1
00:00:02,520 --> 00:00:09,160
Speaker 1: Bloomberg Audio Studios, podcasts, radio news. You're listening to Bloomberg

2
00:00:09,240 --> 00:00:13,760
Speaker 1: BusinessWeek with Carol Masser and Tim Steneveek on Bloomberg Radio.

3
00:00:14,080 --> 00:00:17,160
Speaker 2: As more and more companies adopt AI and more consumers

4
00:00:17,280 --> 00:00:19,599
Speaker 2: use it, the more we learn about the potential flaws

5
00:00:19,920 --> 00:00:23,040
Speaker 2: of the technology. For example, Carol, maybe you remember last

6
00:00:23,079 --> 00:00:27,960
Speaker 2: year when Google's AI suggested that everybody should just beat

7
00:00:28,040 --> 00:00:29,200
Speaker 2: a rock every day.

8
00:00:29,320 --> 00:00:30,240
Speaker 3: Not a good idea.

9
00:00:30,320 --> 00:00:33,479
Speaker 2: Well, they said, rocks have minerals and vitamins that are

10
00:00:33,560 --> 00:00:35,280
Speaker 2: quote important for digestive health.

11
00:00:35,400 --> 00:00:35,720
Speaker 1: Got it.

12
00:00:35,840 --> 00:00:38,400
Speaker 3: I understand that logic, but still not a good idea.

13
00:00:38,520 --> 00:00:40,000
Speaker 2: Or just a few weeks ago, in the Fable app,

14
00:00:40,040 --> 00:00:43,519
Speaker 2: which calls itself the social app for bookworms and binge watchers,

15
00:00:43,520 --> 00:00:46,240
Speaker 2: found itself having to apologize after some of its AI

16
00:00:46,320 --> 00:00:49,400
Speaker 2: generated reader summaries contained what the company ended up calling

17
00:00:49,479 --> 00:00:53,360
Speaker 2: quote very bigoted, racist language. Fable then remove these features

18
00:00:53,880 --> 00:00:57,160
Speaker 2: from the app. It's an evolving world, Yeah absolutely. Padma

19
00:00:57,200 --> 00:00:59,680
Speaker 2: Warrior is founder, president and a CEO of Fable. She's

20
00:00:59,680 --> 00:01:02,880
Speaker 2: also the former CTO of Motorola and Cisco, also the

21
00:01:02,880 --> 00:01:05,760
Speaker 2: former CEO of the US arm of the Chinese ev

22
00:01:05,880 --> 00:01:08,559
Speaker 2: maker Neo. She was on the board of Microsoft until

23
00:01:08,560 --> 00:01:11,280
Speaker 2: twenty twenty three. She's on the board of Spotify as well.

24
00:01:11,319 --> 00:01:14,720
Speaker 2: She joins us this afternoon from Palo Alto Padma. Good

25
00:01:14,720 --> 00:01:15,720
Speaker 2: to have you with us. How are you?

26
00:01:17,000 --> 00:01:18,919
Speaker 4: Thank you for having me. I'm doing well.

27
00:01:19,360 --> 00:01:22,120
Speaker 2: So I'm curious about Fable. A lot of people are

28
00:01:22,160 --> 00:01:26,920
Speaker 2: familiar with it because people use it for rating books,

29
00:01:27,680 --> 00:01:31,840
Speaker 2: finding books, same with stuff we watch on the bigger screen.

30
00:01:33,080 --> 00:01:35,319
Speaker 2: You're the lesson though. The takeaway from AI you guys

31
00:01:35,319 --> 00:01:38,440
Speaker 2: were so quick to come out and apologize for it.

32
00:01:38,880 --> 00:01:41,880
Speaker 2: Take action, be public about it. What were the What

33
00:01:42,240 --> 00:01:43,440
Speaker 2: did you learn from that incident?

34
00:01:44,880 --> 00:01:48,200
Speaker 4: Yeah? I know. Fable is a community platform.

