1
00:00:01,520 --> 00:00:05,000
Speaker 1: Welcome to crash Course, a podcast about business, political, and

2
00:00:05,040 --> 00:00:08,560
Speaker 1: social disruption and what we can learn from it. I'm

3
00:00:08,600 --> 00:00:14,800
Speaker 1: Tim O'Brien. Today's crash course open Ai versus Sam Altman.

4
00:00:15,640 --> 00:00:18,600
Speaker 1: Open Ai, which you may have heard a lot about lately,

5
00:00:19,120 --> 00:00:23,080
Speaker 1: is the company that developed chat Gpt, a wildly popular

6
00:00:23,160 --> 00:00:26,960
Speaker 1: AI bot which you most certainly have heard of. Open

7
00:00:27,000 --> 00:00:30,000
Speaker 1: AI's board of directors recently purged to the company's CEO,

8
00:00:30,400 --> 00:00:36,280
Speaker 1: Sam Altman. High drama ensued, and various stakeholders, employees, investors.

9
00:00:36,400 --> 00:00:40,880
Speaker 1: Microsoft saw to it that Altman was reinstated. The board

10
00:00:40,920 --> 00:00:45,879
Speaker 1: itself then faced a purge. Oh my, this particular collision

11
00:00:45,880 --> 00:00:50,800
Speaker 1: has it all Silicon Valley innovation in Silicon Valley hubris, money,

12
00:00:51,240 --> 00:00:57,160
Speaker 1: managerial snafoos, ugly battles, promising outcomes, and of course, artificial

13
00:00:57,160 --> 00:01:01,000
Speaker 1: intelligence AI is set to transform the world. We're told

14
00:01:02,040 --> 00:01:06,000
Speaker 1: ingenuity and upheaval and open ai offer a way for

15
00:01:06,120 --> 00:01:09,199
Speaker 1: us to consider all of that. So I've invited parme

16
00:01:09,319 --> 00:01:12,000
Speaker 1: Olsen and Dave Lee onto crash Course to help us

17
00:01:12,040 --> 00:01:15,600
Speaker 1: outline the lessons of this particular tale. They are both

18
00:01:15,640 --> 00:01:19,679
Speaker 1: Bloomberg Opinion technology columnists. Parmey It's based in London and

19
00:01:19,760 --> 00:01:21,920
Speaker 1: Davi is in New York and they bring a wealth

20
00:01:21,920 --> 00:01:26,280
Speaker 1: of experience and insight to today's show. Welcome my friends,

21
00:01:26,959 --> 00:01:30,800
Speaker 1: Thanks him, thank you. Let's start with Sam Altman himself.

22
00:01:31,080 --> 00:01:35,160
Speaker 1: Eric Schmidt, Google's former CEO, recently compared Altman to Steve

23
00:01:35,240 --> 00:01:38,880
Speaker 1: Jobs and said Open Eyes Board was wrong to can

24
00:01:38,959 --> 00:01:42,880
Speaker 1: him and I quote Schmidt here, these founder CEO types

25
00:01:42,920 --> 00:01:46,800
Speaker 1: are unusual, they're incredibly valuable, and they change the world.

26
00:01:47,160 --> 00:01:50,880
Speaker 1: Schmidt said that at a recent Axios conference. Dave, what

27
00:01:50,920 --> 00:01:54,200
Speaker 1: do you think of that observation about founder CEOs and

28
00:01:54,240 --> 00:01:57,320
Speaker 1: the sort of idea that there's a unicorn executive out

29
00:01:57,320 --> 00:01:59,000
Speaker 1: there who is indispensable.

30
00:01:59,880 --> 00:02:02,480
Speaker 2: I think, look, that's always been Tilicon Valley view of

31
00:02:02,680 --> 00:02:04,880
Speaker 2: you know, great founders and what they can do, and

32
00:02:04,920 --> 00:02:07,800
Speaker 2: this idea that you should follow a founder's vision rather

33
00:02:07,840 --> 00:02:10,080
Speaker 2: than say put it to committee. And you know, Steve

34
00:02:10,120 --> 00:02:13,160
Speaker 2: Job's obviously the most famous example of that. Kicked out

35
00:02:13,160 --> 00:02:16,560
Speaker 2: of Apple, returned again around a decade later, and of course,

36
00:02:16,760 --> 00:02:19,000
Speaker 2: you know, the rest is history, and what happened with

37
00:02:19,040 --> 00:02:22,320
Speaker 2: Apple surely wouldn't have happened without Steve Jobs coming back.

38
00:02:22,680 --> 00:02:24,959
Speaker 2: And we've seen this since again and again, whether it's

39
00:02:25,000 --> 00:02:28,679
Speaker 2: you know, Mark Zuckerberg or elor Musk or any sort

40
00:02:28,680 --> 00:02:31,000
Speaker 2: of tech founder, even people like Travis Kalonick, you know,

41
00:02:31,040 --> 00:02:35,080
Speaker 2: who founded Uber. There's this sense that there's this founder skill,

42
00:02:35,280 --> 00:02:39,520
Speaker 2: this magic, this uniqueness that should really be sort of

43
00:02:39,560 --> 00:02:42,200
Speaker 2: listened to, and although it can be questioned, you know,

44
00:02:42,240 --> 00:02:44,919
Speaker 2: there's a sort of bias towards always backing the founder.

45
00:02:45,280 --> 00:02:47,720
Speaker 2: Speaking of Sam Altman, of the many comments made about

46
00:02:47,800 --> 00:02:50,360
Speaker 2: Altman over the past week or so, I think, you know,

47
00:02:50,400 --> 00:02:53,679
Speaker 2: Paul Graham, who founded why Combinator, where Sam Moultman sort

48
00:02:53,680 --> 00:02:55,640
Speaker 2: of made his name, said a quote that kind of

49
00:02:55,680 --> 00:02:58,359
Speaker 2: rang out around everyone, and he said, you could parachute

50
00:02:58,400 --> 00:03:01,160
Speaker 2: Sam into an island full of cannon, come back in

51
00:03:01,200 --> 00:03:03,280
Speaker 2: five years, and Sam o wont would come out as king.

52
00:03:03,360 --> 00:03:05,200
Speaker 2: And I think that's kind of the view of him

53
00:03:05,200 --> 00:03:07,440
Speaker 2: in particular, and it does sum up, as you say,

54
00:03:07,480 --> 00:03:09,960
Speaker 2: this culture of revering founders like life.

