1
00:00:08,039 --> 00:00:11,400
Speaker 1: Welcome to another episode of Strictly Business, the podcast in

2
00:00:11,400 --> 00:00:13,720
Speaker 1: which we speak with some of the brightest minds working

3
00:00:13,760 --> 00:00:17,960
Speaker 1: in the media business today. I'm Andrew Wallenstein with Variety.

4
00:00:18,960 --> 00:00:22,079
Speaker 1: With a name like Soul Machines, it's either one of

5
00:00:22,120 --> 00:00:25,920
Speaker 1: two things, some long forgotten seventies R and B group

6
00:00:26,520 --> 00:00:31,639
Speaker 1: or a twenty first century startup creating virtual clones of celebrities. Well,

7
00:00:31,840 --> 00:00:35,080
Speaker 1: we've got the latter on here today on Strictly Business.

8
00:00:35,120 --> 00:00:38,040
Speaker 1: Soul Machines CEO Greg Cross is here to talk about

9
00:00:38,280 --> 00:00:41,199
Speaker 1: a fascinating venture that is marrying state of the art

10
00:00:41,280 --> 00:00:45,599
Speaker 1: animation with artificial intelligence to create something that just might

11
00:00:45,640 --> 00:00:48,199
Speaker 1: be the next big thing in entertainment. We'll be back

12
00:00:48,240 --> 00:00:53,400
Speaker 1: with Greg in just a moment. Hi, this is Andrew

13
00:00:53,440 --> 00:00:58,840
Speaker 1: Wallenstein with Variety's Strictly Business podcast. If you love Varieties podcasts,

14
00:00:58,880 --> 00:01:03,320
Speaker 1: you're going to want to try Variety Intelligence Platform or VIP.

15
00:01:04,240 --> 00:01:07,200
Speaker 1: It's a digital subscription tier on Variety dot com for

16
00:01:07,319 --> 00:01:11,600
Speaker 1: industry professionals to dig deeper into analysis and data that

17
00:01:11,680 --> 00:01:15,360
Speaker 1: helps them be smarter about their business. VIP just launched

18
00:01:15,400 --> 00:01:19,000
Speaker 1: a great new newsletter and offers more special reports than ever,

19
00:01:19,480 --> 00:01:23,720
Speaker 1: So visit Variety dot com. Slash vip save for a

20
00:01:23,840 --> 00:01:30,640
Speaker 1: twenty percent discount. We are back with Greg Cross, CEO

21
00:01:30,920 --> 00:01:34,160
Speaker 1: of Soul Machines, with the first of what was such

22
00:01:34,240 --> 00:01:38,120
Speaker 1: a fascinating interview. It lasted a full hour, so I

23
00:01:38,200 --> 00:01:41,760
Speaker 1: broke it into two thirty minute sections to hear the

24
00:01:41,800 --> 00:01:45,000
Speaker 1: second half hour. Make sure to tune into Strictly Business

25
00:01:45,120 --> 00:01:48,520
Speaker 1: next week. But without further ado, we've got the first

26
00:01:48,560 --> 00:01:52,880
Speaker 1: thirty minutes ready to go right now. Greg, thanks for

27
00:01:52,880 --> 00:01:53,320
Speaker 1: coming in.

28
00:01:53,520 --> 00:01:55,920
Speaker 2: Hi, great to be spending some time with you, and

29
00:01:55,960 --> 00:01:57,800
Speaker 2: you listen to this, it's fantastic great.

30
00:01:58,320 --> 00:02:00,000
Speaker 1: So, you know, this is one of these moments where

31
00:02:00,280 --> 00:02:03,520
Speaker 1: I don't feel like the podcast format does the interview

32
00:02:03,800 --> 00:02:07,480
Speaker 1: justice because you really need to see a Soul Machines

33
00:02:07,560 --> 00:02:10,880
Speaker 1: creation in action to truly get what it's about. But

34
00:02:10,960 --> 00:02:13,160
Speaker 1: you know, we'll soldier on, and the best thing you

35
00:02:13,160 --> 00:02:15,040
Speaker 1: could do is just start us off with a very

36
00:02:15,080 --> 00:02:18,800
Speaker 1: basic explanation of what it is that Sole Machines does.

37
00:02:18,960 --> 00:02:22,080
Speaker 2: Yeah, I mean kind of the name says it at all, really,

38
00:02:22,200 --> 00:02:25,840
Speaker 2: So you know, when I co founded the company, we

39
00:02:25,840 --> 00:02:28,000
Speaker 2: were looking for a name, and you know, the thing

40
00:02:28,000 --> 00:02:30,560
Speaker 2: that we kept coming back to is putting some soul

41
00:02:30,720 --> 00:02:33,880
Speaker 2: into the machines of our future. So you know, I

42
00:02:33,919 --> 00:02:36,240
Speaker 2: love brand names that are quite literal in terms of

43
00:02:36,400 --> 00:02:39,239
Speaker 2: you know what it's all about. So you know, as

44
00:02:39,280 --> 00:02:42,320
Speaker 2: a company, you know, we do a lot of research

45
00:02:42,360 --> 00:02:47,960
Speaker 2: in a very specialized field of artificial intelligence called cognitive modeling.

46
00:02:48,760 --> 00:02:51,760
Speaker 2: So and this is a really enables us to bring

47
00:02:51,800 --> 00:02:55,079
Speaker 2: to life really really advanced form of animation. So that

48
00:02:55,840 --> 00:02:59,520
Speaker 2: cgi characters which we got quite used to know an

49
00:02:59,680 --> 00:03:02,839
Speaker 2: l I the entertainment business, but instead of them being

50
00:03:02,919 --> 00:03:08,120
Speaker 2: animated by pre recorded content and you know this concept

51
00:03:08,160 --> 00:03:11,960
Speaker 2: of linear storytelling, you know, we're literally bringing digital characters

52
00:03:12,040 --> 00:03:16,600
Speaker 2: to life in real time. And we call it autonomous animation.

53
00:03:16,880 --> 00:03:19,040
Speaker 2: And this is what you know, you and I are

54
00:03:19,080 --> 00:03:21,840
Speaker 2: both doing here as we record this podcast. Your brain

55
00:03:21,960 --> 00:03:25,359
Speaker 2: is animating you and bringing you to life in this instance,

56
00:03:25,440 --> 00:03:28,280
Speaker 2: and mine's doing exactly the same. So you know, so

57
00:03:28,760 --> 00:03:31,400
Speaker 2: it's literally, like I mean to paint a picture, it's

58
00:03:31,440 --> 00:03:35,680
Speaker 2: literally like having a zoom conversation or a video conference

59
00:03:35,720 --> 00:03:39,360
Speaker 2: conversation with a digital person rather than a real person.

60
00:03:39,560 --> 00:03:42,480
Speaker 1: So let's give an example with one of your clients.

61
00:03:42,520 --> 00:03:46,320
Speaker 1: There's an interesting mix of celebrities both dead and alive

62
00:03:47,080 --> 00:03:51,000
Speaker 1: in that portfolio. So I for instance, noticed Carmelo Anthony,

63
00:03:51,040 --> 00:03:54,720
Speaker 1: the recently retired basketball player. What can I do with

64
00:03:54,880 --> 00:03:55,960
Speaker 1: virtual Carmelo.

65
00:03:56,480 --> 00:04:01,200
Speaker 2: Yeah, so you know, we started working with Carmalo last

66
00:04:01,320 --> 00:04:06,480
Speaker 2: year and so you know, Digital Camelo. There's a number

67
00:04:06,480 --> 00:04:09,600
Speaker 2: of different types of experiences that we're looking to create

68
00:04:09,680 --> 00:04:13,040
Speaker 2: for for Camalo and his business interests as he moves

69
00:04:13,080 --> 00:04:16,719
Speaker 2: into the sort of into his post playing career. You

70
00:04:16,839 --> 00:04:20,680
Speaker 2: think about it. You know, when you've got an incredible

71
00:04:20,720 --> 00:04:24,080
Speaker 2: athlete like Camelo had a long and successful career, you know,

72
00:04:24,320 --> 00:04:26,480
Speaker 2: how does he stay in touch with his fans now?

73
00:04:26,720 --> 00:04:29,800
Speaker 2: You know now that he's not playing basketball on a

74
00:04:30,600 --> 00:04:33,400
Speaker 2: day to day basis. So for us, this becomes what

75
00:04:33,480 --> 00:04:37,640
Speaker 2: we think of as the future of fan experiences. You know,

76
00:04:37,920 --> 00:04:41,600
Speaker 2: we've all watched Camelo play and amazing basketball games, but

77
00:04:41,839 --> 00:04:45,119
Speaker 2: not many of us get to meet him in real life.