35
00:01:48,480 --> 00:01:52,480
Speaker 5: We focus on helping people find, by track, and share

36
00:01:52,600 --> 00:01:57,640
Speaker 5: all their entertainment interests in one personalized space. We have

37
00:01:57,880 --> 00:02:01,760
Speaker 5: users from all over the world onward and ninety five countries,

38
00:02:02,240 --> 00:02:05,800
Speaker 5: and our mission is to celebrate human creative tea in

39
00:02:05,840 --> 00:02:10,000
Speaker 5: the form of stories. So the incidents you are referring to,

40
00:02:10,280 --> 00:02:14,240
Speaker 5: we were using generative AI to create a summary for

41
00:02:14,320 --> 00:02:19,520
Speaker 5: readers based on their reading lists, and regrettably it served

42
00:02:19,520 --> 00:02:24,240
Speaker 5: a few readers with hurtful, unacceptable commentary related to race.

43
00:02:25,040 --> 00:02:27,200
Speaker 4: We apologized, as you mentioned.

44
00:02:27,000 --> 00:02:30,800
Speaker 5: Very quickly removed these gen ai features from our platform.

45
00:02:31,520 --> 00:02:34,400
Speaker 5: To me, they were very A couple of key lessons

46
00:02:34,440 --> 00:02:39,320
Speaker 5: that we learned. One is it's really important to have

47
00:02:39,360 --> 00:02:42,800
Speaker 5: a human in the loop. In these early days of

48
00:02:42,960 --> 00:02:48,359
Speaker 5: Generator Ai. It is essential to carry out really extensive

49
00:02:48,480 --> 00:02:50,760
Speaker 5: beta testing, and we did a lot of testing, by

50
00:02:50,760 --> 00:02:55,320
Speaker 5: the way, for our feature, but even with all the testing,

51
00:02:55,440 --> 00:02:57,919
Speaker 5: you're bound to have an issue. So I think it's

52
00:02:58,000 --> 00:03:01,079
Speaker 5: really really important to have human in the loop approach.

53
00:03:01,560 --> 00:03:07,280
Speaker 5: The second is transparency and listening to community feedback when

54
00:03:07,440 --> 00:03:11,120
Speaker 5: someone I think our mission, our philosophy at our company

55
00:03:11,200 --> 00:03:14,400
Speaker 5: is even when we do something with our product, which

56
00:03:14,440 --> 00:03:18,040
Speaker 5: generator wai, even if it hurts one person, we take

57
00:03:18,080 --> 00:03:21,400
Speaker 5: it very seriously. So I think that level of transparency

58
00:03:21,440 --> 00:03:24,840
Speaker 5: and being close to listening to the community and users

59
00:03:25,240 --> 00:03:28,520
Speaker 5: and having the ability to react quickly, those are two

60
00:03:28,560 --> 00:03:31,840
Speaker 5: big lessons that we learned and I hope other entrepreneurs

61
00:03:31,880 --> 00:03:32,360
Speaker 5: follow this.

62
00:03:33,600 --> 00:03:35,360
Speaker 3: How much is it that just we're kind of early

63
00:03:35,400 --> 00:03:37,480
Speaker 3: in on this process. I'm not excusing any of it

64
00:03:37,520 --> 00:03:41,880
Speaker 3: that goes, you know, it goes like it did for

65
00:03:41,920 --> 00:03:44,040
Speaker 3: you guys. But is it a case of it's only

66
00:03:44,040 --> 00:03:46,920
Speaker 3: as good as its input, and that is it a

67
00:03:46,960 --> 00:03:49,680
Speaker 3: fair assumption. How do you see you understand the tech world,

68
00:03:49,680 --> 00:03:51,800
Speaker 3: and I'm sure you're looking closely, you know at AI

69
00:03:51,920 --> 00:03:56,280
Speaker 3: right now that the more we put in better, cleaner information,

70
00:03:56,480 --> 00:03:57,760
Speaker 3: the better it gets.

71
00:03:58,600 --> 00:03:58,840
Speaker 4: Yeah.

72
00:03:58,840 --> 00:04:01,480
Speaker 5: Absolutely, And I think there is a lot of confusion.