55
00:03:10,000 --> 00:03:12,120
Speaker 1: I don't know actually how to interpret the idea that

56
00:03:12,120 --> 00:03:13,880
Speaker 1: you're going to parachute him onto an island full of

57
00:03:13,919 --> 00:03:16,040
Speaker 1: cannibals and it'll come out the winter, because that kild

58
00:03:16,120 --> 00:03:20,920
Speaker 1: speak to actually just ambition and being extremely as they

59
00:03:20,960 --> 00:03:24,799
Speaker 1: say in Silicon Valley, goal oriented as opposed necessarily having

60
00:03:24,880 --> 00:03:27,200
Speaker 1: innovative genius, and I think, you know, ideally you want

61
00:03:27,240 --> 00:03:30,440
Speaker 1: a combination of both if things are going to get done.

62
00:03:30,080 --> 00:03:31,959
Speaker 3: But you know, if I can just jump in. It's

63
00:03:32,000 --> 00:03:36,560
Speaker 3: interesting you mentioned Paul Graham, Dave, because Paul Graham, I think,

64
00:03:36,640 --> 00:03:39,880
Speaker 3: is part of the reason why this mentality, this almost

65
00:03:39,960 --> 00:03:44,240
Speaker 3: doctrine exists in Silicon Valley. When he started White Combinator,

66
00:03:44,280 --> 00:03:48,280
Speaker 3: he was like a guru among startups, and his teaching,

67
00:03:48,360 --> 00:03:51,600
Speaker 3: as it were, was that the founder is all that matters.

68
00:03:51,680 --> 00:03:52,080
Speaker 2: In a way.

69
00:03:52,120 --> 00:03:55,320
Speaker 3: The technology doesn't matter what a startups technology would it

70
00:03:55,360 --> 00:03:58,320
Speaker 3: has under the hood. What matters is if the founder

71
00:03:58,560 --> 00:04:01,240
Speaker 3: is a visionary hacker to hype who wants to build

72
00:04:01,240 --> 00:04:04,080
Speaker 3: an empire, and if you, as an investor can find

73
00:04:04,160 --> 00:04:07,440
Speaker 3: that kind of person, then you should give them free rein.

74
00:04:07,760 --> 00:04:10,120
Speaker 3: The board should give them free rein, and they should

75
00:04:10,400 --> 00:04:12,880
Speaker 3: be able to do whatever they need to do in

76
00:04:12,960 --> 00:04:16,479
Speaker 3: order to build the empire. And that's why in most cases,

77
00:04:16,480 --> 00:04:20,279
Speaker 3: someone like Mark Zuckerberg has majority voting share the board

78
00:04:20,320 --> 00:04:22,800
Speaker 3: does whatever he wants pretty much, even through the most

79
00:04:22,800 --> 00:04:27,120
Speaker 3: controversial times. Google's founders also had a lot of control

80
00:04:27,120 --> 00:04:29,840
Speaker 3: of the company, outsized control, so I think a lot

81
00:04:29,839 --> 00:04:32,200
Speaker 3: of it comes down to this kind of ideology that

82
00:04:32,760 --> 00:04:35,800
Speaker 3: Paul kind of created and Sam perpetuated when he took

83
00:04:35,800 --> 00:04:37,559
Speaker 3: over as y combinator head.

84
00:04:38,040 --> 00:04:40,920
Speaker 1: Steve Jobs famously had Bill Campbell as a management coach,

85
00:04:40,960 --> 00:04:43,120
Speaker 1: and Bill Campbell was sort of another one of these

86
00:04:43,160 --> 00:04:46,240
Speaker 1: executives who in a different era was a guru to

87
00:04:46,480 --> 00:04:49,520
Speaker 1: various Silicon Valley titans. But I sort of wonder if

88
00:04:49,520 --> 00:04:53,560
Speaker 1: the Bill Campbell advice managerial advice in the Steve Jobs

89
00:04:53,560 --> 00:04:57,400
Speaker 1: era was qualitatively different than the advice that Paul is

90
00:04:57,440 --> 00:05:00,440
Speaker 1: giving a Sam Altman right now. Is that the case

91
00:05:00,520 --> 00:05:03,200
Speaker 1: Parmi has there been an evolution and the kind of

92
00:05:03,240 --> 00:05:09,040
Speaker 1: advice that nascent Silicon Valley titans are hearing from their advisors.

93
00:05:09,440 --> 00:05:11,280
Speaker 3: I have to admit I'm not familiar with the advice

94
00:05:11,360 --> 00:05:14,920
Speaker 3: that Steve Jobs got directly, but I think maybe there

95
00:05:15,000 --> 00:05:18,640
Speaker 3: is a questioning more and more of whether these founder

96
00:05:18,720 --> 00:05:21,919
Speaker 3: types should have all that control. That may still be

97
00:05:22,040 --> 00:05:24,840
Speaker 3: the case at Facebook, but I mean, this is what

98
00:05:24,880 --> 00:05:28,039
Speaker 3: made open ai so unique, right is that it had

99
00:05:28,080 --> 00:05:31,840
Speaker 3: a board that didn't just have teeth, it actually used them,

100
00:05:32,279 --> 00:05:35,000
Speaker 3: and it used them to fire him. So I don't

101
00:05:35,040 --> 00:05:37,880
Speaker 3: know if that's necessarily changing, but it certainly was very,

102
00:05:37,960 --> 00:05:39,039
Speaker 3: very different to open ai.

103
00:05:39,800 --> 00:05:45,520
Speaker 1: Talk a little bit Parmi about Sam Altman's foundational experiences.

104
00:05:45,680 --> 00:05:48,560
Speaker 1: Prior to coming to open ai. He had been an entrepreneur,

105
00:05:48,720 --> 00:05:52,080
Speaker 1: a young computer with and then lands essentially at y

106
00:05:52,160 --> 00:05:55,440
Speaker 1: Combinator and ends up running it. Yeah, or his bona

107
00:05:55,520 --> 00:05:58,280
Speaker 1: fides before he came to open ai.

108
00:05:58,760 --> 00:06:01,200
Speaker 3: Well, he paid his dues as a startup founder who

109
00:06:01,640 --> 00:06:04,600
Speaker 3: created a failed startup, as so many startup founders do.