78
00:04:45,480 --> 00:04:48,480
Speaker 2: So you think about this way of how we can

79
00:04:48,520 --> 00:04:52,000
Speaker 2: democratize a fan experience and and you can go to

80
00:04:52,279 --> 00:04:54,880
Speaker 2: a website and you could have a face to face

81
00:04:55,000 --> 00:04:57,920
Speaker 2: conversation with what we call we refer to him as

82
00:04:57,960 --> 00:05:01,159
Speaker 2: Digital Mallow as you know, as what we talk about

83
00:05:01,520 --> 00:05:08,279
Speaker 2: as a digital twin. So Digital Mellow, you know, might

84
00:05:08,320 --> 00:05:14,159
Speaker 2: become the brand ambassador of of you know, of a

85
00:05:14,200 --> 00:05:17,560
Speaker 2: big automotive car company for example. And so you know,

86
00:05:17,839 --> 00:05:23,080
Speaker 2: think about digital Mellow being accessible from their e commerce

87
00:05:23,120 --> 00:05:27,240
Speaker 2: store or their website or their smartphone app, and you know,

88
00:05:27,360 --> 00:05:31,200
Speaker 2: he can create this sort of one on one brand experience.

89
00:05:31,320 --> 00:05:35,080
Speaker 1: So Carmelo Virtual Carmelo could sell me a car, he

90
00:05:35,160 --> 00:05:38,480
Speaker 1: could talk to me about the attributes and I don't know,

91
00:05:38,560 --> 00:05:41,320
Speaker 1: maybe strike a deal. And to make it clear, the

92
00:05:41,480 --> 00:05:45,120
Speaker 1: real Carmelo has he's not out the control, he's not

93
00:05:45,200 --> 00:05:49,080
Speaker 1: putting words in his mouth. What he the real Carmelo

94
00:05:49,320 --> 00:05:52,120
Speaker 1: is giving you the rates to do this. But that's

95
00:05:52,120 --> 00:05:52,800
Speaker 1: where it ats.

96
00:05:52,920 --> 00:05:55,559
Speaker 2: Yeah, yeah, and you know we're in a joint venture

97
00:05:55,560 --> 00:05:59,080
Speaker 2: business with Camelo, so you know, so because this is

98
00:05:59,120 --> 00:06:02,440
Speaker 2: a this is a new, really really new type of technology.

99
00:06:02,480 --> 00:06:06,279
Speaker 2: So you know, we're reimagining the way in which fans

100
00:06:06,320 --> 00:06:09,719
Speaker 2: can experience there. You know, they're superheroes there, you know,

101
00:06:09,800 --> 00:06:12,440
Speaker 2: the the you know, the the people that they love

102
00:06:12,520 --> 00:06:16,680
Speaker 2: to watch. So so yeah, real Camelo Anthony. We know

103
00:06:16,800 --> 00:06:21,440
Speaker 2: when we set about creating you know, Digital Camelo Anthony,

104
00:06:21,720 --> 00:06:24,479
Speaker 2: we you know, we know we had some access to

105
00:06:24,480 --> 00:06:28,920
Speaker 2: Camelo for a few hours to capture his likeness and

106
00:06:29,000 --> 00:06:33,600
Speaker 2: his personality, all of which becomes part of the the

107
00:06:33,640 --> 00:06:37,400
Speaker 2: AI systems that we've created. We train an AI voice.

108
00:06:37,520 --> 00:06:42,080
Speaker 2: So a really interesting thing about Digital Mallow is he

109
00:06:42,120 --> 00:06:45,839
Speaker 2: can speak any language fluently. So you know, you think

110
00:06:45,839 --> 00:06:48,640
Speaker 2: about that. You know, one of the biggest NBA markets

111
00:06:48,640 --> 00:06:52,080
Speaker 2: in the world now is China. Yes, so imagine a

112
00:06:52,120 --> 00:06:56,680
Speaker 2: world where Digital Mallow is a brand ambassador for a

113
00:06:56,760 --> 00:07:02,320
Speaker 2: Chinese car company speaking Mandarin to fans of the NBA

114
00:07:02,400 --> 00:07:07,560
Speaker 2: in China. So you know, Digital Mellow could become a

115
00:07:07,640 --> 00:07:10,480
Speaker 2: commentator on you know, what's going on in the world

116
00:07:10,480 --> 00:07:13,280
Speaker 2: of basketball at any point in time, and he can

117
00:07:13,320 --> 00:07:18,640
Speaker 2: interact with fans, you know from Syracuse, you know, the

118
00:07:18,640 --> 00:07:21,560
Speaker 2: biggest Mellow fans in the world, the Syracuse fans. Or

119
00:07:21,600 --> 00:07:25,200
Speaker 2: he could interact with, as I said, fans in Europe

120
00:07:25,960 --> 00:07:28,440
Speaker 2: in there in the native language. So I mean, these

121
00:07:28,480 --> 00:07:31,640
Speaker 2: are some of the amazing opportunities that exist as we

122
00:07:31,720 --> 00:07:34,120
Speaker 2: think about, you know, the way in which we can

123
00:07:34,160 --> 00:07:36,320
Speaker 2: create fan experiences in the future.

124
00:07:36,400 --> 00:07:37,920
Speaker 1: And I want to be clear, like, you know, you're

125
00:07:37,960 --> 00:07:40,560
Speaker 1: not some guy in a in a basement here. You're

126
00:07:40,880 --> 00:07:43,560
Speaker 1: first of all an entrepreneur who has you know, sold

127
00:07:43,600 --> 00:07:48,120
Speaker 1: companies to Apple and been around and you've raised well

128
00:07:48,160 --> 00:07:52,160
Speaker 1: over one hundred million for this business. You've got representation

129
00:07:52,280 --> 00:07:57,240
Speaker 1: at c AAA and absolutely and besides Mellow there is

130
00:07:57,600 --> 00:08:00,280
Speaker 1: you've got the likeness of Marilyn Monroe, which I assume

131
00:08:00,320 --> 00:08:02,720
Speaker 1: he negotiated with her estate Like this is real.

132
00:08:02,960 --> 00:08:07,120
Speaker 2: Yeah yeah, so yeah. So from a company point of view,

133
00:08:07,120 --> 00:08:10,160
Speaker 2: the sort of two of us founded the company. So

134
00:08:10,240 --> 00:08:14,200
Speaker 2: my co founder, doctor Mark Sega, came out of the

135
00:08:14,520 --> 00:08:20,200
Speaker 2: incredible Wetter Digital company. Mark won technical Oscars for the

136
00:08:20,240 --> 00:08:23,119
Speaker 2: animation work that he did on the original Avatar movie.

137
00:08:23,160 --> 00:08:26,560
Speaker 2: So if we think back to that movie, it was

138
00:08:27,280 --> 00:08:30,440
Speaker 2: one of the first times that we've seen CGI characters

139
00:08:30,480 --> 00:08:34,440
Speaker 2: capable of micro human expressions. We fell in love with

140
00:08:34,480 --> 00:08:38,520
Speaker 2: those giant blue characters because they expressed emotion. We connected

141
00:08:38,600 --> 00:08:41,480
Speaker 2: them because they expressed an emotion in an incredible way.

