

647
00:37:14,120 --> 00:37:16,680
Speaker 1: or not you use an Android phone. This means that

648
00:37:16,760 --> 00:37:19,640
Speaker 1: Duplex doesn't just know what time you want to meet

649
00:37:19,680 --> 00:37:23,279
Speaker 1: your tender date at that Tapas place. It also potentially

650
00:37:23,320 --> 00:37:27,360
Speaker 1: knows every other single thing Google knows about your life.

651
00:37:27,760 --> 00:37:30,640
Speaker 1: And Google is so widespread that you do not have

652
00:37:30,680 --> 00:37:33,520
Speaker 1: to have a Google account for this to touch you. Second,

653
00:37:33,840 --> 00:37:38,360
Speaker 1: a criminal, a government, a stalker, a prankster, some Internet

654
00:37:38,360 --> 00:37:43,279
Speaker 1: troll could potentially cheat Duplex out of giving up your

655
00:37:43,320 --> 00:37:47,680
Speaker 1: information in a phone call. Imagine Duplex somehow being fooled

656
00:37:47,680 --> 00:37:51,440
Speaker 1: into thinking it's calling a restaurant to make a delivery

657
00:37:51,600 --> 00:37:55,080
Speaker 1: order and then boom, somebody has your credit card information,

658
00:37:55,200 --> 00:37:57,160
Speaker 1: expiration date, and security code.

659
00:37:57,440 --> 00:38:00,200
Speaker 2: There were actually a really great, a whole bunch of

660
00:38:00,200 --> 00:38:03,720
Speaker 2: great comments in the comment section on this gizmoto article

661
00:38:04,000 --> 00:38:06,840
Speaker 2: by Alex Kranz, and one of them was the idea of, like,

662
00:38:06,960 --> 00:38:09,640
Speaker 2: how this technology could be used by this term I

663
00:38:09,640 --> 00:38:12,719
Speaker 2: love bad actors, you know, so for example, like what

664
00:38:12,880 --> 00:38:16,359
Speaker 2: if you had a pizza place that wanted to use

665
00:38:16,400 --> 00:38:21,319
Speaker 2: this kind of technology to flood arrival pizza company with

666
00:38:21,560 --> 00:38:25,160
Speaker 2: fake calls and keep their phones tied up? And I

667
00:38:25,200 --> 00:38:27,120
Speaker 2: don't know that was a particularly one in sort of

668
00:38:27,120 --> 00:38:30,040
Speaker 2: a far fetched one, but the idea, the implications being

669
00:38:30,280 --> 00:38:35,000
Speaker 2: that you could use this for scamming easily, big time,

670
00:38:35,280 --> 00:38:38,399
Speaker 2: especially with older folks that maybe aren't going to key

671
00:38:38,480 --> 00:38:40,480
Speaker 2: in to the fact that this is not a real person,

672
00:38:40,880 --> 00:38:45,759
Speaker 2: and you could possibly automate and scale scamming in a

673
00:38:45,800 --> 00:38:48,279
Speaker 2: way that would far surpass the way it's done now,

674
00:38:48,320 --> 00:38:51,360
Speaker 2: because even the least tech savvy people can tell that

675
00:38:51,440 --> 00:38:55,239
Speaker 2: you're hearing a automated voice, you know, spamming you to

676
00:38:55,480 --> 00:38:57,840
Speaker 2: try to, you know, sign up for a cruise or

677
00:38:57,880 --> 00:38:58,520
Speaker 2: something like that.

678
00:38:59,160 --> 00:39:03,560
Speaker 1: Absolutely, and to follow up on Kranz's because we said

679
00:39:03,560 --> 00:39:06,160
Speaker 1: it was threefold, So to follow up on the third thing,

680
00:39:06,320 --> 00:39:11,720
Speaker 1: which I found perhaps the most disturbing here, No forget,

681
00:39:11,760 --> 00:39:15,600
Speaker 1: perhaps it is the most disturbing thing. A person could

682
00:39:15,719 --> 00:39:21,960
Speaker 1: hijack your duplex account and essentially function as you until

683
00:39:22,000 --> 00:39:24,879
Speaker 1: such a point as they're caught. So imagine learning your

684
00:39:24,880 --> 00:39:27,960
Speaker 1: credit and savings have been wiped out because your duplex

685
00:39:28,040 --> 00:39:32,080
Speaker 1: called in a series of untraceable transfers to some Caribbean island,

686
00:39:32,160 --> 00:39:36,320
Speaker 1: some offshore account. There's nothing that anyone could do because

687
00:39:36,440 --> 00:39:39,800
Speaker 1: legally that thing would be functioning as you. Crans and

688
00:39:39,880 --> 00:39:43,080
Speaker 1: other critics of this technology are also concerned that at

689
00:39:43,120 --> 00:39:47,760
Speaker 1: the initial presentation, Google build this as another NEDO feature

690
00:39:47,840 --> 00:39:50,319
Speaker 1: and they didn't say anything about privacy or concerns. It

691
00:39:50,360 --> 00:39:53,080
Speaker 1: was very much, look what we can do, not hey

692
00:39:53,120 --> 00:39:54,920
Speaker 1: should we Just.