110
00:06:04,720 --> 00:06:07,320
Speaker 3: He created a startup called looped, worked on it for

111
00:06:07,440 --> 00:06:09,520
Speaker 3: a number of years, devoted his life to it, and

112
00:06:09,560 --> 00:06:12,760
Speaker 3: then eventually sold it to another company, which you might

113
00:06:12,800 --> 00:06:15,000
Speaker 3: say as a success, but in some ways it wasn't

114
00:06:15,040 --> 00:06:17,440
Speaker 3: seen that way. And then he went on to become

115
00:06:17,440 --> 00:06:20,840
Speaker 3: an advisor to Why Combinator, which is like basically one

116
00:06:20,839 --> 00:06:25,000
Speaker 3: of the world's the world's most successful startup accelerator of

117
00:06:25,000 --> 00:06:27,680
Speaker 3: all time. You've had companies like Reddit and Stripe and

118
00:06:27,760 --> 00:06:31,520
Speaker 3: Airbnb all came through that, And you know, he became

119
00:06:31,760 --> 00:06:36,400
Speaker 3: known for having this real ambitious view and pushing startups

120
00:06:36,440 --> 00:06:38,320
Speaker 3: to be more ambitious. So he famously said to the

121
00:06:38,320 --> 00:06:41,800
Speaker 3: founders of Airbnb, you should change the numbers in your

122
00:06:41,839 --> 00:06:45,400
Speaker 3: pitch deck to investors from millions to billions, and then

123
00:06:45,440 --> 00:06:47,480
Speaker 3: he kind of went into this stage where he got

124
00:06:47,520 --> 00:06:51,480
Speaker 3: really into futuristic technology, and he was kind of gazing

125
00:06:51,520 --> 00:06:54,600
Speaker 3: into the future with things like mind uploading and kind

126
00:06:54,600 --> 00:06:58,359
Speaker 3: of uploading his consciousness into cloud computers and exploring the

127
00:06:58,440 --> 00:07:01,720
Speaker 3: universe he's talked about up publicly. And he then started

128
00:07:01,720 --> 00:07:06,800
Speaker 3: investing in very futuristic technology like nuclear fusion, life extension technology,

129
00:07:06,960 --> 00:07:09,520
Speaker 3: and he pushed y Combinator in that direction, But his

130
00:07:09,640 --> 00:07:13,320
Speaker 3: heart really was in AI when he had apparently he

131
00:07:13,480 --> 00:07:16,120
Speaker 3: had this kind of realization at some point in his

132
00:07:16,200 --> 00:07:20,360
Speaker 3: late twenties that humans aren't really all that unique and special,

133
00:07:20,440 --> 00:07:24,120
Speaker 3: and our brains can be simulated by machines. And if

134
00:07:24,120 --> 00:07:27,520
Speaker 3: that's the case, we can build a machine that is

135
00:07:27,720 --> 00:07:30,720
Speaker 3: as unique and powerful as the human brain. And that's

136
00:07:30,760 --> 00:07:33,560
Speaker 3: what he set out to do. But with this extra

137
00:07:34,200 --> 00:07:37,280
Speaker 3: kind of level of idealism around spreading all the benefits

138
00:07:37,320 --> 00:07:38,680
Speaker 3: of that to humanity.

139
00:07:38,720 --> 00:07:40,400
Speaker 1: Well, we could go down so many roads here. This

140
00:07:40,440 --> 00:07:43,200
Speaker 1: reminds me of the Silicon Valley chatter around the Singularity

141
00:07:43,640 --> 00:07:46,239
Speaker 1: that the Google guys were so fascinated with. It's interesting

142
00:07:46,280 --> 00:07:49,680
Speaker 1: to me that if you get enough technological success the

143
00:07:49,720 --> 00:07:52,160
Speaker 1: next step in your thinking is how do you perpetuate

144
00:07:52,240 --> 00:07:55,200
Speaker 1: yourself forever inside of a machine that Maybe that's a

145
00:07:55,240 --> 00:07:58,960
Speaker 1: conversation for another day, Dave. You know, the idea of

146
00:07:59,320 --> 00:08:02,880
Speaker 1: highly tailed motivated people being given free reign so they

147
00:08:02,920 --> 00:08:06,760
Speaker 1: can be creative isn't limited to Silicon Valley. Obviously. There's

148
00:08:06,800 --> 00:08:10,240
Speaker 1: a long history of it in the business world, you know,

149
00:08:10,280 --> 00:08:13,200
Speaker 1: going back to the Rockefellers and Henry Ford and in

150
00:08:13,360 --> 00:08:16,400
Speaker 1: completely different industries. And it's in the arts, right. We

151
00:08:16,520 --> 00:08:20,120
Speaker 1: know that film directors and painters and others are sort

152
00:08:20,120 --> 00:08:23,560
Speaker 1: of seen as masters or mistresses of their own destinies

153
00:08:23,600 --> 00:08:26,640
Speaker 1: and people shouldn't intrude on that. Does all of that

154
00:08:27,280 --> 00:08:30,200
Speaker 1: fairly accrue to Sam Altman in this case? Is he

155
00:08:30,760 --> 00:08:35,680
Speaker 1: someone presiding over a company who needs a board to

156
00:08:35,760 --> 00:08:37,400
Speaker 1: support him but not get in his way.

157
00:08:38,040 --> 00:08:42,400
Speaker 2: Well, I think it's interesting, you know, thinking about Altman's upbringing,

158
00:08:42,440 --> 00:08:45,320
Speaker 2: I guess through why combinates, because what that process did,

159
00:08:45,760 --> 00:08:48,320
Speaker 2: I think is gained him a lot of allies and

160
00:08:48,360 --> 00:08:50,560
Speaker 2: a reputation as being, you know, this real sort of

161
00:08:50,640 --> 00:08:53,480
Speaker 2: network guy in tech, and I think that has allowed

162
00:08:53,520 --> 00:08:56,040
Speaker 2: him to move into these future roles with a certain

163
00:08:56,120 --> 00:08:58,520
Speaker 2: level of trust. I thought it was really remarkable. When

164
00:08:58,520 --> 00:09:01,280
Speaker 2: we were hearing just about being fired, one of the

165
00:09:01,280 --> 00:09:03,480
Speaker 2: people to immediately sort of go on to Twitter was

166
00:09:03,520 --> 00:09:06,680
Speaker 2: Brian Chesky, the CEO of Airbnb, which was a Y

167
00:09:06,760 --> 00:09:11,280
Speaker 2: Combinator company, immediately jumping to sam Untiner's defense and almost

168
00:09:11,360 --> 00:09:13,960
Speaker 2: kind of reporting on the situation himself, which was bizarre.

169
00:09:14,000 --> 00:09:16,520
Speaker 2: So I think, you know, although these founders do get

170
00:09:16,520 --> 00:09:18,800
Speaker 2: this kind of reverence, it does have to be earned,

171
00:09:18,800 --> 00:09:20,559
Speaker 2: and I think the way that Sam Oltman has earned

172
00:09:20,559 --> 00:09:22,520
Speaker 2: that is not so much to be seen as this

173
00:09:22,640 --> 00:09:26,839
Speaker 2: kind of visionary genius necessarily, but this kind of smart networker,

174
00:09:27,000 --> 00:09:28,880
Speaker 2: this smart sort of bringing together of people. And the

175
00:09:28,880 --> 00:09:32,280
Speaker 2: thing that made Y combinators stand out was this really

176
00:09:32,480 --> 00:09:36,080
Speaker 2: fast acceleration. When companies would join y Combinator, they had

177
00:09:36,080 --> 00:09:39,360
Speaker 2: to say a particular metric that they were going to grow,

178
00:09:39,440 --> 00:09:41,600
Speaker 2: and every two weeks they'd checked that it was you know,