142
00:08:41,520 --> 00:08:44,280
Speaker 2: Mark then went on to do Kong with su Peter

143
00:08:44,400 --> 00:08:46,400
Speaker 2: Jackson using the same sort of technology, and then he

144
00:08:46,400 --> 00:08:49,520
Speaker 2: set up a research lab at the University of Auckland

145
00:08:49,520 --> 00:08:53,679
Speaker 2: in twenty twelve and started building the digital brain models

146
00:08:53,720 --> 00:08:58,040
Speaker 2: at underpin the work that we do today. So I

147
00:08:58,080 --> 00:09:01,600
Speaker 2: met Mark in twenty sixteen bun the company out of

148
00:09:02,480 --> 00:09:05,520
Speaker 2: the university at that point, and we've raised over one

149
00:09:05,559 --> 00:09:08,760
Speaker 2: hundred and thirty five million dollars and venture capital from

150
00:09:08,880 --> 00:09:12,079
Speaker 2: some of the from some of the leading venture funds

151
00:09:12,120 --> 00:09:15,200
Speaker 2: in the world. Soft Bank is an investor, Tamasik, the

152
00:09:15,240 --> 00:09:20,199
Speaker 2: big Singapore global investor Horizons, Selena Chow, and the people

153
00:09:20,200 --> 00:09:23,319
Speaker 2: that have invested in companies like Spotify and Zoom, to

154
00:09:23,400 --> 00:09:25,320
Speaker 2: name a couple. So yeah, I mean, we you know,

155
00:09:25,360 --> 00:09:28,440
Speaker 2: we've you know, this is really really deep tech research

156
00:09:28,520 --> 00:09:32,839
Speaker 2: and you know, and it's attracting the interests of some

157
00:09:32,880 --> 00:09:35,280
Speaker 2: of the biggest celebrities in the world. You talked about

158
00:09:35,640 --> 00:09:41,360
Speaker 2: the project we have with Maryland. The digital rights of

159
00:09:41,400 --> 00:09:44,880
Speaker 2: the name, likeness and image of Maryland are held by

160
00:09:44,960 --> 00:09:48,080
Speaker 2: Authentic Brands Group for a bigger entertainment group I guess

161
00:09:48,080 --> 00:09:51,719
Speaker 2: headquartered out of New York, and actually they own a

162
00:09:51,840 --> 00:09:55,120
Speaker 2: number of you know, the name like images of some

163
00:09:55,200 --> 00:09:59,480
Speaker 2: like pretty iconic legends, you know, Muhammadali, Elvis Presley. So

164
00:10:00,320 --> 00:10:03,120
Speaker 2: Marilyn Monroe is one of the next projects we're literally

165
00:10:03,200 --> 00:10:07,640
Speaker 2: just about to start working on and bringing life to

166
00:10:07,720 --> 00:10:12,239
Speaker 2: that project. So Mark Twan and another one of our celebrities,

167
00:10:12,640 --> 00:10:15,920
Speaker 2: and we've chosen different celebrities from different fields. So Mark

168
00:10:16,240 --> 00:10:20,120
Speaker 2: is a K pop superstar, you know, in the world

169
00:10:20,160 --> 00:10:22,439
Speaker 2: of music, you know, I mean Mark is you know,

170
00:10:22,480 --> 00:10:25,560
Speaker 2: it might not be as well known in America, but

171
00:10:26,040 --> 00:10:30,360
Speaker 2: he's one of CIA's biggest musicians that they rap. And

172
00:10:30,480 --> 00:10:31,280
Speaker 2: so where does a.

173
00:10:31,320 --> 00:10:33,680
Speaker 1: K pop fan encounter the virtual Mark?

174
00:10:34,480 --> 00:10:37,800
Speaker 2: How literally encounter them? All over the world, you know,

175
00:10:37,920 --> 00:10:40,480
Speaker 2: K pop is, I mean it is a global phenomenon.

176
00:10:40,520 --> 00:10:44,120
Speaker 2: But you know guys like Mark will full of will

177
00:10:44,160 --> 00:10:48,280
Speaker 2: full of soccer stadium in the Philippines or in Jakata

178
00:10:49,559 --> 00:10:52,440
Speaker 2: just to hear them answer questions on a stage, not perform,

179
00:10:52,640 --> 00:10:55,960
Speaker 2: just ask questions. So imagine a world where they can,

180
00:10:56,400 --> 00:10:58,679
Speaker 2: you know, they can sit in a live stream and

181
00:10:58,760 --> 00:11:03,760
Speaker 2: ask questions directly to digital marked one in their own language.

182
00:11:03,960 --> 00:11:06,240
Speaker 2: I mean, that's the sort of magic we envisage for

183
00:11:06,280 --> 00:11:06,720
Speaker 2: the future.

184
00:11:07,200 --> 00:11:09,839
Speaker 1: And is this and maybe this isn't an either or

185
00:11:09,880 --> 00:11:12,440
Speaker 1: a question, but I'll ask it, is this meant to

186
00:11:12,520 --> 00:11:16,040
Speaker 1: be a one to one interaction or a one to

187
00:11:16,240 --> 00:11:17,360
Speaker 1: many interaction?

188
00:11:17,679 --> 00:11:20,959
Speaker 2: Both? But it's actually I mean, our technology is all

189
00:11:21,000 --> 00:11:24,440
Speaker 2: about one to one interaction, you know. So you know,

190
00:11:24,640 --> 00:11:29,520
Speaker 2: human interaction is about emotional connection. Sure, so we communicate

191
00:11:29,559 --> 00:11:31,640
Speaker 2: in lots of different ways as human beings, but the

192
00:11:31,640 --> 00:11:35,960
Speaker 2: most you know, the most personal, the most connective interactions

193
00:11:36,000 --> 00:11:38,640
Speaker 2: are those one to one, you know, one to one

194
00:11:38,720 --> 00:11:44,320
Speaker 2: face to face interactions where you're not just hearing words,

195
00:11:44,440 --> 00:11:49,560
Speaker 2: you're actually understanding, you know, the full all of the

196
00:11:49,600 --> 00:11:54,160
Speaker 2: elements of that nonverbal communication. So an emotional connection is

197
00:11:54,160 --> 00:11:59,040
Speaker 2: incredibly important to us. We remember things that we connect,

198
00:11:59,040 --> 00:12:02,280
Speaker 2: you know, and people that we're emotionally connect with. So yeah,

199
00:12:02,520 --> 00:12:04,800
Speaker 2: it is for us a lot of this about one

200
00:12:04,880 --> 00:12:08,160
Speaker 2: to one interaction. We also do a huge amount of

201
00:12:08,200 --> 00:12:11,480
Speaker 2: work in the in the in the brand space, so

202
00:12:11,600 --> 00:12:17,559
Speaker 2: creating digital ambassadors and digital influences for big brands, General Motors,

203
00:12:18,160 --> 00:12:24,320
Speaker 2: Cadillac are examples of a client of ours in the

204
00:12:24,320 --> 00:12:31,679
Speaker 2: financial space JP Morgan, Chase, MasterCard, Fidelity, and so this

205
00:12:31,760 --> 00:12:37,000
Speaker 2: is helping big brands imagine how they can create an

206
00:12:37,000 --> 00:12:42,240
Speaker 2: emotional connection in an increasingly digital world, because if you

207
00:12:42,320 --> 00:12:46,880
Speaker 2: think about brand connections in the digital world of e commerce,

208
00:12:48,040 --> 00:12:51,839
Speaker 2: they've been marginalized because in many respects, e commerce is

209
00:12:51,840 --> 00:12:57,719
Speaker 2: about efficient transactions, where brands historically have always created these

210
00:12:57,760 --> 00:13:01,040
Speaker 2: emotional connections using different forms of me media, whether it's

211
00:13:01,080 --> 00:13:05,199
Speaker 2: print media, whether it's television, as advertising as a media,

212
00:13:05,240 --> 00:13:09,600
Speaker 2: but brand values. The value of a brand is measured

213
00:13:09,600 --> 00:13:12,080
Speaker 2: by the level of emotional connection they can have with

214
00:13:12,120 --> 00:13:15,680
Speaker 2: the target audiences, and this provides that one to one

215
00:13:16,040 --> 00:13:19,280
Speaker 2: personalized brand experience in a digital world.

216
00:13:20,400 --> 00:13:23,960
Speaker 1: So the thing that concerns me, though, and I wonder

217
00:13:24,040 --> 00:13:27,600
Speaker 1: if concerns you know, the people you pitch your services

218
00:13:27,640 --> 00:13:30,920
Speaker 1: to that you know, maybe they're agents of celebrities. Is

219
00:13:31,320 --> 00:13:34,240
Speaker 1: I feel like we're always hearing stories about let's call

220
00:13:34,280 --> 00:13:37,960
Speaker 1: it technology going rogue. I remember there was a virtual

221
00:13:38,120 --> 00:13:42,520
Speaker 1: artist fan Mecha last year who sounded somewhat similar in

222
00:13:42,520 --> 00:13:46,960
Speaker 1: this kind of technology spouted racist things. I feel like

223
00:13:47,000 --> 00:13:50,160
Speaker 1: it's almost inevitable. We always hear these things. Do you

224
00:13:50,480 --> 00:13:53,439
Speaker 1: encounter these concerns and what do you say to those

225
00:13:53,520 --> 00:13:54,440
Speaker 1: you're trying to win over?