73
00:04:01,680 --> 00:04:05,320
Speaker 5: People sometimes confuse AI the generative AI. I think it's

74
00:04:05,360 --> 00:04:09,320
Speaker 5: important to perhaps distinguish a little bit, at least broadly,

75
00:04:09,680 --> 00:04:13,560
Speaker 5: the difference between the two. You know, I think as

76
00:04:13,600 --> 00:04:16,120
Speaker 5: you guys probably know, and I'm not sure how much

77
00:04:16,320 --> 00:04:17,720
Speaker 5: how much people realize this.

78
00:04:17,839 --> 00:04:18,159
Speaker 4: Though.

79
00:04:18,600 --> 00:04:22,400
Speaker 5: AI processes data to analyze it and provide us with

80
00:04:22,560 --> 00:04:27,920
Speaker 5: predictions and insight, for example, performing tasks like identifying objects

81
00:04:27,920 --> 00:04:32,640
Speaker 5: in images, converting speech to tests, recommendations, all of this thing,

82
00:04:33,200 --> 00:04:37,400
Speaker 5: generative AI is different. It creates new content based on

83
00:04:37,920 --> 00:04:41,960
Speaker 5: the data that it has learned from. So generative AI

84
00:04:42,080 --> 00:04:46,240
Speaker 5: can create text, can create images, can create music videos.

85
00:04:45,839 --> 00:04:46,320
Speaker 4: All of this.

86
00:04:46,920 --> 00:04:50,880
Speaker 5: I find that generative AI it's developing very quickly and

87
00:04:51,040 --> 00:04:55,800
Speaker 5: very fast, and it's really really important because as I said,

88
00:04:56,160 --> 00:05:01,440
Speaker 5: even the best AI needs human oversight at the stage.

89
00:05:01,960 --> 00:05:04,679
Speaker 5: As you pointed out, it's only as good as the input.

90
00:05:04,760 --> 00:05:08,720
Speaker 5: So it's really important to train generative AI algorithms and

91
00:05:08,839 --> 00:05:12,679
Speaker 5: actually even AI on data sets that are culturally literate

92
00:05:12,880 --> 00:05:16,840
Speaker 5: and sensitive, because even as humans, we are evolving, right,

93
00:05:17,000 --> 00:05:20,080
Speaker 5: Like language that was used years ago is not acceptable

94
00:05:20,120 --> 00:05:25,000
Speaker 5: to us anymore, and it's really important to train our

95
00:05:25,120 --> 00:05:30,760
Speaker 5: algorithms and artificial intelligence so that it doesn't inadvertently promote

96
00:05:30,800 --> 00:05:32,800
Speaker 5: harmful stereotypes and biases.

97
00:05:34,440 --> 00:05:36,560
Speaker 3: So, okay and Tim and I are just kind of

98
00:05:36,600 --> 00:05:38,760
Speaker 3: like listening to So. I love first of all, your

99
00:05:38,880 --> 00:05:41,880
Speaker 3: distinction between generative AI and AI, and you're right to

100
00:05:41,880 --> 00:05:43,880
Speaker 3: make the distinction. AI has been around for a long

101
00:05:43,920 --> 00:05:48,599
Speaker 3: time in terms of what happened on your app, and

102
00:05:48,680 --> 00:05:54,200
Speaker 3: the information that certainly was not welcomed was that gen

103
00:05:54,279 --> 00:05:56,840
Speaker 3: ai or regular ai, that.

104
00:05:56,920 --> 00:05:57,600
Speaker 4: Was jen Ai.

105
00:05:58,240 --> 00:06:01,120
Speaker 5: We were saying gen Ai in three of our features

106
00:06:01,120 --> 00:06:04,000
Speaker 5: on the product, and we had issue with one of them,

107
00:06:04,080 --> 00:06:07,240
Speaker 5: but we actually pulled all three features back. We launched

108
00:06:07,279 --> 00:06:11,320
Speaker 5: two of them back without Genai and using human creativity,

109
00:06:11,880 --> 00:06:15,200
Speaker 5: so that our community actually is actually more happy now

110
00:06:15,240 --> 00:06:16,240
Speaker 5: with those features.