693
00:39:54,800 --> 00:39:58,120
Speaker 2: Like it's like that line from Jurassic Park. Your scientists

694
00:39:58,120 --> 00:40:00,440
Speaker 2: were spending so much time thinking whether they could, They

695
00:40:00,480 --> 00:40:02,239
Speaker 2: never thought if they should.

696
00:40:02,360 --> 00:40:04,680
Speaker 1: And crime finds a way.

697
00:40:04,760 --> 00:40:08,040
Speaker 2: It does. It does. But you know, and you may

698
00:40:08,080 --> 00:40:10,120
Speaker 2: think this is an alarmist, but it's just like we're

699
00:40:10,160 --> 00:40:12,000
Speaker 2: just a hop skipping a jump away from this stuff.

700
00:40:12,000 --> 00:40:15,480
Speaker 2: And obviously this technology is not ready for market. No, no, no,

701
00:40:15,480 --> 00:40:17,000
Speaker 2: one's saying that this is just kind of like a

702
00:40:17,120 --> 00:40:20,280
Speaker 2: Nido party trick they did at their Io Developers conference.

703
00:40:20,480 --> 00:40:24,680
Speaker 2: But yeah, you're right, because if it's literally the voice

704
00:40:24,719 --> 00:40:28,400
Speaker 2: inside your phone, then it's going to have access to

705
00:40:28,400 --> 00:40:30,960
Speaker 2: all the data that's in your phone. And depending on

706
00:40:31,040 --> 00:40:33,400
Speaker 2: how careful you are with this stuff, it could have

707
00:40:33,560 --> 00:40:37,280
Speaker 2: every bit of identifying information that a human banker would

708
00:40:37,320 --> 00:40:41,200
Speaker 2: ask you to give to confirm this crazy wire transfer

709
00:40:41,239 --> 00:40:42,440
Speaker 2: that you want to do. What do they ask you for?

710
00:40:42,520 --> 00:40:45,000
Speaker 2: Ben they ask you for like the last four digits

711
00:40:45,040 --> 00:40:48,399
Speaker 2: of your social security number, your mailing address, your name,

712
00:40:48,520 --> 00:40:51,239
Speaker 2: maybe your mother's maiden name, or some secret word some

713
00:40:51,440 --> 00:40:54,640
Speaker 2: security going to be on your phone in some way,

714
00:40:54,680 --> 00:40:55,640
Speaker 2: shape or form.

715
00:40:55,520 --> 00:40:59,359
Speaker 1: Likely, and many people email themselves those answers so when

716
00:40:59,400 --> 00:41:02,560
Speaker 1: they're online and they can check back, including passwords. So

717
00:41:03,120 --> 00:41:05,919
Speaker 1: it's a very important point to make that it has

718
00:41:06,120 --> 00:41:08,520
Speaker 1: this duplex. Stuff has not been assigned any kind of

719
00:41:08,560 --> 00:41:11,760
Speaker 1: roll out date. People are still speculating just how close

720
00:41:11,840 --> 00:41:15,719
Speaker 1: or how far Google is from making this technology viable

721
00:41:15,800 --> 00:41:18,960
Speaker 1: outside of testing situations and later available to the public.

722
00:41:19,200 --> 00:41:21,600
Speaker 1: But the problem is this is not the only game

723
00:41:21,640 --> 00:41:26,280
Speaker 1: in town. Sure, audio impersonation is spooky, but what about video.

724
00:41:26,400 --> 00:41:29,319
Speaker 1: Do you remember back during the hunt for Osama bin

725
00:41:29,400 --> 00:41:33,080
Speaker 1: Laden where various people were arguing that the bin Laden

726
00:41:33,080 --> 00:41:36,439
Speaker 1: and propaganda videos wasn't the real guy, but someone else

727
00:41:36,440 --> 00:41:37,640
Speaker 1: impersonating him or.

728
00:41:37,640 --> 00:41:40,600
Speaker 2: Something like that. Sure, or like a Snapchat filter. Right, Yes,

729
00:41:40,600 --> 00:41:43,200
Speaker 2: it's kidding, but I'm not because all of the crazy tech.

730
00:41:43,360 --> 00:41:46,080
Speaker 2: It blows my mind how much research and development they

731
00:41:46,080 --> 00:41:48,279
Speaker 2: put into the Snapchat filter. Oh yeah, and the more

732
00:41:48,320 --> 00:41:50,239
Speaker 2: you see them, the more realistic they get, and the

733
00:41:50,239 --> 00:41:53,160
Speaker 2: more I could see them turning into some pretty nefarious

734
00:41:53,200 --> 00:41:56,000
Speaker 2: ways of mapping people's faces and you know, making you

735
00:41:56,040 --> 00:41:58,319
Speaker 2: look like someone else entirely and making it not just

736
00:41:58,360 --> 00:42:02,279
Speaker 2: a silly mustache, but totally believable doppelganger.