226
00:13:54,720 --> 00:13:58,160
Speaker 2: Yeah, I mean, look, you know this debate we you know,

227
00:13:58,200 --> 00:14:01,280
Speaker 2: I mean artificial intelligence. I mean last year was something

228
00:14:01,360 --> 00:14:03,560
Speaker 2: that was in the world of technology, in the world

229
00:14:03,559 --> 00:14:07,560
Speaker 2: of research labs, and now artificial intelligence that thanks to chat,

230
00:14:07,600 --> 00:14:10,560
Speaker 2: gpt is and app we all have on our smartphones now,

231
00:14:10,679 --> 00:14:15,520
Speaker 2: so we all use even though we artificial intelligence touched

232
00:14:15,600 --> 00:14:18,400
Speaker 2: us every day previously, and if we bought something on

233
00:14:18,440 --> 00:14:22,080
Speaker 2: Amazon AI was driving the algorithms in the background, or

234
00:14:22,120 --> 00:14:25,720
Speaker 2: on social media, it's driving the advertising algorithms in the background.

235
00:14:25,760 --> 00:14:29,120
Speaker 2: So we weren't making a choice to engage with AI

236
00:14:29,560 --> 00:14:33,200
Speaker 2: as consumers. Now we are, So you know, that debate

237
00:14:33,240 --> 00:14:36,040
Speaker 2: has really really come to the forefront. Is you know,

238
00:14:36,280 --> 00:14:38,320
Speaker 2: are the robots coming? Are they going to kill us?

239
00:14:38,320 --> 00:14:40,880
Speaker 2: Are they going to rule over us? And I guess

240
00:14:40,880 --> 00:14:43,120
Speaker 2: we're you know, so there's two parts of this. So one,

241
00:14:43,280 --> 00:14:46,560
Speaker 2: I think the debate's really really good. It's great to

242
00:14:46,680 --> 00:14:51,320
Speaker 2: have both sides of that debate, and it's great to

243
00:14:51,320 --> 00:14:54,000
Speaker 2: have so many people engaging in that debate because when

244
00:14:54,040 --> 00:14:58,480
Speaker 2: you think about the last big technology wave that came along,

245
00:14:58,480 --> 00:15:01,440
Speaker 2: which was the social media wave, you know, we just

246
00:15:01,880 --> 00:15:04,120
Speaker 2: came along for the right, you know, and then we

247
00:15:04,280 --> 00:15:07,760
Speaker 2: learned about you know, the perils of fate news and

248
00:15:07,840 --> 00:15:09,640
Speaker 2: all of the sort that kind of happened to us.

249
00:15:09,640 --> 00:15:13,480
Speaker 2: So as we move into this world where artificial intelligence

250
00:15:13,600 --> 00:15:16,440
Speaker 2: is becoming a bigger and bigger part of it, I

251
00:15:16,520 --> 00:15:18,960
Speaker 2: think it's a fantastic thing that there is so much

252
00:15:19,040 --> 00:15:23,000
Speaker 2: debate going on and people are expressing their concern because

253
00:15:23,000 --> 00:15:26,320
Speaker 2: that puts you know, an absolute focus on you know,

254
00:15:26,480 --> 00:15:29,480
Speaker 2: providers like us to make it safe you know, and

255
00:15:29,920 --> 00:15:32,800
Speaker 2: you know and to you know, and so you've seen

256
00:15:32,960 --> 00:15:35,680
Speaker 2: worth you know, as we've all started to use these

257
00:15:35,760 --> 00:15:40,120
Speaker 2: large language models. They you know, I mean, they hallucinate,

258
00:15:40,280 --> 00:15:42,640
Speaker 2: so they make stuff up because they're just predicting what

259
00:15:42,680 --> 00:15:44,680
Speaker 2: word comes next, So they make stuff up from time

260
00:15:44,760 --> 00:15:48,320
Speaker 2: to time. But the rate of learning, you know, is incredible.

261
00:15:48,400 --> 00:15:52,600
Speaker 2: The safeguards that are there that have been built into that,

262
00:15:52,680 --> 00:15:56,600
Speaker 2: and the safeguards that you can train into it, you know,

263
00:15:56,720 --> 00:16:00,440
Speaker 2: quite substantial, you know. Within you know, the large language

264
00:16:00,440 --> 00:16:02,880
Speaker 2: models and chat GPT sort of hit the market really,

265
00:16:02,920 --> 00:16:05,440
Speaker 2: I guess at the beginning of this year, and already

266
00:16:05,440 --> 00:16:09,880
Speaker 2: we've got something you know, evolving called a private language model. Yeah.

267
00:16:10,000 --> 00:16:13,280
Speaker 2: So a private language model is where you can train

268
00:16:14,040 --> 00:16:18,600
Speaker 2: ah if you're like a closed version of knowledge or content.

269
00:16:18,800 --> 00:16:22,960
Speaker 2: So you know, a digital Camelo Anthony can only speak

270
00:16:23,200 --> 00:16:27,080
Speaker 2: about the content you train it on, but I can

271
00:16:27,080 --> 00:16:31,000
Speaker 2: still speak in an incredibly conversational way. So you know,

272
00:16:31,080 --> 00:16:34,120
Speaker 2: then this idea of having to build safeguards and you know,

273
00:16:34,520 --> 00:16:37,240
Speaker 2: and keep things careful. And then in the example you gave,

274
00:16:37,560 --> 00:16:41,160
Speaker 2: you know, I can remember reading about it, and you know,

275
00:16:41,680 --> 00:16:46,240
Speaker 2: I think you know, you know, they didn't really think

276
00:16:46,360 --> 00:16:51,040
Speaker 2: through the ethics of what they were doing and doing

277
00:16:51,080 --> 00:16:53,560
Speaker 2: it and you know, doing it and executing it in

278
00:16:53,560 --> 00:16:56,160
Speaker 2: the most responsible way. I think, you know, from you know,

279
00:16:56,200 --> 00:16:59,200
Speaker 2: they got you know, you know, I think a pretty

280
00:16:59,200 --> 00:17:01,880
Speaker 2: heavily criticized My memory serves me, right, you know, by

281
00:17:02,000 --> 00:17:05,040
Speaker 2: being a bunch of white tech bros, you know, trying

282
00:17:05,040 --> 00:17:08,840
Speaker 2: to build a you know, an African American American rapper

283
00:17:09,320 --> 00:17:12,160
Speaker 2: and so got into all sorts of cultural and sensitive

284
00:17:12,240 --> 00:17:14,560
Speaker 2: is insensitivities and things like that, and you know, very

285
00:17:14,600 --> 00:17:17,960
Speaker 2: very quickly got taken down. So in the world in

286
00:17:18,000 --> 00:17:20,159
Speaker 2: which we live, you know, you have to be incredibly

287
00:17:20,160 --> 00:17:22,760
Speaker 2: thoughtful about how you engage you know, how we think

288
00:17:22,800 --> 00:17:25,480
Speaker 2: about it. I mean, there are areas of business that

289
00:17:25,520 --> 00:17:27,360
Speaker 2: we don't want to be involved in. I don't want

290
00:17:27,440 --> 00:17:31,120
Speaker 2: to produce digital twins and politicians, Okay, I mean that's

291
00:17:31,200 --> 00:17:33,320
Speaker 2: that's that's an example, I mean, because I mean we

292
00:17:33,400 --> 00:17:36,720
Speaker 2: can look at, you know, what, how fake news has

293
00:17:36,760 --> 00:17:39,520
Speaker 2: impacted the democratic democratic process.

294
00:17:40,520 --> 00:17:43,240
Speaker 1: What about a virtual or a digital twin of our

295
00:17:43,280 --> 00:17:44,440
Speaker 1: news anchor.