111
00:06:16,520 --> 00:06:20,039
Speaker 2: Do you think you can relaunch the genai feature eventually?

112
00:06:21,640 --> 00:06:22,800
Speaker 4: I think it'll take us.

113
00:06:23,160 --> 00:06:26,280
Speaker 5: I think for us we need to be really cautious because,

114
00:06:26,320 --> 00:06:29,160
Speaker 5: as I said, if you look at the statistics, right,

115
00:06:29,240 --> 00:06:33,680
Speaker 5: it was like ten possible issues out of thirty million

116
00:06:33,839 --> 00:06:36,240
Speaker 5: that we did. But even that small ratio, I think

117
00:06:36,279 --> 00:06:39,440
Speaker 5: the interesting thing for me, at least with genai, the

118
00:06:39,520 --> 00:06:42,520
Speaker 5: testing can't just be quantified and say we had a

119
00:06:42,600 --> 00:06:46,200
Speaker 5: very small percentage of failure, so we should release this feature,

120
00:06:46,839 --> 00:06:50,120
Speaker 5: because as I said, even one harmful comment and one

121
00:06:50,200 --> 00:06:54,080
Speaker 5: hurtful comment is bad, and so how do you find

122
00:06:54,120 --> 00:06:57,080
Speaker 5: that in millions and millions that you're creating. So I

123
00:06:57,120 --> 00:07:00,000
Speaker 5: think at this stage we will probably not use genai

124
00:07:00,160 --> 00:07:03,440
Speaker 5: for those sorts of features, and I think we will

125
00:07:03,440 --> 00:07:06,200
Speaker 5: continue to work on it to train our models and

126
00:07:06,240 --> 00:07:08,479
Speaker 5: algorithms in the right way.

127
00:07:09,160 --> 00:07:13,680
Speaker 3: Do you think that the government should be more involved

128
00:07:13,800 --> 00:07:18,200
Speaker 3: or should be very closely involved with oversight of AI.

129
00:07:20,160 --> 00:07:20,640
Speaker 4: I'm not sure.

130
00:07:21,160 --> 00:07:24,040
Speaker 5: Yeah, I think it's difficult, I think, but so much

131
00:07:24,160 --> 00:07:27,920
Speaker 5: broad usage, I don't know if government should be involved,

132
00:07:27,920 --> 00:07:30,640
Speaker 5: but there should certainly be guardrailed. I think it's up

133
00:07:30,720 --> 00:07:34,840
Speaker 5: to technology development teams and companies to put those guardrails

134
00:07:34,840 --> 00:07:38,040
Speaker 5: and safeguards because there's broad sets of use cases. They're

135
00:07:38,120 --> 00:07:41,960
Speaker 5: very specific sets of use cases. But yeah, definitely there

136
00:07:42,000 --> 00:07:45,360
Speaker 5: should be policy that guides us at an overarching level

137
00:07:46,120 --> 00:07:49,440
Speaker 5: so that we could use that like anything we govern

138
00:07:49,560 --> 00:07:52,560
Speaker 5: right with technology. I think that is the role that

139
00:07:52,640 --> 00:07:53,600
Speaker 5: government should play.

140
00:07:54,920 --> 00:07:56,760
Speaker 2: What about when it comes to TikTok. It's in the

141
00:07:56,800 --> 00:08:00,520
Speaker 2: news right now, obviously because of that Supreme Court ruling

142
00:08:00,520 --> 00:08:02,880
Speaker 2: that we got. The future of the fate of the

143
00:08:02,920 --> 00:08:05,200
Speaker 2: app is uncertain. What are your thoughts on it.

144
00:08:06,520 --> 00:08:11,280
Speaker 5: We actually are a welcoming community to a big part

145
00:08:11,320 --> 00:08:13,760
Speaker 5: of TikTok called book talk. Book talk is one of

146
00:08:13,760 --> 00:08:17,120
Speaker 5: the largest communities on TikTok, and the last report I

147
00:08:17,200 --> 00:08:22,240
Speaker 5: read it said hashtag book talk is used by billions

148
00:08:22,280 --> 00:08:23,440
Speaker 5: of times across the board.