737
00:42:02,680 --> 00:42:06,919
Speaker 1: Yeah, and we've seen Look, visual manipulation is a tail

738
00:42:07,040 --> 00:42:10,480
Speaker 1: as old as journalism. We've seen doctor photos of plenty

739
00:42:10,560 --> 00:42:15,399
Speaker 1: purporting to be evidence of UFOs, giant skeletons, ghosts, and

740
00:42:15,760 --> 00:42:18,279
Speaker 1: we as a species have known for a long time

741
00:42:18,320 --> 00:42:21,480
Speaker 1: that photos can be faked. Someone with a handy knack

742
00:42:21,600 --> 00:42:27,520
Speaker 1: for photoshop, Paul can really work. Wonders and nowadays, we've

743
00:42:27,520 --> 00:42:30,160
Speaker 1: gone beyond the touch ups of still images.

744
00:42:30,280 --> 00:42:32,040
Speaker 2: Have you seen that doctored stuff They don't want you

745
00:42:32,160 --> 00:42:36,480
Speaker 2: to know logo there's some something fishy going on with

746
00:42:36,560 --> 00:42:37,480
Speaker 2: that hand.

747
00:42:37,480 --> 00:42:43,440
Speaker 1: Suspicious indeed, but again no spoilers, right, So if you're

748
00:42:43,520 --> 00:42:46,600
Speaker 1: wondering fans of Westworld, yes, that maze is for you.

749
00:42:47,239 --> 00:42:50,640
Speaker 1: As The Verge reported in July of last year, twenty seventeen,

750
00:42:50,680 --> 00:42:54,320
Speaker 1: we're recording this twenty eighteen July, researchers at the University

751
00:42:54,360 --> 00:42:58,320
Speaker 1: of Washington invented a tool that takes audio files, converts

752
00:42:58,360 --> 00:43:02,080
Speaker 1: them to realistic mouth movements, and then graphs those onto

753
00:43:02,160 --> 00:43:05,480
Speaker 1: existing video. The end result of this is someone saying

754
00:43:05,560 --> 00:43:08,520
Speaker 1: something or appearing to say something that they never actually said.

755
00:43:08,719 --> 00:43:11,560
Speaker 1: And the scary part is it's really convincing. It's not

756
00:43:11,680 --> 00:43:14,840
Speaker 1: like meme level give Funny a lot of their earlier

757
00:43:14,880 --> 00:43:18,560
Speaker 1: examples used footage of former President Barack Obama. Did you

758
00:43:18,600 --> 00:43:19,240
Speaker 1: see this stuff?

759
00:43:19,320 --> 00:43:20,719
Speaker 2: Well, dude, there, I mean, and I don't know, I

760
00:43:20,800 --> 00:43:23,680
Speaker 2: keep harping on this, but there was a Barack Obama

761
00:43:23,760 --> 00:43:28,680
Speaker 2: snapchat filter and it it does have that Uncanny Valley look.

762
00:43:29,080 --> 00:43:31,640
Speaker 2: But first of all, problematic, you know, for a lot

763
00:43:31,640 --> 00:43:35,240
Speaker 2: of reasons sure to put on the visage of someone

764
00:43:35,280 --> 00:43:37,720
Speaker 2: that's a different race than you and play a character

765
00:43:37,840 --> 00:43:38,479
Speaker 2: and you know.

766
00:43:38,640 --> 00:43:40,839
Speaker 1: Oh, just Snapchat users were using it.

767
00:43:40,880 --> 00:43:42,480
Speaker 2: That's what I'm saying. It was literally a Snapchat film.

768
00:43:42,520 --> 00:43:44,920
Speaker 2: There's also another one that was a Bob Marley you know,

769
00:43:45,080 --> 00:43:48,839
Speaker 2: with like but but it's pretty damn realistic. Other than

770
00:43:49,080 --> 00:43:51,280
Speaker 2: there's things they probably did on purpose to make it

771
00:43:51,400 --> 00:43:54,520
Speaker 2: less realistic. But that kind of mapping quality that you're

772
00:43:54,560 --> 00:43:57,320
Speaker 2: talking about, the implications there are nutty.