296
00:17:45,800 --> 00:17:47,639
Speaker 2: As long I mean, and once again, we don't do

297
00:17:47,760 --> 00:17:50,439
Speaker 2: digital twins of anybody unless we hone the name likeness

298
00:17:50,440 --> 00:17:52,600
Speaker 2: and image and we you know, we have a relationship

299
00:17:52,600 --> 00:17:55,720
Speaker 2: with those persons. But yeah, I mean, yeah, that's I mean,

300
00:17:55,720 --> 00:17:58,600
Speaker 2: that's something that you know, might be you know, might

301
00:17:58,640 --> 00:18:02,400
Speaker 2: be amazing to think about as as we move forward

302
00:18:02,480 --> 00:18:05,000
Speaker 2: into this world of digital content. Because if you think

303
00:18:05,040 --> 00:18:08,879
Speaker 2: about the world of the news anchor that you know,

304
00:18:08,920 --> 00:18:12,080
Speaker 2: and we go back to the television you know, the

305
00:18:12,160 --> 00:18:14,919
Speaker 2: old olden days and the television area where we you know,

306
00:18:14,960 --> 00:18:18,200
Speaker 2: we all tuned into the six o'clock news, the prime

307
00:18:18,280 --> 00:18:21,639
Speaker 2: time news to because this was a trusted face, a

308
00:18:21,680 --> 00:18:24,639
Speaker 2: trusted voice, a source of the truth. You know, you

309
00:18:24,640 --> 00:18:27,520
Speaker 2: could imagine the need for I mean, we do imagine

310
00:18:27,520 --> 00:18:29,720
Speaker 2: the need for you know, we need more trusted voices

311
00:18:29,760 --> 00:18:35,399
Speaker 2: and trusted brands, you know, and that emotional connection. I

312
00:18:35,440 --> 00:18:38,320
Speaker 2: think that's a big part of you know, what's missing

313
00:18:38,320 --> 00:18:40,160
Speaker 2: in some of the digital worlds. So yeah, we think

314
00:18:40,400 --> 00:18:42,719
Speaker 2: we spend a lot of time thinking about how we

315
00:18:42,760 --> 00:18:45,400
Speaker 2: create those types of connections. You know, when you talk

316
00:18:45,480 --> 00:18:49,440
Speaker 2: to digital mallow and we're very clearful you're not we're

317
00:18:49,480 --> 00:18:50,879
Speaker 2: not trying to trick you that you're talking to the

318
00:18:50,920 --> 00:18:53,399
Speaker 2: real Camelo Anthony. This is a digital version.

319
00:18:53,440 --> 00:18:56,040
Speaker 1: I noticed he sort of identifies that he is not

320
00:18:56,320 --> 00:18:57,160
Speaker 1: the real Yeah.

321
00:18:57,760 --> 00:18:59,800
Speaker 2: I mean that's really important because if you think about

322
00:18:59,800 --> 00:19:03,080
Speaker 2: the concept of establishing an emotional connection that's based on

323
00:19:03,119 --> 00:19:06,040
Speaker 2: a level of trust and integrity, and you don't get

324
00:19:06,080 --> 00:19:09,960
Speaker 2: that when you're trying to full somebody that you're the

325
00:19:10,000 --> 00:19:12,680
Speaker 2: real thing when you're not. You know, So the way

326
00:19:12,720 --> 00:19:17,760
Speaker 2: in which we can provide education and in so many

327
00:19:17,760 --> 00:19:21,320
Speaker 2: different fields, AI is already having such a positive impact,

328
00:19:21,359 --> 00:19:26,720
Speaker 2: and so it's always easy to focus on one side

329
00:19:26,720 --> 00:19:29,840
Speaker 2: of the debate and forget about, you know, the amazing

330
00:19:30,400 --> 00:19:32,280
Speaker 2: and the incredible things we do a lot of work

331
00:19:32,320 --> 00:19:35,520
Speaker 2: in as well as working in entertainment with celebrities and

332
00:19:35,520 --> 00:19:37,600
Speaker 2: big brands, we do a lot of work in health

333
00:19:37,600 --> 00:19:39,840
Speaker 2: care and education. And you know, when you stop and

334
00:19:39,880 --> 00:19:42,760
Speaker 2: think about health care and education and sectors, we don't

335
00:19:42,800 --> 00:19:45,800
Speaker 2: have enough real people in those sectors at the moment.

336
00:19:45,840 --> 00:19:48,600
Speaker 2: We don't have enough teachers in many communities, we don't

337
00:19:49,800 --> 00:19:51,959
Speaker 2: you know, in this post COVID era, we don't have

338
00:19:52,080 --> 00:19:56,439
Speaker 2: enough nurses and healthcare professionals. So, I mean, so ways

339
00:19:56,480 --> 00:20:01,760
Speaker 2: in which artificial intelligence, the ways and which sort of

340
00:20:01,760 --> 00:20:05,160
Speaker 2: digital people that we create can make a difference.

341
00:20:08,560 --> 00:20:10,639
Speaker 1: We'll be right back. We're going to talk more with

342
00:20:10,760 --> 00:20:14,560
Speaker 1: Soul Machines CEO Greg Cross in just a moment. Hi,

343
00:20:14,640 --> 00:20:19,399
Speaker 1: this is Andrew Wallenstein with Variety's Strictly Business podcast. If

344
00:20:19,440 --> 00:20:21,880
Speaker 1: you love Varieties podcasts, you're going to want to try

345
00:20:22,119 --> 00:20:27,920
Speaker 1: Variety Intelligence Platform or VIP. It's a digital subscription tier

346
00:20:27,960 --> 00:20:31,560
Speaker 1: on Variety dot com for industry professionals to dig deeper

347
00:20:31,600 --> 00:20:35,160
Speaker 1: into analysis and data that helps them be smarter about

348
00:20:35,200 --> 00:20:38,840
Speaker 1: their business. VIP just launched a great new newsletter and

349
00:20:38,880 --> 00:20:42,840
Speaker 1: offers more special reports than ever, So visit Variety dot

350
00:20:42,960 --> 00:20:50,120
Speaker 1: com slash vip save for a twenty percent discount. Now

351
00:20:50,119 --> 00:20:55,240
Speaker 1: we're back with Soul Machines CEO Greg Cross. Greg, you know,

352
00:20:55,280 --> 00:20:57,919
Speaker 1: we were just talking about the safety issue and I

353
00:20:57,960 --> 00:21:01,879
Speaker 1: hear you're taking great care here. The technology is going

354
00:21:01,920 --> 00:21:04,159
Speaker 1: to get better and better. But at the end of

355
00:21:04,200 --> 00:21:06,320
Speaker 1: the day, there's a bit of a paradox here because

356
00:21:06,680 --> 00:21:10,119
Speaker 1: at the end of the day they're autonomous, and I

357
00:21:10,119 --> 00:21:14,399
Speaker 1: guess they're safeguards, but I still put myself in the

358
00:21:14,400 --> 00:21:17,119
Speaker 1: shoes of say an agent you were pitching your services to,

359
00:21:17,800 --> 00:21:24,080
Speaker 1: and can you truly guarantee that my you know, this

360
00:21:24,240 --> 00:21:27,400
Speaker 1: digital double of the celebrity is not going to get

361
00:21:27,440 --> 00:21:31,960
Speaker 1: my real celebrity in trouble, you know, what do you

362
00:21:32,240 --> 00:21:35,800
Speaker 1: possibly say to you know, placate those fears?

363
00:21:36,000 --> 00:21:38,520
Speaker 2: Yeah, you know, so the first thing, real celebrities can

364
00:21:38,520 --> 00:21:40,520
Speaker 2: do a pretty good job of getting that in trouble

365
00:21:40,560 --> 00:21:43,760
Speaker 2: in their own rights. So, I mean, you know, the

366
00:21:43,760 --> 00:21:48,960
Speaker 2: digital world often mimics the real world. So so look,

367
00:21:49,440 --> 00:21:52,920
Speaker 2: you know, there are I mean, there are simple safeguards

368
00:21:52,720 --> 00:21:54,800
Speaker 2: that we can put in place. So when we create

369
00:21:54,840 --> 00:22:00,399
Speaker 2: a digital personality, for you know, like a Camelo Anthony,

370
00:22:01,240 --> 00:22:04,960
Speaker 2: the digital personality we configure doesn't include his ability to

371
00:22:05,000 --> 00:22:10,119
Speaker 2: express any negative emotions. So you know, physically, you know that,

372
00:22:10,400 --> 00:22:14,800
Speaker 2: you know, we're going to protect the public persona and

373
00:22:14,840 --> 00:22:21,399
Speaker 2: the public brand of Mallow through a personality that, you know,

374
00:22:21,840 --> 00:22:26,399
Speaker 2: a digital personality that has no ability to express negative emotion.

375
00:22:26,480 --> 00:22:27,919
Speaker 1: But then let me argue the other side of it,

376
00:22:27,960 --> 00:22:31,359
Speaker 1: which is you put in too many safeguards. The animation

377
00:22:31,760 --> 00:22:36,000
Speaker 1: doesn't seem real enough, Like so do you sort of

378
00:22:36,080 --> 00:22:37,200
Speaker 1: is it a double edged sword?