179
00:09:41,679 --> 00:09:44,240
Speaker 2: two x four x and so forth. And I think

180
00:09:44,240 --> 00:09:46,280
Speaker 2: that's the ethos that Sam Wontman has brought. Now the

181
00:09:46,320 --> 00:09:48,600
Speaker 2: question is, you know, was that something that needs a

182
00:09:48,600 --> 00:09:50,719
Speaker 2: board to rain in or is that something that a

183
00:09:50,760 --> 00:09:52,680
Speaker 2: board needs to sort of sit back and say, okay,

184
00:09:52,720 --> 00:09:55,400
Speaker 2: off you go and you know, I think sam Moultman

185
00:09:55,480 --> 00:09:58,959
Speaker 2: must have been mindful of the reputation of founders, particularly

186
00:09:58,960 --> 00:10:01,480
Speaker 2: people like Mark Zuckerberg, who you know, the board of

187
00:10:01,520 --> 00:10:03,760
Speaker 2: Facebook could never fire Mark zuckerbo they don't have the

188
00:10:03,800 --> 00:10:06,080
Speaker 2: power to do that. I think he was mindful of

189
00:10:06,200 --> 00:10:07,920
Speaker 2: how that is seen, and I think he was trying

190
00:10:07,960 --> 00:10:10,439
Speaker 2: to make a statement by saying that not only did

191
00:10:10,440 --> 00:10:13,520
Speaker 2: he not have any financial stake in open ai, but

192
00:10:13,679 --> 00:10:16,319
Speaker 2: he was serving as the behest of the boards. As

193
00:10:16,320 --> 00:10:21,520
Speaker 2: it turned out, his reputation ended up being more influential

194
00:10:21,600 --> 00:10:23,640
Speaker 2: than the votes of the board open AI. So even

195
00:10:23,720 --> 00:10:27,160
Speaker 2: once they made that decision, the rallying and support around

196
00:10:27,160 --> 00:10:30,560
Speaker 2: sam Oltman, thanks to his reputation within the tech industry

197
00:10:30,600 --> 00:10:33,839
Speaker 2: built over the past several years, meant that was a

198
00:10:33,880 --> 00:10:36,240
Speaker 2: bigger protection for him than having board votes would have

199
00:10:36,280 --> 00:10:36,760
Speaker 2: been anyway.

200
00:10:37,360 --> 00:10:39,320
Speaker 1: Did you want to jump in there, Permie, Yeah.

201
00:10:39,360 --> 00:10:41,440
Speaker 3: And I think that's a great point about the amount

202
00:10:41,440 --> 00:10:45,199
Speaker 3: of protection that he has. Silicon Valley is such a bubble,

203
00:10:45,360 --> 00:10:48,640
Speaker 3: and if you're a founder, if you're successful and you've

204
00:10:48,640 --> 00:10:51,040
Speaker 3: made a very valuable company, and you're very wealthy and

205
00:10:51,240 --> 00:10:54,360
Speaker 3: you're a billionaire, that you have to behave so badly

206
00:10:54,480 --> 00:10:57,480
Speaker 3: to fall from Silicon Valley's graces, and Elon Musk is

207
00:10:57,520 --> 00:11:01,040
Speaker 3: a prime example of that. So I Meantman's return as

208
00:11:01,080 --> 00:11:03,360
Speaker 3: CEO as a prime example of that. But I just

209
00:11:03,400 --> 00:11:05,280
Speaker 3: wanted to go back to one point you mentioned, Tim.

210
00:11:05,320 --> 00:11:08,199
Speaker 3: You were kind of casting our minds back to the

211
00:11:08,280 --> 00:11:11,280
Speaker 3: Rockefellers and people in the arts who have this kind

212
00:11:11,320 --> 00:11:14,120
Speaker 3: of power as founders and as leaders. But I really

213
00:11:14,160 --> 00:11:17,240
Speaker 3: think that in this situation, someone like Sam Altman and

214
00:11:17,240 --> 00:11:21,040
Speaker 3: the leaders of big tech companies are so different to

215
00:11:21,160 --> 00:11:24,560
Speaker 3: any of that, just because of the scale and wealth

216
00:11:24,600 --> 00:11:28,680
Speaker 3: and resources that these companies have, which are unprecedented in history.

217
00:11:28,920 --> 00:11:33,840
Speaker 3: I mean, Google today reaches something like four billion users worldwide.

218
00:11:33,840 --> 00:11:37,679
Speaker 3: That's half the global population. You know, although Chatgypt has

219
00:11:37,679 --> 00:11:41,079
Speaker 3: something like two hundred million regular users. I think Microsoft services,

220
00:11:41,120 --> 00:11:44,000
Speaker 3: which are going to use more and more open AI technology,

221
00:11:44,559 --> 00:11:46,680
Speaker 3: touch more than a billion people around the world. So

222
00:11:47,040 --> 00:11:50,040
Speaker 3: we're talking about a few people, these founders, with an

223
00:11:50,040 --> 00:11:53,160
Speaker 3: incredible amount of free reign, who have like an incredible

224
00:11:53,200 --> 00:11:56,400
Speaker 3: global influence at the same time, more so than any

225
00:11:56,559 --> 00:11:58,000
Speaker 3: empire even in history.

226
00:11:58,520 --> 00:12:01,640
Speaker 1: And now, of course, the product they have in hand

227
00:12:02,320 --> 00:12:05,920
Speaker 1: is a revolutionary and transformative product. With Google, you know,

228
00:12:06,000 --> 00:12:10,240
Speaker 1: in its pre AI phase, it was a ubiquitous search

229
00:12:10,240 --> 00:12:12,720
Speaker 1: engine that sort of opened the world up to its users.

230
00:12:13,160 --> 00:12:15,720
Speaker 1: No one really worried about it taking the world over.

231
00:12:15,880 --> 00:12:18,040
Speaker 1: You know, it was an information provider. It could be

232
00:12:18,160 --> 00:12:20,920
Speaker 1: a you know, assessed pool of disinformation too. And we'll

233
00:12:20,920 --> 00:12:22,880
Speaker 1: get into this morning. I think our conversations, so I

234
00:12:22,880 --> 00:12:27,920
Speaker 1: think this particular product AI and technology is shaping people's

235
00:12:28,000 --> 00:12:31,320
Speaker 1: views in different ways as well. But Dave, tell me

236
00:12:31,360 --> 00:12:31,840
Speaker 1: what you think.