773
00:43:57,760 --> 00:44:01,160
Speaker 1: Yeah, And luckily for everybody I just got spooked out,

774
00:44:01,200 --> 00:44:06,359
Speaker 1: We have some good news here. First, they didn't just

775
00:44:06,719 --> 00:44:10,080
Speaker 1: choose Barack Obama because they had some sort of ideological

776
00:44:10,120 --> 00:44:12,080
Speaker 1: thing or they were like we like this guy or

777
00:44:12,080 --> 00:44:14,360
Speaker 1: we hate this guy or anything. They did it because

778
00:44:14,400 --> 00:44:17,360
Speaker 1: a high profile individual, like a celebrity or a president

779
00:44:17,600 --> 00:44:20,160
Speaker 1: will have a ton more high quality video and audio

780
00:44:20,160 --> 00:44:24,120
Speaker 1: footage to pull from. And also, this is this thing

781
00:44:24,280 --> 00:44:28,440
Speaker 1: takes it's a huge attrition process. They had to have

782
00:44:28,480 --> 00:44:32,000
Speaker 1: at least seventeen hours of footage just to get started.

783
00:44:32,400 --> 00:44:33,440
Speaker 1: And they say, so.

784
00:44:33,400 --> 00:44:35,520
Speaker 2: There's another neural net kind of learning.

785
00:44:35,160 --> 00:44:39,160
Speaker 1: Exactly yap learning stuffing. So this is a problem. They

786
00:44:39,200 --> 00:44:42,520
Speaker 1: say that their goals are wholly good hearted. We've got

787
00:44:42,520 --> 00:44:45,359
Speaker 1: a quote here the team behind the works say they

788
00:44:45,400 --> 00:44:48,000
Speaker 1: hope it could be used to improve video chat tools

789
00:44:48,080 --> 00:44:53,120
Speaker 1: like Skype. Users could collect footage of themselves speaking, use

790
00:44:53,160 --> 00:44:55,359
Speaker 1: it to train the software, and then when they need

791
00:44:55,360 --> 00:44:57,759
Speaker 1: to talk to someone video on their side would be

792
00:44:57,840 --> 00:45:01,680
Speaker 1: generated automatically, just use their voice. This would help in

793
00:45:01,760 --> 00:45:05,040
Speaker 1: situations where someone's Internet is shaky, or if they're trying

794
00:45:05,080 --> 00:45:06,520
Speaker 1: to save mobile data.

795
00:45:07,000 --> 00:45:09,400
Speaker 2: But that's what Google's saying about Duplex. It's just a

796
00:45:09,480 --> 00:45:13,000
Speaker 2: nifty little, handy dandy tool to help you book your

797
00:45:13,040 --> 00:45:15,200
Speaker 2: hair appointment so you don't have to talk to people

798
00:45:15,239 --> 00:45:17,359
Speaker 2: on the phone. They never talk about the.

799
00:45:18,000 --> 00:45:22,960
Speaker 1: Yeah, yeah, exactly, and this look can't. We can't ascribe motive.

800
00:45:23,040 --> 00:45:26,200
Speaker 1: We can't say these people are lying to you, but

801
00:45:26,320 --> 00:45:29,400
Speaker 1: we can say that this sort of technology is a

802
00:45:29,480 --> 00:45:33,520
Speaker 1: Pandora's jar, And once that lid is unscrewed, there's no

803
00:45:34,040 --> 00:45:37,440
Speaker 1: There is absolutely no realistic way to prevent both the

804
00:45:37,520 --> 00:45:41,080
Speaker 1: spread of these faked segments yeah, and the spread of

805
00:45:41,200 --> 00:45:45,399
Speaker 1: things being accused unjustly of fake segments. This poses some

806
00:45:45,480 --> 00:45:49,040
Speaker 1: inherent dangers for journalism. We already see how easily a

807
00:45:49,080 --> 00:45:52,480
Speaker 1: completely fake story can proliferate on Facebook.

808
00:45:52,120 --> 00:45:55,600
Speaker 2: A bot generated story, yeah times like literally an AI

809
00:45:55,880 --> 00:45:59,960
Speaker 2: generated Twitter account, for example, that can mimic the style

810
00:46:00,000 --> 00:46:05,359
Speaker 2: aisle of someone of a particular ideology or whatever. Here's

811
00:46:05,360 --> 00:46:07,720
Speaker 2: another aspect of it. There's an article from the Information

812
00:46:07,840 --> 00:46:11,759
Speaker 2: dot Com. Google's controversial voice assistant could talk its way

813
00:46:12,120 --> 00:46:17,080
Speaker 2: into call centers because, like we're saying earlier, if you

814
00:46:17,120 --> 00:46:19,560
Speaker 2: get an automated voice on the phone, you know it's

815
00:46:19,600 --> 00:46:22,040
Speaker 2: not going to be actually very helpful at all. You're

816
00:46:22,080 --> 00:46:24,640
Speaker 2: probably just going to blast past it. And they know that.

817
00:46:24,719 --> 00:46:26,640
Speaker 2: And there are a lot of people that have their

818
00:46:26,760 --> 00:46:29,719
Speaker 2: jobs being that person that it gets passed off too.

819
00:46:30,239 --> 00:46:35,000
Speaker 2: So if there were a more successful voice recognition and

820
00:46:35,080 --> 00:46:37,520
Speaker 2: communication tool like this, it could put a lot of

821
00:46:37,520 --> 00:46:38,360
Speaker 2: people out of jobs.