149
00:08:23,480 --> 00:08:25,000
Speaker 4: It's a very large community.

150
00:08:25,760 --> 00:08:29,080
Speaker 5: We definitely feel we're a safe haven for book talkers

151
00:08:29,120 --> 00:08:32,960
Speaker 5: on fable. Actually, people we do have quite a lot

152
00:08:32,960 --> 00:08:35,839
Speaker 5: of book talk influences today on Fable and a lot of.

153
00:08:35,760 --> 00:08:37,559
Speaker 4: People who use both platforms.

154
00:08:38,200 --> 00:08:41,080
Speaker 5: Many people that are currently on TikTok in book talk

155
00:08:41,160 --> 00:08:45,000
Speaker 5: call us the coziest, called fable as the cosiest corner

156
00:08:45,040 --> 00:08:47,520
Speaker 5: of the Internet. So we hope to be that and

157
00:08:47,559 --> 00:08:49,920
Speaker 5: we hope to welcome communities like that. I think the

158
00:08:49,960 --> 00:08:51,959
Speaker 5: good thing about TikTok, and I think what it has

159
00:08:52,040 --> 00:08:56,000
Speaker 5: shown us is that there is a power to creation

160
00:08:56,360 --> 00:09:00,320
Speaker 5: and community and that is sort of like neededeople to

161
00:09:00,400 --> 00:09:02,000
Speaker 5: express themselves, belong in.

162
00:09:01,920 --> 00:09:04,920
Speaker 4: A community, and be with like minded people.

163
00:09:05,640 --> 00:09:08,000
Speaker 5: So we definitely hope we will be the haven for

164
00:09:08,080 --> 00:09:11,559
Speaker 5: people that are currently on TikTok if it gets banned

165
00:09:11,640 --> 00:09:12,959
Speaker 5: or it is not allowed to.

166
00:09:12,880 --> 00:09:15,360
Speaker 2: Be used in You have two million users around the world.

167
00:09:15,360 --> 00:09:17,240
Speaker 2: Have you seen an uptake in the last couple of weeks.

168
00:09:17,240 --> 00:09:21,400
Speaker 2: Have people come into the platform in anticipation of this, Yes, we've.

169
00:09:21,240 --> 00:09:24,840
Speaker 5: Seen a huge uptick in the last four weeks, actually

170
00:09:24,840 --> 00:09:26,400
Speaker 5: even before when you started.

171
00:09:27,040 --> 00:09:28,800
Speaker 4: So we are just really hoping.

172
00:09:28,520 --> 00:09:32,800
Speaker 5: That this continues and we can host and welcome these

173
00:09:32,840 --> 00:09:34,160
Speaker 5: communities onto Fable.

174
00:09:34,240 --> 00:09:35,960
Speaker 2: So it sounds like a ban of TikTok would be

175
00:09:35,960 --> 00:09:37,240
Speaker 2: really good for Fable.

176
00:09:38,600 --> 00:09:41,120
Speaker 5: I think either way, like I said, we were we

177
00:09:41,360 --> 00:09:44,400
Speaker 5: had a lot of book talkers come to Fable.

178
00:09:44,480 --> 00:09:45,800
Speaker 4: Even with book talk, it is.

179
00:09:45,880 --> 00:09:49,200
Speaker 5: Very complimentary in the use case, so it's not like

180
00:09:49,280 --> 00:09:51,480
Speaker 5: we are routing for that to be banned or anything.

181
00:09:52,520 --> 00:09:56,200
Speaker 5: I think for us, TikTok continues, that's great because that

182
00:09:56,280 --> 00:09:59,880
Speaker 5: community will thrive and we will thrive with it. Unfortunately

183
00:10:00,160 --> 00:10:02,880
Speaker 5: that community goes away. We want to be the alternative

184
00:10:02,960 --> 00:10:03,960
Speaker 5: for people who love.