379
00:22:37,480 --> 00:22:39,199
Speaker 2: No, not really, because at the end of the day,

380
00:22:39,200 --> 00:22:43,160
Speaker 2: it's human beings. We tend to respond, you know, mainly

381
00:22:43,200 --> 00:22:47,679
Speaker 2: to positive emotional constructs. An emotional connection is formed, you know,

382
00:22:48,119 --> 00:22:52,159
Speaker 2: from the simple act of saying hello and smiling. You know,

383
00:22:52,320 --> 00:22:57,280
Speaker 2: that's what triggers the dopamine and serotonin in your brain,

384
00:22:57,480 --> 00:23:01,640
Speaker 2: which you know creates this positive connections that in these

385
00:23:01,680 --> 00:23:05,520
Speaker 2: positive emotions, you know, our brain much prefers that that state.

386
00:23:06,200 --> 00:23:10,639
Speaker 2: Then the agitated angry states that you know, you know,

387
00:23:11,040 --> 00:23:13,000
Speaker 2: you know, are the flip side of that. So no,

388
00:23:13,160 --> 00:23:16,199
Speaker 2: I don't know, I don't see that. You know that,

389
00:23:16,920 --> 00:23:21,199
Speaker 2: you know, we're not effectively taking away the power to connect.

390
00:23:21,280 --> 00:23:23,080
Speaker 2: I mean, because I say the power to connect comes

391
00:23:23,119 --> 00:23:25,560
Speaker 2: from positive emotion, not so much negotive emotion.

392
00:23:25,840 --> 00:23:28,680
Speaker 1: So you've been around since twenty sixteen with this company,

393
00:23:28,720 --> 00:23:30,399
Speaker 1: and so I'm wondering, as we sit here in twenty

394
00:23:30,400 --> 00:23:33,720
Speaker 1: twenty three, how come I'm not seeing you know, thousands

395
00:23:33,720 --> 00:23:37,480
Speaker 1: of these digital mellows for every other NBA player or

396
00:23:37,520 --> 00:23:40,159
Speaker 1: every other TV star, movie star, musician.

397
00:23:40,480 --> 00:23:43,879
Speaker 2: Sure, and you know we're just at the beginning of

398
00:23:44,240 --> 00:23:48,520
Speaker 2: this era, you know, to create I mean, the cost

399
00:23:48,560 --> 00:23:53,320
Speaker 2: of and the process is to create these incredibly realistic

400
00:23:53,400 --> 00:23:58,440
Speaker 2: CGI characters, you know, you know, you go back several years,

401
00:23:58,880 --> 00:24:01,720
Speaker 2: you know, and the cost of creating CGI movies, I

402
00:24:01,720 --> 00:24:03,920
Speaker 2: mean these are I mean, it's eye watering in terms

403
00:24:03,920 --> 00:24:07,640
Speaker 2: of the ability to create CGI characters, and it can

404
00:24:07,680 --> 00:24:10,600
Speaker 2: cost hundreds of thousands of dollars per minute of animation

405
00:24:11,240 --> 00:24:14,719
Speaker 2: to create pre recorded content of the quality required in

406
00:24:14,800 --> 00:24:18,560
Speaker 2: the movie industry and the games industry. So there's been

407
00:24:18,600 --> 00:24:20,400
Speaker 2: a huge amount of research. I mean we're ten years

408
00:24:20,440 --> 00:24:23,359
Speaker 2: into our research program, you know, and a lot of

409
00:24:23,400 --> 00:24:26,639
Speaker 2: effort to actually reduce that cost. So I mean we

410
00:24:26,760 --> 00:24:28,760
Speaker 2: can now you know, when we started, I'll give you

411
00:24:29,680 --> 00:24:32,760
Speaker 2: some idea. When we started, we did our first digital twin.

412
00:24:32,880 --> 00:24:36,880
Speaker 2: We created Jack Nicklas, the golfer, the golfer, the golfing legend.

413
00:24:37,400 --> 00:24:40,720
Speaker 2: Jack wanted to be first, I mean seriously, and he's

414
00:24:40,760 --> 00:24:43,479
Speaker 2: eighty three years old and he had no qualms. I

415
00:24:43,520 --> 00:24:46,080
Speaker 2: want to be first, but I want to be thirty

416
00:24:46,119 --> 00:24:49,119
Speaker 2: eight years old again. And that's what we did. So

417
00:24:49,160 --> 00:24:52,600
Speaker 2: we made Jack thirty eight years old. And why did

418
00:24:52,600 --> 00:24:54,640
Speaker 2: he pick thirty eight Because Jack, I mean, that's when

419
00:24:54,680 --> 00:24:56,440
Speaker 2: I was playing my best golf. I was in the

420
00:24:56,800 --> 00:25:00,000
Speaker 2: prime of my you know, when he won the British

421
00:25:00,080 --> 00:25:01,879
Speaker 2: Open for the last time at nineteen seventy eight. So

422
00:25:01,880 --> 00:25:04,440
Speaker 2: we made a digital twiner of Jack at thirty eight

423
00:25:04,480 --> 00:25:08,280
Speaker 2: years old for him. When we did that project, it

424
00:25:08,320 --> 00:25:11,840
Speaker 2: took us six months to build digital jack. You know,

425
00:25:12,480 --> 00:25:18,160
Speaker 2: with our most recent celebrity that we've announced, the digital

426
00:25:18,200 --> 00:25:20,240
Speaker 2: twinter Mark Twine, the K pop star I was telling

427
00:25:20,280 --> 00:25:22,520
Speaker 2: you about, that took less than four weeks.

428
00:25:23,119 --> 00:25:24,320
Speaker 1: So you're getting better at there.

429
00:25:24,359 --> 00:25:27,760
Speaker 2: So it's just getting faster and faster and faster.

430
00:25:28,200 --> 00:25:30,480
Speaker 1: And is it getting cheaper and cheaper, because if it's not,

431
00:25:30,680 --> 00:25:31,800
Speaker 1: how does this a business?

432
00:25:31,880 --> 00:25:34,080
Speaker 2: Yeah? Of course, I mean, of course it gets cheaper

433
00:25:34,119 --> 00:25:36,040
Speaker 2: because you know, I mean, if you think about the

434
00:25:36,080 --> 00:25:39,199
Speaker 2: sort of CGI characters you know that appear in the

435
00:25:39,240 --> 00:25:41,320
Speaker 2: movie and the games entry, they're all built by hand.

436
00:25:41,960 --> 00:25:44,600
Speaker 2: So yeah, what we're focused on, you know, is we're

437
00:25:44,640 --> 00:25:47,639
Speaker 2: automating the production process. And so if you go to

438
00:25:47,720 --> 00:25:52,080
Speaker 2: our website and you can talk to Nova. So Nova

439
00:25:52,520 --> 00:25:56,119
Speaker 2: is not a digital celebrity, she's our brand ambassador and

440
00:25:56,160 --> 00:26:01,600
Speaker 2: she's a synthetic digital person. She is she's not made

441
00:26:01,600 --> 00:26:04,280
Speaker 2: off the likeness of a real person, you know. So

442
00:26:04,440 --> 00:26:07,920
Speaker 2: we we have a creative suite for our synthetic digital people.

443
00:26:07,920 --> 00:26:10,919
Speaker 2: That means literally anybody in the world can create a

444
00:26:11,040 --> 00:26:17,040
Speaker 2: high quality, a hyper realistic CGI character in minutes. You know,

445
00:26:17,280 --> 00:26:22,480
Speaker 2: different different ages, you know, either gender or blended genders,

446
00:26:24,000 --> 00:26:26,959
Speaker 2: different ethnicities, they can you know, you can choose different

447
00:26:27,440 --> 00:26:30,600
Speaker 2: accents and different voices and different languages that they speak.

448
00:26:30,600 --> 00:26:33,160
Speaker 2: All of this is exists in the world of AI today.

449
00:26:33,240 --> 00:26:37,480
Speaker 2: I mean, you know, when we work with our partners

450
00:26:37,520 --> 00:26:42,040
Speaker 2: to create a synthetic digital voice for for Jack Nicholas,

451
00:26:42,960 --> 00:26:45,160
Speaker 2: we can take his English voice and turn it into

452
00:26:45,240 --> 00:26:49,800
Speaker 2: Japanese and into Mandarin. You know, Jack is about to

453
00:26:49,800 --> 00:26:54,200
Speaker 2: become a brand ambassador for a golf merchandising company in Tokyo.