822
00:46:39,000 --> 00:46:41,320
Speaker 1: That's a very good point. It could get us closer

823
00:46:41,360 --> 00:46:44,840
Speaker 1: to not the post work economy, but the post worker economy.

824
00:46:45,440 --> 00:46:49,720
Speaker 1: And up to now we've talked about these two forms

825
00:46:49,760 --> 00:46:53,680
Speaker 1: of impersonation is discrete in different things, right, audio on

826
00:46:53,680 --> 00:46:56,920
Speaker 1: one side, video on another. But what if they become combined.

827
00:46:57,120 --> 00:47:00,880
Speaker 1: What if a digital impersonation of a human being, even

828
00:47:00,960 --> 00:47:03,240
Speaker 1: you knowl or even you Paul, or even me could

829
00:47:03,280 --> 00:47:06,520
Speaker 1: exist online with no one but you and the people

830
00:47:06,600 --> 00:47:09,799
Speaker 1: you meet in person knowing the difference. I mean, let's

831
00:47:09,800 --> 00:47:12,239
Speaker 1: think about it. It would sound like us, it would

832
00:47:12,320 --> 00:47:15,840
Speaker 1: look like us, and if it pulled from our online

833
00:47:16,080 --> 00:47:18,600
Speaker 1: data footprint, it would also know a lot of stuff

834
00:47:18,640 --> 00:47:22,040
Speaker 1: about us, including the relationships we already have with other

835
00:47:22,080 --> 00:47:25,560
Speaker 1: people and how we interact with them. So it's possible,

836
00:47:25,840 --> 00:47:30,839
Speaker 1: for instance, that a fake version of Matt Frederick writs

837
00:47:31,239 --> 00:47:37,400
Speaker 1: to us from Massachusetts foreshadowing there and responds to the

838
00:47:37,480 --> 00:47:40,120
Speaker 1: three of us the way that the real Matt would,

839
00:47:40,160 --> 00:47:42,560
Speaker 1: and we wouldn't know the difference. We'd be sending weird

840
00:47:42,800 --> 00:47:45,799
Speaker 1: memes and thumbs up and doing our inside jokes, and

841
00:47:45,840 --> 00:47:47,319
Speaker 1: it would already know all of those.

842
00:47:47,400 --> 00:47:49,160
Speaker 2: Well, let's be honest. I mean a lot of times

843
00:47:49,239 --> 00:47:51,880
Speaker 2: text conversation and email conversations are already kind of in

844
00:47:51,960 --> 00:47:54,520
Speaker 2: shorthand or in some kind of a little bit more terse,

845
00:47:54,680 --> 00:47:56,160
Speaker 2: not mean, but just kind of like we're trying to

846
00:47:56,200 --> 00:47:58,600
Speaker 2: get it done. That's why we depend on that method

847
00:47:58,600 --> 00:48:01,839
Speaker 2: of communication over talking, because it's a lot more boom

848
00:48:01,880 --> 00:48:03,880
Speaker 2: boom boom, let's get it done and move on. I

849
00:48:03,880 --> 00:48:06,400
Speaker 2: would think it would be easy ish for an AI

850
00:48:06,560 --> 00:48:10,319
Speaker 2: to mimic those kinds of little quick exchanges and not

851
00:48:10,560 --> 00:48:12,120
Speaker 2: raise a red flag for you or I.

852
00:48:12,360 --> 00:48:14,480
Speaker 1: That's a great point because even if they get something wrong,

853
00:48:14,520 --> 00:48:16,080
Speaker 1: we would just think, oh, Matt must.

854
00:48:15,920 --> 00:48:18,920
Speaker 2: Be in a hurry typo or you know, fat finger whatever.

855
00:48:19,000 --> 00:48:21,839
Speaker 1: Yeah, and auto correct.

856
00:48:21,640 --> 00:48:26,040
Speaker 2: Autocorrect exactly like I've sent texts and like seen how

857
00:48:26,080 --> 00:48:28,000
Speaker 2: mangled they were, But be like, oh, he'll know what

858
00:48:28,000 --> 00:48:30,040
Speaker 2: I mean. Yes, like let let it ride.

859
00:48:30,239 --> 00:48:32,680
Speaker 1: Yeah, yeah, yeah. Because again it goes to time. You know,

860
00:48:32,840 --> 00:48:35,160
Speaker 1: it would be alarmist for us to say this would

861
00:48:35,200 --> 00:48:39,040
Speaker 1: plausibly happen to the average person in the near future,

862
00:48:39,280 --> 00:48:42,600
Speaker 1: but for a high value target like a politician, a celebrity,

863
00:48:42,800 --> 00:48:46,359
Speaker 1: a controversial business person and so on, it's completely within

864
00:48:46,400 --> 00:48:49,560
Speaker 1: the realm of possibility. And speaking of alarmists, let's open

865
00:48:49,640 --> 00:48:51,879
Speaker 1: wide the doors of science fiction and do a bit

866
00:48:51,920 --> 00:48:56,319
Speaker 1: of speculation here, completely unfounded. Imagine a world where the

867
00:48:56,360 --> 00:49:00,920
Speaker 1: only communication you can trust becomes face to face in person.