185
00:10:03,840 --> 00:10:06,760
Speaker 4: Books and TV shows and stories in general.

186
00:10:07,679 --> 00:10:11,800
Speaker 3: What would people maybe find surprising about your app and

187
00:10:12,360 --> 00:10:15,000
Speaker 3: the kind of activity that goes on. What is it

188
00:10:15,360 --> 00:10:18,920
Speaker 3: that you think might be a bit surprising A.

189
00:10:19,240 --> 00:10:21,959
Speaker 5: You know, I think that's a great question. I started

190
00:10:22,040 --> 00:10:25,800
Speaker 5: this company because I really wanted to bring the technology

191
00:10:25,840 --> 00:10:29,400
Speaker 5: world and the creative world together. At Fable, we say

192
00:10:29,440 --> 00:10:31,600
Speaker 5: we want to build a tech company with the soul

193
00:10:31,679 --> 00:10:35,160
Speaker 5: of an artist. So I feel we are very proud

194
00:10:35,200 --> 00:10:37,600
Speaker 5: of our design. We are very proud of the UI

195
00:10:37,800 --> 00:10:43,880
Speaker 5: that we create. It's beautiful. We celebrate human creation. Yet

196
00:10:43,920 --> 00:10:46,560
Speaker 5: we are a tech company. We're based, you know, I'm

197
00:10:46,600 --> 00:10:50,240
Speaker 5: based in Silicon Valley, our team is distributed. I think

198
00:10:50,240 --> 00:10:53,439
Speaker 5: people would find us to be unique in being able

199
00:10:53,480 --> 00:10:57,240
Speaker 5: to bring both of those together. The other thing that

200
00:10:57,320 --> 00:11:00,440
Speaker 5: I feel we are unique is we truly are a

201
00:11:00,559 --> 00:11:03,640
Speaker 5: platform if you love and every human loves stories. Right

202
00:11:03,679 --> 00:11:05,400
Speaker 5: at the end of the day, human beings are all

203
00:11:05,440 --> 00:11:08,240
Speaker 5: about the story we tell. If you love stories, we

204
00:11:08,320 --> 00:11:09,960
Speaker 5: are the destination for you.

205
00:11:10,000 --> 00:11:12,760
Speaker 3: Just got about twenty five seconds left here. How do

206
00:11:12,800 --> 00:11:14,760
Speaker 3: you make money and are you profitable? Just quickly?

207
00:11:15,760 --> 00:11:17,920
Speaker 4: We are very still early. We're a startup.

208
00:11:18,840 --> 00:11:22,160
Speaker 5: We you know, we are not yet quite monetizing the platform.

209
00:11:22,200 --> 00:11:23,560
Speaker 4: We'll begin to do that effort.

210
00:11:23,600 --> 00:11:26,360
Speaker 5: Today, the platform is free, which makes it easy for

211
00:11:26,440 --> 00:11:27,199
Speaker 5: people to join.

212
00:11:28,080 --> 00:11:31,199
Speaker 3: Will it be the subscription based ultimately or is that

213
00:11:31,280 --> 00:11:34,000
Speaker 3: the that's the revenue stream going forward quickly?

214
00:11:34,559 --> 00:11:38,360
Speaker 5: Yes. Yes, we will launch a subscription here soon and

215
00:11:38,440 --> 00:11:41,640
Speaker 5: hope to add traditional value added features for people to

216
00:11:41,720 --> 00:11:45,319
Speaker 5: track their links, to track of their progress, you know,

217
00:11:45,480 --> 00:11:47,880
Speaker 5: keep track of what interests their life, et cetera.

218
00:11:48,160 --> 00:11:50,920
Speaker 3: Got it come back? We'd love to hear more over

219
00:11:50,920 --> 00:11:54,360
Speaker 3: the year. Padna Warrior, founder president CEO of Fable joining

220
00:11:54,440 --> 00:11:57,120
Speaker 3: us on this Friday. This is Bloomberg Business Week