454
00:26:54,280 --> 00:26:55,920
Speaker 2: I mean, Jack in real life doesn't speak a word

455
00:26:55,920 --> 00:26:59,000
Speaker 2: of Japanese. His digital thirty eight year old digital twin

456
00:26:59,280 --> 00:27:02,879
Speaker 2: speaks fluid Japanese in Jack's voice. So you know these

457
00:27:03,320 --> 00:27:05,119
Speaker 2: you know, these are you know, so all of the

458
00:27:05,160 --> 00:27:07,480
Speaker 2: stuff gets faster and faster and faster at this point

459
00:27:07,520 --> 00:27:10,200
Speaker 2: in time. And you know, and the big breakthroughs in

460
00:27:10,280 --> 00:27:12,800
Speaker 2: AI that we've seen with the large language models like

461
00:27:12,880 --> 00:27:16,400
Speaker 2: Chad GPT have just accelerated that at a dramatic rate.

462
00:27:16,600 --> 00:27:18,919
Speaker 1: So fast forward for me for a little like, do

463
00:27:19,000 --> 00:27:21,920
Speaker 1: you see two years from now or five years from now,

464
00:27:22,520 --> 00:27:26,000
Speaker 1: you know, every Hollywood celebrity is going to have a

465
00:27:26,080 --> 00:27:27,200
Speaker 1: digital double.

466
00:27:26,960 --> 00:27:30,359
Speaker 2: Like absolutely, you know, absolutely, I mean you know, I mean,

467
00:27:30,520 --> 00:27:32,399
Speaker 2: you know, if you think back to the early days

468
00:27:32,440 --> 00:27:35,400
Speaker 2: of social media, you know, I mean that's a good yeah.

469
00:27:35,400 --> 00:27:38,200
Speaker 2: I mean not everybody, not every celebrity had an Instagram

470
00:27:38,240 --> 00:27:43,560
Speaker 2: account or a Facebook account. Now everybody does, so yeah,

471
00:27:43,600 --> 00:27:46,000
Speaker 2: I mean, this is this is a process of you know,

472
00:27:46,160 --> 00:27:50,120
Speaker 2: the technology becoming you know, a bigger and bigger part

473
00:27:50,160 --> 00:27:54,520
Speaker 2: of people's lives. You know, within the sort of time

474
00:27:54,560 --> 00:27:56,840
Speaker 2: frames you were just talking about. Then you know, you

475
00:27:56,880 --> 00:27:59,440
Speaker 2: will have a choice to make a digital twin of yourself,

476
00:28:00,640 --> 00:28:03,879
Speaker 2: a hyper realistic digital twin of yourself, you know, I

477
00:28:03,880 --> 00:28:04,800
Speaker 2: mean today, you.

478
00:28:04,760 --> 00:28:05,960
Speaker 1: Know, Wafe would be thrilled.

479
00:28:07,440 --> 00:28:09,040
Speaker 2: You know, you'll be able to take ten years off

480
00:28:09,040 --> 00:28:12,000
Speaker 2: your age if you want. You know, you you know,

481
00:28:12,160 --> 00:28:15,040
Speaker 2: you'll be able to have you know, long hair if

482
00:28:15,080 --> 00:28:17,400
Speaker 2: you want, or you know, blonde hair. I mean, these

483
00:28:17,440 --> 00:28:18,600
Speaker 2: are you know, I mean, these are some of the

484
00:28:18,600 --> 00:28:21,040
Speaker 2: things that that will that that you know, and you'll

485
00:28:21,040 --> 00:28:23,920
Speaker 2: be able to train that digital person and the personality

486
00:28:23,920 --> 00:28:27,040
Speaker 2: you want them to represent yourself and just by talking

487
00:28:27,040 --> 00:28:29,359
Speaker 2: to them and interacting with them. And you know, and

488
00:28:29,400 --> 00:28:32,520
Speaker 2: while you know, creating a custom voice today you know,

489
00:28:32,960 --> 00:28:37,160
Speaker 2: you know takes you know, a few hours of voice recording,

490
00:28:37,520 --> 00:28:39,600
Speaker 2: you know, in the future, it will take a few

491
00:28:39,640 --> 00:28:42,320
Speaker 2: minutes or a few seconds of voice recording to train

492
00:28:42,560 --> 00:28:44,680
Speaker 2: at a custom voice. So you know, these are you know,

493
00:28:44,720 --> 00:28:48,480
Speaker 2: so all of these technologies will you know, democratize in

494
00:28:48,560 --> 00:28:51,680
Speaker 2: the same way that you know, social media has enabled

495
00:28:51,760 --> 00:28:55,440
Speaker 2: us to stay in touch with our our favorite celebrities

496
00:28:55,920 --> 00:28:57,040
Speaker 2: you know online today.

497
00:28:57,240 --> 00:29:00,480
Speaker 1: Speaking of social media, you know, the celebrity has got

498
00:29:00,480 --> 00:29:05,720
Speaker 1: there after a generation of influencers or creators figured that

499
00:29:05,880 --> 00:29:08,880
Speaker 1: all that and then the celebrities came in. Your business

500
00:29:08,920 --> 00:29:13,240
Speaker 1: is now also catering to the influencer crowd with this

501
00:29:13,800 --> 00:29:16,160
Speaker 1: new product that you have out. Explain how that one.

502
00:29:16,120 --> 00:29:19,320
Speaker 2: Is, yeah, I mean, yeah, you know, this is the

503
00:29:19,320 --> 00:29:22,600
Speaker 2: Digital DNA Studio, which is a creative suite, so you know,

504
00:29:22,840 --> 00:29:26,920
Speaker 2: which means people can create a digital influencer themselves. They

505
00:29:26,960 --> 00:29:29,920
Speaker 2: can they can create skills for it, they can attach

506
00:29:30,040 --> 00:29:36,000
Speaker 2: it to a GPT engine or a conversational knowledge base,

507
00:29:36,920 --> 00:29:38,840
Speaker 2: and then they can deploy it in an e commerce

508
00:29:38,880 --> 00:29:41,520
Speaker 2: store or a web store or a smartphone app. So

509
00:29:41,600 --> 00:29:43,959
Speaker 2: this is you know, this is the way we you know,

510
00:29:44,320 --> 00:29:45,920
Speaker 2: we are going about, you know, one of the ways

511
00:29:45,960 --> 00:29:48,920
Speaker 2: we're going about building our business more on the commercial side.

512
00:29:48,960 --> 00:29:51,720
Speaker 2: You know, when you come to celebrities. You know, you've

513
00:29:51,720 --> 00:29:54,000
Speaker 2: got to know, you take a lot more time and

514
00:29:54,040 --> 00:29:57,000
Speaker 2: spend a lot you know, you actually want to replicate

515
00:29:57,280 --> 00:30:01,120
Speaker 2: you know, you know the celebrity, you as accurately as

516
00:30:01,160 --> 00:30:05,800
Speaker 2: you can. Their mannerisms, there, the behaviors that are identifiable

517
00:30:06,640 --> 00:30:10,360
Speaker 2: you know in real life. You you that requires more work.

518
00:30:11,320 --> 00:30:13,719
Speaker 1: And on that note, to some degree, I listen to

519
00:30:13,720 --> 00:30:16,160
Speaker 1: this and I say to myself, why even bother with

520
00:30:16,240 --> 00:30:19,680
Speaker 1: the celebrities you can create you For instance, you're talking

521
00:30:19,680 --> 00:30:24,480
Speaker 1: about Novaya. Give Nova a singing voice, gives Nova acting talent.

522
00:30:24,880 --> 00:30:27,920
Speaker 1: There is this whole branch of synthetic media where they

523
00:30:28,000 --> 00:30:29,600
Speaker 1: try to create their own are you.