237
00:12:32,280 --> 00:12:34,480
Speaker 2: Well, I think is also worth stressing in terms of

238
00:12:34,679 --> 00:12:37,240
Speaker 2: Oltman's value to open AI. You know, this is a

239
00:12:37,280 --> 00:12:39,640
Speaker 2: company that people are talking about being valued at around

240
00:12:39,640 --> 00:12:42,480
Speaker 2: eighty six billion dollars, and you know, a lot of

241
00:12:42,480 --> 00:12:45,920
Speaker 2: this is because of the personality of Sam Oltman. The

242
00:12:46,040 --> 00:12:47,800
Speaker 2: value of open II comes from the fact that for

243
00:12:47,840 --> 00:12:50,360
Speaker 2: the past year, every week or so, it seemed like

244
00:12:50,440 --> 00:12:53,040
Speaker 2: Sam Martman was shaking hands with a world leader. It

245
00:12:53,160 --> 00:12:55,480
Speaker 2: was sitting there in front of Congress talking about AI,

246
00:12:55,600 --> 00:12:59,600
Speaker 2: talking about regulation, and so there is value. There's clear

247
00:12:59,679 --> 00:13:01,920
Speaker 2: value of what Sam Mortman is to open AYE. And

248
00:13:01,960 --> 00:13:04,800
Speaker 2: I think it's reputation and the deal with Microsoft and

249
00:13:05,040 --> 00:13:07,600
Speaker 2: the fact that it was and should still be positioned

250
00:13:07,600 --> 00:13:09,800
Speaker 2: as the front runner here is down to the fact

251
00:13:09,800 --> 00:13:13,640
Speaker 2: that he has this ability to communicate what the company

252
00:13:13,640 --> 00:13:16,880
Speaker 2: wants to do clearly, and that doesn't necessarily come all

253
00:13:16,880 --> 00:13:18,920
Speaker 2: that naturally to other people in Silicon Valley. So there

254
00:13:18,960 --> 00:13:21,000
Speaker 2: is an intrinsic value in him himself well, and.

255
00:13:21,040 --> 00:13:23,880
Speaker 1: You know, all of the Steve Jobs comparison, Steve Jobs

256
00:13:23,960 --> 00:13:27,400
Speaker 1: was not an engineer. Steve Jobs wasn't an inventor of technology.

257
00:13:27,400 --> 00:13:31,120
Speaker 1: That was Steve Wozniak, his early partner, and Jobs was

258
00:13:31,200 --> 00:13:34,440
Speaker 1: the person who was capable of verticulating the kind of

259
00:13:34,520 --> 00:13:37,679
Speaker 1: Nirvana vision of a laptop and every desk and then

260
00:13:37,679 --> 00:13:40,120
Speaker 1: a mobile phone in every hand. And he was a

261
00:13:40,160 --> 00:13:42,439
Speaker 1: good manager, and he was good at ultimately at sort

262
00:13:42,440 --> 00:13:45,800
Speaker 1: of harnessing creative people, and that is a skill as well.

263
00:13:45,880 --> 00:13:48,679
Speaker 1: But what still feels very mysterious to me, you guys,

264
00:13:48,760 --> 00:13:51,480
Speaker 1: is why he got pushed out to begin with. Correct

265
00:13:51,480 --> 00:13:52,960
Speaker 1: me on this if I'm not up to date, But

266
00:13:53,040 --> 00:13:56,160
Speaker 1: at this point my understanding is the only kind of

267
00:13:56,200 --> 00:13:58,720
Speaker 1: public reason that's been given for why the board pushed

268
00:13:58,760 --> 00:14:01,760
Speaker 1: him out is that board, in their own words more

269
00:14:01,800 --> 00:14:04,760
Speaker 1: or less, and I'm synopsizing, felt that he wasn't being

270
00:14:04,880 --> 00:14:08,280
Speaker 1: candid with them in his communications with the board, and

271
00:14:08,320 --> 00:14:12,000
Speaker 1: there is now I think an Altman commissioned investigation into

272
00:14:12,440 --> 00:14:14,840
Speaker 1: what led up to that. It's being led by Larry

273
00:14:14,840 --> 00:14:18,400
Speaker 1: Summers and some other board members, which also kind of

274
00:14:18,480 --> 00:14:21,000
Speaker 1: speaks to whether or not that actually be an independent investigation.

275
00:14:21,480 --> 00:14:23,840
Speaker 1: But anyway, how did we get here? Why did the

276
00:14:23,880 --> 00:14:29,800
Speaker 1: board end up feeling concerned about this wonder kid in their.

277
00:14:29,680 --> 00:14:33,840
Speaker 3: Miss Well, I think, first of all, no one still knows,

278
00:14:33,960 --> 00:14:36,760
Speaker 3: apart from it seems the people on the board were,

279
00:14:36,800 --> 00:14:39,600
Speaker 3: and possibly Sam because given a couple of his first

280
00:14:39,600 --> 00:14:42,680
Speaker 3: public interviews, he will not answer questions about why he

281
00:14:42,800 --> 00:14:45,800
Speaker 3: was fired. But I think the reason because it's all

282
00:14:45,800 --> 00:14:48,240
Speaker 3: of speculation now, but it seems like the reason kind

283
00:14:48,240 --> 00:14:50,280
Speaker 3: of goes back to what we were just discussing about

284
00:14:50,280 --> 00:14:54,120
Speaker 3: the outsized influence and power of these tech founders. It

285
00:14:54,160 --> 00:14:58,280
Speaker 3: goes unchecked in most cases, and from what people have

286
00:14:58,360 --> 00:15:01,800
Speaker 3: been saying and reporting, the board uneasy essentially with Sam's

287
00:15:01,840 --> 00:15:05,080
Speaker 3: behavior outside of open Ai and some of these ventures

288
00:15:05,080 --> 00:15:08,040
Speaker 3: that he was pursuing. So for example, the iris scanning

289
00:15:08,080 --> 00:15:12,360
Speaker 3: company world Coin, which aimed to scan the eyes of

290
00:15:12,400 --> 00:15:15,640
Speaker 3: billions of people around the world in order to identify

291
00:15:15,680 --> 00:15:18,720
Speaker 3: them when the Internet becomes flooded with bots, and also

292
00:15:18,760 --> 00:15:22,520
Speaker 3: to help distribute the trillions of dollars of wealth that

293
00:15:22,560 --> 00:15:25,800
Speaker 3: Altman believed would come about through the attainment of AGI.