868
00:49:01,640 --> 00:49:04,400
Speaker 1: Imagine a world where you are accused of a crime

869
00:49:04,480 --> 00:49:07,360
Speaker 1: you didn't commit, but you don't have a rock solid

870
00:49:07,440 --> 00:49:10,320
Speaker 1: alibi of your activities at the time of the alleged crime,

871
00:49:10,480 --> 00:49:13,480
Speaker 1: and surprise, surprise, they have you on camera committing it

872
00:49:13,560 --> 00:49:16,040
Speaker 1: and then confessing to it, and you cannot prove it

873
00:49:16,080 --> 00:49:19,360
Speaker 1: is not you would this mean that eventually video evidence

874
00:49:19,400 --> 00:49:23,520
Speaker 1: becomes inadmissible in court? Does this technology pose an existential

875
00:49:23,600 --> 00:49:27,520
Speaker 1: threat to the fabric of digital reality? Is it really

876
00:49:27,600 --> 00:49:29,959
Speaker 1: us talking to you right now?

877
00:49:30,120 --> 00:49:32,120
Speaker 2: Yeah, man, I don't even know anymore.

878
00:49:32,320 --> 00:49:32,759
Speaker 1: I don't know.

879
00:49:33,040 --> 00:49:37,080
Speaker 2: And to answer your hypothetical question earlier, yes, I think

880
00:49:37,120 --> 00:49:39,880
Speaker 2: this is absolutely a threat, and we rely on things

881
00:49:39,960 --> 00:49:42,520
Speaker 2: like video evidence. But I could see far enough removed

882
00:49:42,560 --> 00:49:46,160
Speaker 2: from this particular technological time and place that could be

883
00:49:46,200 --> 00:49:48,439
Speaker 2: something that is just a thing of the past. Man,

884
00:49:48,520 --> 00:49:49,040
Speaker 2: you know what I mean?

885
00:49:49,239 --> 00:49:49,680
Speaker 1: Yeah?

886
00:49:49,840 --> 00:49:52,040
Speaker 2: Can you picture it like I'm trying to think of

887
00:49:52,040 --> 00:49:54,799
Speaker 2: of an analog of something that used to be infallible

888
00:49:55,120 --> 00:49:58,520
Speaker 2: and now as a completely up for grabs, like.

889
00:49:58,440 --> 00:49:59,520
Speaker 1: I don't know, polygraphs.

890
00:49:59,600 --> 00:50:02,320
Speaker 2: Yeah, exactly, a great example. I don't know. This is

891
00:50:02,320 --> 00:50:04,399
Speaker 2: a little bit more of an ephemeral example. But even

892
00:50:04,400 --> 00:50:07,320
Speaker 2: something like the press, or like the news you know now,

893
00:50:07,480 --> 00:50:09,239
Speaker 2: you know, used to to be able to depend on

894
00:50:09,280 --> 00:50:12,080
Speaker 2: some level of rigor and truth in any kind of

895
00:50:12,080 --> 00:50:14,239
Speaker 2: news reporting you see, And now, as you say, because

896
00:50:14,280 --> 00:50:17,960
Speaker 2: of the Internet, we can see stuff that's completely generated

897
00:50:18,280 --> 00:50:22,279
Speaker 2: by artificial intelligence that tricks people all the time into

898
00:50:22,320 --> 00:50:24,400
Speaker 2: believing that it's you know, God's truth.

899
00:50:24,680 --> 00:50:29,040
Speaker 1: And it's insane. So I wonder you know, I have

900
00:50:29,160 --> 00:50:31,279
Speaker 1: questions for you to No, I don't want to put

901
00:50:31,280 --> 00:50:34,080
Speaker 1: you on the spot, but this is this is one

902
00:50:34,120 --> 00:50:36,760
Speaker 1: of the only conversations we can know is actually happening

903
00:50:36,840 --> 00:50:37,719
Speaker 1: between us, right.

904
00:50:37,800 --> 00:50:39,520
Speaker 2: I mean, we are sitting here looking at each other

905
00:50:39,560 --> 00:50:42,759
Speaker 2: in the eyes. We are definitely both humans because as

906
00:50:42,760 --> 00:50:47,799
Speaker 2: far as I know, we don't have human stand ins

907
00:50:47,840 --> 00:50:50,080
Speaker 2: that are believable enough to trick either one of us.