524
00:30:29,640 --> 00:30:32,080
Speaker 2: Yeah, yeah, I mean of course I mean and yeah, yeah,

525
00:30:32,120 --> 00:30:34,160
Speaker 2: of course you know, and so we will, without a doubt,

526
00:30:34,160 --> 00:30:37,440
Speaker 2: we will have creators who go and create their version

527
00:30:37,440 --> 00:30:40,640
Speaker 2: of Nova, you know, our digital influencer and they and

528
00:30:40,680 --> 00:30:43,000
Speaker 2: they create it, and they will create audiences for them

529
00:30:43,040 --> 00:30:46,080
Speaker 2: on social media, which they will in turn be able

530
00:30:46,080 --> 00:30:48,120
Speaker 2: to monetize and you know, they will be able to

531
00:30:48,120 --> 00:30:51,240
Speaker 2: click on that and then go have a live conversation

532
00:30:51,320 --> 00:30:52,920
Speaker 2: and if they know and they might be able to,

533
00:30:52,920 --> 00:30:54,960
Speaker 2: you know, they might end up live streaming, you know,

534
00:30:55,040 --> 00:30:58,240
Speaker 2: their their digital influencer on you know, on a streaming

535
00:30:58,280 --> 00:31:01,840
Speaker 2: platform and enabling fair to pay to interact with their

536
00:31:02,400 --> 00:31:04,240
Speaker 2: you know, with their digital influence. I mean, there's some

537
00:31:04,280 --> 00:31:07,560
Speaker 2: great examples of it out there already where you know,

538
00:31:07,640 --> 00:31:11,640
Speaker 2: we've seen Brazil is a big market for these digital

539
00:31:11,680 --> 00:31:14,840
Speaker 2: influences and then these digital ambassadors. So you know, one

540
00:31:14,840 --> 00:31:17,360
Speaker 2: of the biggest, one of the most successful ones is

541
00:31:18,440 --> 00:31:23,040
Speaker 2: a digital influencer called Lou of Margalou. She represents one

542
00:31:23,040 --> 00:31:27,600
Speaker 2: of the largest retail chains in Brazil and they've had

543
00:31:27,640 --> 00:31:30,480
Speaker 2: Lou out there and she's mainly she's a CGI character.

544
00:31:30,560 --> 00:31:33,520
Speaker 2: She's a made up CGI character. She's not a real person.

545
00:31:34,600 --> 00:31:36,640
Speaker 2: And she's been around for I think ten odd years.

546
00:31:36,640 --> 00:31:40,680
Speaker 2: She has over thirty million followers on social media, and

547
00:31:40,720 --> 00:31:44,120
Speaker 2: she has gone past just being a brand ambassador for

548
00:31:44,400 --> 00:31:48,640
Speaker 2: you know, the retail brand that she represents. She's become

549
00:31:48,880 --> 00:31:53,040
Speaker 2: a social influencer in Brazil. You know, she is outspoken

550
00:31:53,080 --> 00:31:56,480
Speaker 2: on women's rights in Brazil, for example. So you know,

551
00:31:56,520 --> 00:31:58,160
Speaker 2: I mean, this is you know, I mean, but we

552
00:31:58,200 --> 00:32:00,280
Speaker 2: see this on a regular basis. I Mean, some of

553
00:32:00,320 --> 00:32:02,600
Speaker 2: the biggest celebrities in the world have hundreds of millions

554
00:32:02,640 --> 00:32:06,120
Speaker 2: of followers. They have a percentage of those followers, and

555
00:32:06,600 --> 00:32:09,000
Speaker 2: I feel like they have a personal connection with those

556
00:32:09,040 --> 00:32:13,080
Speaker 2: celebrities just because they're able to like or post comments

557
00:32:13,120 --> 00:32:16,080
Speaker 2: against their social media posts. So, you know, a lot

558
00:32:16,120 --> 00:32:19,720
Speaker 2: of these digital behaviors around, you know, this world of

559
00:32:19,880 --> 00:32:23,600
Speaker 2: celebrity and this world of influences you know, you know

560
00:32:23,640 --> 00:32:26,880
Speaker 2: already exists and except you know worth when you create

561
00:32:26,880 --> 00:32:29,240
Speaker 2: a digital twin, you make it more intimate, you make

562
00:32:29,280 --> 00:32:33,680
Speaker 2: it more connective, you make it more believable, and that

563
00:32:34,000 --> 00:32:36,360
Speaker 2: you know what what we're we're incredibly excited by that

564
00:32:36,480 --> 00:32:39,480
Speaker 2: asn't as an opportunity. So you know, you can imagine,

565
00:32:39,680 --> 00:32:44,600
Speaker 2: you know, if we're you know, we're working with you

566
00:32:44,640 --> 00:32:47,520
Speaker 2: know Hollywood star, you know, I mean we've talked about

567
00:32:47,520 --> 00:32:50,760
Speaker 2: Marilyn Mony, but if we're talking about we have a

568
00:32:50,760 --> 00:32:53,840
Speaker 2: Hollywood star who's got a big back catalog of films.

569
00:32:53,920 --> 00:32:56,640
Speaker 2: You know, you could your or you know one of

570
00:32:56,680 --> 00:32:59,720
Speaker 2: the production studios who's created a movie franchise, you know

571
00:32:59,760 --> 00:33:03,200
Speaker 2: where they have multiple installments of that movie franchise, and

572
00:33:03,400 --> 00:33:07,400
Speaker 2: imagine bringing that actor to life as that character and

573
00:33:07,560 --> 00:33:10,720
Speaker 2: fans of that movie franchise can have a day to

574
00:33:10,800 --> 00:33:13,960
Speaker 2: day relationship with that character and you know, they can

575
00:33:14,000 --> 00:33:17,880
Speaker 2: get backstories and insights and feel connected to you know,

576
00:33:17,960 --> 00:33:20,360
Speaker 2: on a you know, literally on a day to day basis,

577
00:33:20,440 --> 00:33:23,440
Speaker 2: rather than having to wait twelve months or you know,

578
00:33:23,480 --> 00:33:27,240
Speaker 2: between a franchise installment. So these are you I mean,

579
00:33:27,240 --> 00:33:29,480
Speaker 2: these are some of the ways in which you know,

580
00:33:29,520 --> 00:33:32,120
Speaker 2: we think about fan engagement. I mean, if you can

581
00:33:32,200 --> 00:33:37,040
Speaker 2: to get into the world of professional sports games like football,

582
00:33:37,280 --> 00:33:40,520
Speaker 2: you know, the NFL has an interesting challenge with the

583
00:33:40,640 --> 00:33:43,760
Speaker 2: kids of today. They don't sit still for four hours

584
00:33:44,240 --> 00:33:46,760
Speaker 2: on a Sunday afternoon to watch a football game. So

585
00:33:47,200 --> 00:33:49,920
Speaker 2: you know, how do you curate and present content to

586
00:33:49,960 --> 00:33:52,200
Speaker 2: them in a form fact that and in a way

587
00:33:52,800 --> 00:33:56,239
Speaker 2: that interest them and keeps them connected to the game. So,

588
00:33:56,640 --> 00:33:58,440
Speaker 2: you know, these are some of the things that we

589
00:33:58,520 --> 00:34:01,160
Speaker 2: spend a lot of time thinking about and talking to,

590
00:34:01,680 --> 00:34:04,280
Speaker 2: you know, not just the celebrities, but the big media

591
00:34:04,280 --> 00:34:07,360
Speaker 2: companies and the big content providers who you know, ultimately

592
00:34:07,400 --> 00:34:11,439
Speaker 2: you know, in the world of professional sports, that's what funds, uh,

593
00:34:12,280 --> 00:34:15,000
Speaker 2: that's what funds of sports makes the will go round

594
00:34:15,000 --> 00:34:18,160
Speaker 2: for these things. So I mean, so you know, these

595
00:34:18,200 --> 00:34:20,840
Speaker 2: are some of the ways in which you know, we reimagine,

596
00:34:21,000 --> 00:34:23,319
Speaker 2: you know, we think about the future of entertainment and

597
00:34:23,360 --> 00:34:27,440
Speaker 2: the way in which you will engage fans with personalized

598
00:34:27,480 --> 00:34:30,320
Speaker 2: content and emotional connection in the future.

599
00:34:31,080 --> 00:34:32,920
Speaker 1: Well, I don't know about you, but listening to this,

600
00:34:33,120 --> 00:34:35,920
Speaker 1: my brain is spinning. I'm sure my listener's brands are

601
00:34:35,920 --> 00:34:38,480
Speaker 1: going to be spinning. Greg. Thank you so much for

602
00:34:38,520 --> 00:34:41,120
Speaker 1: coming in and talking about Sole Machines. This is going

603
00:34:41,120 --> 00:34:42,640
Speaker 1: to be an interesting company to watch.

604
00:34:43,480 --> 00:34:45,279
Speaker 2: Thank you so much for having me. It's been it's

605
00:34:45,280 --> 00:34:46,600
Speaker 2: been an absolute blasting year.

606
00:34:52,719 --> 00:34:55,400
Speaker 1: Well that was part one of my interview with Greg Cross,

607
00:34:55,520 --> 00:34:58,440
Speaker 1: CEO of Sole Machines. Come back next week to the

608
00:34:58,440 --> 00:35:02,080
Speaker 1: Strictly Business podcast for part two of the interview. There's

609
00:35:02,080 --> 00:35:04,360
Speaker 1: so much more with Greg. You're not gonna want to

610
00:35:04,440 --> 00:35:05,680
Speaker 1: miss see you then,