294
00:15:26,360 --> 00:15:30,560
Speaker 3: He was also pursuing, according to reports, creating a so

295
00:15:30,640 --> 00:15:34,120
Speaker 3: called quote iPhone for AI with Johnny Ive who was

296
00:15:34,160 --> 00:15:37,240
Speaker 3: the former lead designer at Apple. And he was also

297
00:15:37,400 --> 00:15:41,480
Speaker 3: on top of that, reportedly talking to Middle Eastern sovereign

298
00:15:41,520 --> 00:15:44,080
Speaker 3: wealth funds to raise something like ten billion dollars to

299
00:15:44,120 --> 00:15:46,880
Speaker 3: build a chip company. This is a lot of fingers

300
00:15:46,920 --> 00:15:49,760
Speaker 3: to have in different pies, and I think it sounds

301
00:15:49,760 --> 00:15:52,240
Speaker 3: to me like the board was just overall kind of

302
00:15:52,320 --> 00:15:55,840
Speaker 3: uneasy with where he was going with all these different

303
00:15:55,920 --> 00:15:57,880
Speaker 3: ventures and whether he was going to use open AI's

304
00:15:57,960 --> 00:16:00,800
Speaker 3: technology for any of that. So when they say he

305
00:16:00,920 --> 00:16:03,600
Speaker 3: wasn't candid, it seems like they were kind of caught

306
00:16:03,640 --> 00:16:05,200
Speaker 3: on the back foot with some of this stuff, and

307
00:16:05,240 --> 00:16:08,760
Speaker 3: they didn't feel like they were performing their fiduciary duty

308
00:16:08,760 --> 00:16:11,760
Speaker 3: to humanity, which was actually their mission as the board.

309
00:16:12,160 --> 00:16:14,040
Speaker 1: Yeah, I want to get into some of this in

310
00:16:14,080 --> 00:16:17,840
Speaker 1: more depth, you know, specifically this particular board's unusual mission.

311
00:16:18,280 --> 00:16:20,120
Speaker 1: But one of the other things that's been out there, Dave,

312
00:16:20,240 --> 00:16:23,120
Speaker 1: is it's been reported and then dismissed by others that

313
00:16:23,160 --> 00:16:27,720
Speaker 1: there was this mysterious Project Q being hatched inside open

314
00:16:27,800 --> 00:16:31,880
Speaker 1: AI that maybe ran a foul of this idea of

315
00:16:32,080 --> 00:16:35,040
Speaker 1: AI being harnessed ultimately for the good. It's unclear to

316
00:16:35,040 --> 00:16:37,800
Speaker 1: me but that the board had concerns about this product.

317
00:16:38,280 --> 00:16:40,840
Speaker 1: Others have said, actually, there were no letters written about it.

318
00:16:40,840 --> 00:16:43,320
Speaker 1: So the reporting has been very murky around this, and

319
00:16:43,400 --> 00:16:44,960
Speaker 1: I don't know that we should rely on it, but

320
00:16:45,040 --> 00:16:49,640
Speaker 1: it gets to this issue of the board being concerned about,

321
00:16:49,760 --> 00:16:51,920
Speaker 1: you know, the idea of bots controlling the world and

322
00:16:51,960 --> 00:16:55,520
Speaker 1: open an AI losing its mission along the way.

323
00:16:56,040 --> 00:16:58,720
Speaker 2: Right right, I mean the ultimate endgame concern here is

324
00:16:58,760 --> 00:17:01,920
Speaker 2: that an AIO created that poses a real risk to

325
00:17:02,000 --> 00:17:05,160
Speaker 2: the human race, and any step along that journey makes

326
00:17:05,160 --> 00:17:08,199
Speaker 2: people pretty pretty nervous. Now, the question around whether this

327
00:17:08,359 --> 00:17:11,639
Speaker 2: was a factor in Oltmand's firing, as you say, that

328
00:17:11,760 --> 00:17:14,000
Speaker 2: still mrk it at this point, and I think actually

329
00:17:14,000 --> 00:17:16,760
Speaker 2: it sort of stresses this issue, which is that had

330
00:17:16,800 --> 00:17:20,320
Speaker 2: the board that forced Oltman out, had it had reasoning

331
00:17:20,480 --> 00:17:23,080
Speaker 2: and public explanations for why it did, that we could

332
00:17:23,080 --> 00:17:25,840
Speaker 2: have avoided what's now going to be, you know, this

333
00:17:26,040 --> 00:17:29,840
Speaker 2: endless filling of the void with possibly conspiracies of other

334
00:17:29,920 --> 00:17:33,240
Speaker 2: you know, sort of hyperboles around what open ai is developing. Now.

335
00:17:33,320 --> 00:17:37,360
Speaker 2: Q Star, this sort of rumored project that supposedly can

336
00:17:37,600 --> 00:17:40,879
Speaker 2: solve basic math problems. I mean, for starters, calling it

337
00:17:40,960 --> 00:17:44,000
Speaker 2: Q anything seems like a mistake. It's very bonred like,

338
00:17:44,480 --> 00:17:47,120
Speaker 2: I mean, it's remarkable. And just days later we Amazon

339
00:17:47,200 --> 00:17:49,600
Speaker 2: named one of it's a's Q as well, which I

340
00:17:49,640 --> 00:17:52,360
Speaker 2: just find staggering that these companies would be so sort

341
00:17:52,359 --> 00:17:54,720
Speaker 2: of mindless in doing that. But that's another that's another

342
00:17:54,760 --> 00:17:55,440
Speaker 2: matter altogether.

343
00:17:55,600 --> 00:17:58,600
Speaker 1: It's like naming something X well, right.

344
00:17:58,600 --> 00:18:00,680
Speaker 2: Yeah, I mean you might as well call killer AI.

345
00:18:00,800 --> 00:18:02,520
Speaker 2: I mean, it's just going to make people sort of

346
00:18:02,560 --> 00:18:04,920
Speaker 2: nervous or just conspiratorial, which you know, we have enough

347
00:18:04,960 --> 00:18:07,360
Speaker 2: of that right now in the world, of course. But look,

348
00:18:07,400 --> 00:18:08,879
Speaker 2: so the issue is, you know, it's very hard to

349
00:18:08,920 --> 00:18:11,360
Speaker 2: sort of sometimes for people who aren't experts in AI

350
00:18:11,440 --> 00:18:13,920
Speaker 2: to draw this line between something that might be able

351
00:18:13,960 --> 00:18:16,760
Speaker 2: to do basic math and something that might destroy the universe.

352
00:18:17,160 --> 00:18:20,280
Speaker 2: But supposedly there was some concern within open ai that

353
00:18:20,680 --> 00:18:23,480
Speaker 2: this was you know, a significant break through, a significant step,

354
00:18:23,960 --> 00:18:26,560
Speaker 2: and you know, speaking of x Elon Musk was in

355
00:18:26,640 --> 00:18:29,960
Speaker 2: New York recently talking about the firing of Samultman, and

356
00:18:30,000 --> 00:18:32,840
Speaker 2: he made what I thought was unusually lately a pretty

357
00:18:32,880 --> 00:18:35,280
Speaker 2: reasonable point. He was saying, Look, either the board needs

358
00:18:35,320 --> 00:18:38,480
Speaker 2: to be clear about these concerns and allow the public

359
00:18:38,520 --> 00:18:40,359
Speaker 2: to know what they were, or if there wasn't a

360
00:18:40,400 --> 00:18:42,359
Speaker 2: good reason, they need to you know, well they have