908
00:50:50,239 --> 00:50:52,600
Speaker 1: It's like in Terminator, the first ones you could tell

909
00:50:52,640 --> 00:50:56,600
Speaker 1: the difference, right, So, so what do you have any

910
00:50:56,640 --> 00:51:01,200
Speaker 1: thoughts on the likelihood of this kind of technolog progressing

911
00:51:01,480 --> 00:51:05,719
Speaker 1: or being used to disrupt the spread of reliable information.

912
00:51:05,880 --> 00:51:07,719
Speaker 1: Do you think it's a definite Do you think it

913
00:51:07,800 --> 00:51:12,080
Speaker 1: might happen? Do you think it's a little bit sensationalistic?

914
00:51:12,120 --> 00:51:12,640
Speaker 1: What do you think?

915
00:51:12,800 --> 00:51:15,319
Speaker 2: I think it absolutely could happen. I think it's another

916
00:51:15,400 --> 00:51:20,320
Speaker 2: one of those things that's kind of dangled out there

917
00:51:20,600 --> 00:51:22,840
Speaker 2: like a piece of bait on a fishing hook for

918
00:51:22,960 --> 00:51:26,120
Speaker 2: the American consumer to bite onto and be oh, this

919
00:51:26,160 --> 00:51:28,080
Speaker 2: will change my life for the better. This will make

920
00:51:28,120 --> 00:51:30,640
Speaker 2: it so easy for me to not have to make

921
00:51:30,680 --> 00:51:34,239
Speaker 2: hair appointments or schedule oil changes or what have you. Yeah,

922
00:51:34,280 --> 00:51:36,960
Speaker 2: and then you you know, once the buy in happens,

923
00:51:37,040 --> 00:51:39,560
Speaker 2: then it started. It opens that Pandora's jar that you

924
00:51:39,640 --> 00:51:42,080
Speaker 2: keep talking about. So yeah, no, there's no doubt about it.

925
00:51:42,120 --> 00:51:44,839
Speaker 2: The more access we give our devices and the more

926
00:51:44,920 --> 00:51:48,400
Speaker 2: centralized our information is. And I guess by centralized, I

927
00:51:48,480 --> 00:51:52,760
Speaker 2: just mean within a particular service where it has access

928
00:51:52,800 --> 00:51:55,600
Speaker 2: to all of it. And that's what that's what it takes. Right,

929
00:51:56,120 --> 00:51:58,200
Speaker 2: Why does series suck? Okay, let's just I just want to get

930
00:51:58,200 --> 00:51:59,680
Speaker 2: into this because I was a rag on Apple for

931
00:51:59,680 --> 00:52:03,280
Speaker 2: a minute. Siri is not a good personal assistant because

932
00:52:03,320 --> 00:52:07,640
Speaker 2: it doesn't learn. It doesn't keep track of what you say.

933
00:52:07,719 --> 00:52:10,920
Speaker 2: It's starting from scratch every time you ask it a question. Right,

934
00:52:11,000 --> 00:52:13,400
Speaker 2: This is Siri, Apple's personal assistant, which is one of

935
00:52:13,440 --> 00:52:17,120
Speaker 2: the first ones to hit the market but fizzled ridiculously

936
00:52:17,200 --> 00:52:21,680
Speaker 2: because of Google's and Amazon's Alexa and stuff. Because those

937
00:52:22,280 --> 00:52:25,359
Speaker 2: actually learn and are connected to the Internet and learn

938
00:52:25,400 --> 00:52:29,200
Speaker 2: your vocal patterns and learn your preferences and have access

939
00:52:29,239 --> 00:52:32,640
Speaker 2: to that bigger network, whereas Siri is just kind of dumb.

940
00:52:32,760 --> 00:52:33,840
Speaker 2: It doesn't really do that.

941
00:52:34,120 --> 00:52:36,240
Speaker 1: They can tell you stuff about Apple, you can.

942
00:52:36,120 --> 00:52:37,719
Speaker 2: Tell you a lot of stuff about the history of

943
00:52:38,120 --> 00:52:40,759
Speaker 2: the company, and it can do Google searches for you.

944
00:52:40,800 --> 00:52:44,320
Speaker 2: But if you said, hey, Siri, what time is this movie?

945
00:52:44,600 --> 00:52:48,440
Speaker 2: It'll probably give you some tangentially related answers, but not

946
00:52:48,600 --> 00:52:49,719
Speaker 2: the exact answer you want.

947
00:52:49,960 --> 00:52:52,359
Speaker 1: Time was invented shortly after The Big Bang.

948
00:52:52,560 --> 00:52:54,480
Speaker 2: I say, Siri, tell me a joke. That's always fun.