361
00:18:42,440 --> 00:18:44,240
Speaker 2: stood down now because then you know they needed to

362
00:18:44,240 --> 00:18:46,760
Speaker 2: be accountable for that error. But I think either way

363
00:18:46,800 --> 00:18:49,119
Speaker 2: there needs to be a feeling of this vacuum around

364
00:18:49,560 --> 00:18:52,479
Speaker 2: what exactly the concerns may have been. Because still one

365
00:18:52,480 --> 00:18:54,480
Speaker 2: of the big unknowns in all this is, you know,

366
00:18:54,520 --> 00:18:58,080
Speaker 2: one of open AI's co founders, Ilias skav who was

367
00:18:58,119 --> 00:19:00,240
Speaker 2: the chief scientist open AI and was on the board,

368
00:19:00,280 --> 00:19:03,159
Speaker 2: and he was the sort of pivotal changed mind on

369
00:19:03,200 --> 00:19:05,320
Speaker 2: the board that meant the board had a majority to

370
00:19:05,400 --> 00:19:08,639
Speaker 2: fire Altman. We don't quite know what it was that

371
00:19:08,720 --> 00:19:10,760
Speaker 2: concerned him so much or why he changed his mind.

372
00:19:10,800 --> 00:19:14,480
Speaker 2: I mean, he's subsequently come along and said he regretted

373
00:19:14,960 --> 00:19:18,120
Speaker 2: the decision, and Sam Moultman, in part of his statement

374
00:19:18,160 --> 00:19:20,439
Speaker 2: about returning to the company, Sam Moltman said that he

375
00:19:20,520 --> 00:19:23,200
Speaker 2: harbors no ill will towards him, And you know, I've

376
00:19:23,200 --> 00:19:26,280
Speaker 2: never heard that phrase said with much sincerity. So I'm

377
00:19:26,320 --> 00:19:28,679
Speaker 2: kind of keen to see where he lives. Yeah, I

378
00:19:28,680 --> 00:19:31,359
Speaker 2: mean it's like a football owner saying they backed the manager.

379
00:19:31,400 --> 00:19:31,520
Speaker 3: You know.

380
00:19:31,560 --> 00:19:34,080
Speaker 2: It's sort of ominous sign that things might not be

381
00:19:34,119 --> 00:19:36,640
Speaker 2: so well. And so this is one of these details

382
00:19:37,119 --> 00:19:39,480
Speaker 2: that we need to sort of hear more about. And

383
00:19:39,560 --> 00:19:41,920
Speaker 2: I think open AYE, if it wants to be at

384
00:19:41,960 --> 00:19:44,520
Speaker 2: the front of talking to governments and being this hugely

385
00:19:44,560 --> 00:19:48,159
Speaker 2: an entry company, the trade off has to be that

386
00:19:48,240 --> 00:19:51,040
Speaker 2: it is more public, perhaps than tech companies would typically

387
00:19:51,119 --> 00:19:53,680
Speaker 2: be in the past. It doesn't get to hide behind

388
00:19:53,720 --> 00:19:56,639
Speaker 2: this super secrecy that Silicon Valley is more comfortable with.

389
00:19:56,880 --> 00:19:59,200
Speaker 2: It has to tell us what was going on here.

390
00:19:59,200 --> 00:20:01,119
Speaker 1: We should also know. And Elon Musk says he was

391
00:20:01,160 --> 00:20:05,520
Speaker 1: an early board member at opmen AI along with Sam Altman.

392
00:20:05,560 --> 00:20:09,200
Speaker 1: Before Sam became CEO, he was on the board and

393
00:20:09,240 --> 00:20:12,120
Speaker 1: as Parmi noted earlier, Elon has run rough shot over

394
00:20:12,119 --> 00:20:14,240
Speaker 1: his own board. We're going to take a break in

395
00:20:14,280 --> 00:20:16,760
Speaker 1: a sec But this idea that AI is going to

396
00:20:16,800 --> 00:20:19,840
Speaker 1: take over the world and that open AI is sitting

397
00:20:19,880 --> 00:20:22,879
Speaker 1: on top of a Pandora's box, is that warranted. Parmi.

398
00:20:23,560 --> 00:20:25,359
Speaker 3: First of all, I would say an answer to that

399
00:20:25,480 --> 00:20:30,119
Speaker 3: question that I think open ai secretly loves that people

400
00:20:30,200 --> 00:20:34,720
Speaker 3: think that because it makes their technology. However, but they

401
00:20:34,800 --> 00:20:37,240
Speaker 3: keep it under wraps, it makes their technology seem that

402
00:20:37,359 --> 00:20:40,120
Speaker 3: much more powerful. You know, for people to fear it

403
00:20:40,160 --> 00:20:42,840
Speaker 3: doesn't mean they don't want to pay for it necessarily.

404
00:20:42,960 --> 00:20:45,879
Speaker 3: Companies kind of see that almost as a kind of

405
00:20:46,040 --> 00:20:50,320
Speaker 3: nice attractive feature to software. But I think I would

406
00:20:50,400 --> 00:20:54,320
Speaker 3: say that with something like this new development with q starr,

407
00:20:54,400 --> 00:20:57,120
Speaker 3: for example, that open ai is working on this ability

408
00:20:57,160 --> 00:21:00,719
Speaker 3: to do great school math. You so this means that

409
00:21:00,760 --> 00:21:04,360
Speaker 3: AI can reason, because when you do math, you have

410
00:21:04,440 --> 00:21:06,919
Speaker 3: to figure out steps to solve a problem, and it's

411
00:21:06,960 --> 00:21:09,879
Speaker 3: a little bit like solving problems in real life. And

412
00:21:09,960 --> 00:21:15,360
Speaker 3: so right now, chatgybt is amazing because it can generate language,

413
00:21:15,359 --> 00:21:17,359
Speaker 3: it can generate text, and a lot of that is

414
00:21:17,480 --> 00:21:21,560
Speaker 3: just statistical based predictions. But if a model can do

415
00:21:21,600 --> 00:21:24,520
Speaker 3: things much more strategically, if it can plan, if it

416
00:21:24,560 --> 00:21:28,399
Speaker 3: can solve problems, many people say in the AI field

417
00:21:28,400 --> 00:21:32,080
Speaker 3: that this is a step towards more general intelligence. And

418
00:21:32,160 --> 00:21:34,679
Speaker 3: the reason why I'm saying that in response to your

419
00:21:34,760 --> 00:21:38,480
Speaker 3: question is because although I don't think AGI is necessarily

420
00:21:38,520 --> 00:21:41,080
Speaker 3: going to take over the world. I personally don't subscribe

421
00:21:41,119 --> 00:21:44,800
Speaker 3: to the rogue AI theory, certainly not anytime soon. But

422
00:21:44,920 --> 00:21:48,960
Speaker 3: this idea of humans outsourcing more than just tasks, but