949
00:52:54,640 --> 00:52:58,200
Speaker 2: But no, but Google assistants will do that because it

950
00:52:58,320 --> 00:53:01,480
Speaker 2: learns your habits and figure years out what kinds of

951
00:53:01,560 --> 00:53:04,120
Speaker 2: questions you're asking. So that's what I'm saying. When we

952
00:53:04,160 --> 00:53:07,719
Speaker 2: start getting into that where we're tapped into this larger network,

953
00:53:08,320 --> 00:53:11,240
Speaker 2: it does seem like a slippery slope. But yeah, people,

954
00:53:11,280 --> 00:53:13,839
Speaker 2: even the most paranoid of my friends, some of them,

955
00:53:13,840 --> 00:53:15,680
Speaker 2: are all about these home assistants. And it's thing, this

956
00:53:15,719 --> 00:53:18,000
Speaker 2: conversation we always have where it's like, I'll give it

957
00:53:18,080 --> 00:53:20,879
Speaker 2: up if I can tell a little box to turn

958
00:53:20,920 --> 00:53:21,719
Speaker 2: on my lights for me.

959
00:53:22,239 --> 00:53:25,759
Speaker 1: Ah. Yeah, I was given one of those things, and

960
00:53:25,800 --> 00:53:28,480
Speaker 1: I plug it in when I'm cooking and unplug it afterwards.

961
00:53:28,800 --> 00:53:31,840
Speaker 1: So it probably thinks that I am always cooking, or

962
00:53:32,200 --> 00:53:35,480
Speaker 1: just don't have power at my house, which I'm fine with,

963
00:53:35,600 --> 00:53:39,719
Speaker 1: But we want to hear from you, assuming again to

964
00:53:39,800 --> 00:53:42,640
Speaker 1: the earlier question, is really us talking to you now?

965
00:53:42,680 --> 00:53:44,240
Speaker 1: What would you like to tell us? Are the robot

966
00:53:44,360 --> 00:53:48,880
Speaker 1: versions of us impersonating the biological versions of us? Do

967
00:53:48,920 --> 00:53:52,920
Speaker 1: you have a home assistant? Do you use this facial

968
00:53:52,960 --> 00:53:56,920
Speaker 1: recognition software that is so often marketed as a free

969
00:53:57,440 --> 00:54:02,400
Speaker 1: recreational pursuit. Do you believe that the future will be

970
00:54:02,640 --> 00:54:07,759
Speaker 1: fraught with things that make us question digital reality? Is

971
00:54:07,800 --> 00:54:10,480
Speaker 1: there a way to combat it? Should there be? You

972
00:54:10,520 --> 00:54:13,640
Speaker 1: can find us on Instagram. You can find us on Facebook.

973
00:54:13,680 --> 00:54:16,440
Speaker 1: You can find us on Twitter. You can find some

974
00:54:16,640 --> 00:54:19,000
Speaker 1: version of us. I feel like now we have to

975
00:54:19,040 --> 00:54:21,320
Speaker 1: say you can find some version of us on there.

976
00:54:21,640 --> 00:54:24,200
Speaker 1: You can also, of course, visit one of our favorite

977
00:54:24,200 --> 00:54:26,920
Speaker 1: places on the internet, our Facebook page. Here's where it

978
00:54:26,920 --> 00:54:32,040
Speaker 1: gets crazy, where you can see Matt Knowle and I, well, what.

979
00:54:31,920 --> 00:54:34,400
Speaker 2: Do we do there? Where on the internet?

980
00:54:34,920 --> 00:54:35,759
Speaker 1: Here's where it gets great?

981
00:54:35,880 --> 00:54:39,719
Speaker 2: Oh, all kinds of stuff. We lurk sometimes sometimes we

982
00:54:39,840 --> 00:54:43,120
Speaker 2: respond and hop around in the threads and post our

983
00:54:43,120 --> 00:54:45,839
Speaker 2: own little gifts and memes, and we have a good

984
00:54:45,840 --> 00:54:47,640
Speaker 2: time in there. It's a lot of fun and.

985
00:54:48,360 --> 00:54:51,319
Speaker 3: That's the end of this classic episode. If you have

986
00:54:51,400 --> 00:54:55,480
Speaker 3: any thoughts or questions about this episode, you can get

987
00:54:55,520 --> 00:54:57,920
Speaker 3: into contact with us in a number of different ways.

988
00:54:58,120 --> 00:54:59,680
Speaker 3: One of the best is to give us a call

989
00:54:59,760 --> 00:55:03,960
Speaker 3: or no is one eight three three st d WYTK.

990
00:55:04,360 --> 00:55:06,160
Speaker 3: If you don't want to do that, you can send

991
00:55:06,239 --> 00:55:07,640
Speaker 3: us a good old fashioned email.

992
00:55:07,880 --> 00:55:12,000
Speaker 1: We are conspiracy at iHeartRadio dot com.

993
00:55:12,200 --> 00:55:14,240
Speaker 3: Stuff they Don't want you to Know is a production

994
00:55:14,360 --> 00:55:18,879
Speaker 3: of iHeartRadio. For more podcasts from iHeartRadio, visit the iHeartRadio app,

995
00:55:18,960 --> 00:55:21,840
Speaker 3: Apple Podcasts, or wherever you listen to your favorite shows.