423
00:21:49,080 --> 00:21:53,840
Speaker 3: actual responsibility to AI is certainly going to be happening

424
00:21:53,880 --> 00:21:56,840
Speaker 3: in the next couple of years with these kinds of

425
00:21:56,880 --> 00:22:01,480
Speaker 3: developments like q Star, and also Google's new model called Gemini,

426
00:22:01,560 --> 00:22:06,000
Speaker 3: which hasn't released yet, also has the sexpertise and strategic planning,

427
00:22:06,280 --> 00:22:08,320
Speaker 3: so it's kind of similar to what open ai is

428
00:22:08,359 --> 00:22:10,840
Speaker 3: working on, and I think that's going to be really

429
00:22:10,880 --> 00:22:14,040
Speaker 3: interesting to see how business leaders, how anyone uses these

430
00:22:14,119 --> 00:22:17,320
Speaker 3: kinds of systems not just to generate text or to

431
00:22:17,359 --> 00:22:20,080
Speaker 3: ask advice, but to actually carry out some of our

432
00:22:20,200 --> 00:22:22,800
Speaker 3: day to day work tasks and responsibilities.

433
00:22:24,000 --> 00:22:25,680
Speaker 1: On that note, we're going to take a quick break

434
00:22:25,680 --> 00:22:27,600
Speaker 1: to hear from a sponsor and then we'll come right back.

435
00:22:33,600 --> 00:22:35,800
Speaker 1: We're back with parme Olsen and Dave Lee, and we're

436
00:22:35,840 --> 00:22:40,160
Speaker 1: discussing all of the recent upheaval surrounding open ai and

437
00:22:40,200 --> 00:22:44,840
Speaker 1: its impact on its product chat GPT. Dave sam Altman

438
00:22:45,040 --> 00:22:49,120
Speaker 1: was the CEO and on the board at OpenAI. He's

439
00:22:49,160 --> 00:22:52,800
Speaker 1: now just the CEO. How do we interpret that? Did

440
00:22:52,800 --> 00:22:56,600
Speaker 1: he win the power struggle that ensued after he was

441
00:22:56,600 --> 00:22:59,399
Speaker 1: forced out and brought back. Does he have more influence now?

442
00:22:59,440 --> 00:23:02,320
Speaker 1: Does he as less influence? Can we even know?

443
00:23:03,320 --> 00:23:05,119
Speaker 2: Yeah? I mean, look, I think a lot depends on

444
00:23:05,160 --> 00:23:06,800
Speaker 2: a couple of things. I mean, one of them being

445
00:23:07,160 --> 00:23:10,000
Speaker 2: how does the rest of this board look like in

446
00:23:10,080 --> 00:23:12,440
Speaker 2: the coming weeks and months. You know, the interim board

447
00:23:12,480 --> 00:23:14,560
Speaker 2: of three is going to expand possibly through as many

448
00:23:14,560 --> 00:23:17,199
Speaker 2: as nine or they haven't quite confirmed the shape up

449
00:23:17,240 --> 00:23:19,800
Speaker 2: of that, and there's going to be lots of sort

450
00:23:19,800 --> 00:23:21,719
Speaker 2: of questions that's about what the makeup of that board's

451
00:23:21,760 --> 00:23:24,280
Speaker 2: going to be. I mean, already the board has lost

452
00:23:24,320 --> 00:23:27,160
Speaker 2: its only women. It's now an all male interim boards,

453
00:23:27,119 --> 00:23:29,320
Speaker 2: So there's going to be pressure to reflect a diversity there.

454
00:23:29,320 --> 00:23:31,280
Speaker 2: But then also there's going to have to be people

455
00:23:31,280 --> 00:23:36,120
Speaker 2: who are incredibly versed in safety round AI, around policy

456
00:23:36,160 --> 00:23:38,600
Speaker 2: and AI, people that are going to be at least

457
00:23:38,640 --> 00:23:41,520
Speaker 2: seen as potentially being able to hold someone like Sam

458
00:23:41,560 --> 00:23:43,760
Speaker 2: Altman to account or at least sort of reign in

459
00:23:43,760 --> 00:23:46,760
Speaker 2: perhaps some of those commercial interests that might sort of

460
00:23:46,880 --> 00:23:49,360
Speaker 2: provide conflicts or just kind of have him going full

461
00:23:49,400 --> 00:23:51,639
Speaker 2: steam ahead. So I think it we'll know more about

462
00:23:51,640 --> 00:23:54,360
Speaker 2: sam Altman's power. Once we know more about that board.

463
00:23:54,520 --> 00:23:57,240
Speaker 1: Party you were mentioning earlier, how sam Alton was allowed

464
00:23:57,280 --> 00:24:00,600
Speaker 1: to pursue all these other ventures outside of open Ai,

465
00:24:00,760 --> 00:24:04,160
Speaker 1: including companies that possibly would have ended up as competitors

466
00:24:04,640 --> 00:24:08,000
Speaker 1: getting funding for those rather than possibly getting funding for

467
00:24:08,280 --> 00:24:12,600
Speaker 1: open Ai, just all of the myriad financial and professional conflicts.

468
00:24:12,720 --> 00:24:15,640
Speaker 1: One of the other things that I find very strange

469
00:24:15,640 --> 00:24:19,200
Speaker 1: in the story is that open ai is a nonprofit

470
00:24:19,359 --> 00:24:24,960
Speaker 1: company housing a for profit subsidiary that offers a product

471
00:24:25,080 --> 00:24:29,760
Speaker 1: Chat GPT for free unless you want extra Chat GPT features,

472
00:24:30,000 --> 00:24:31,879
Speaker 1: and then as a consumer, you have to pay extra

473
00:24:31,920 --> 00:24:36,320
Speaker 1: to get those. This confuses me, even in that structure

474
00:24:36,320 --> 00:24:38,520
Speaker 1: at one point, the company at evaluation of about eighty

475
00:24:38,560 --> 00:24:42,280
Speaker 1: six billion dollars, isn't this structure in it of itself

476
00:24:42,480 --> 00:24:43,720
Speaker 1: bound to cause problems.

477
00:24:44,480 --> 00:24:47,600
Speaker 3: One thing I underestimated about this whole story when it

478
00:24:47,680 --> 00:24:51,280
Speaker 3: happened was how much power this board had, And I

479
00:24:51,359 --> 00:24:54,440
Speaker 3: underestimated and didn't fully appreciate the fact that they really

480
00:24:54,480 --> 00:24:58,520
Speaker 3: could fire sam Altman. But the idea of trying to

481
00:24:58,560 --> 00:25:02,800
Speaker 3: combine for profit the nonprofit is not completely unusual In

482
00:25:02,880 --> 00:25:06,760
Speaker 3: Silicon Valley. Mozilla is connected with the Mozilla Foundation that

483
00:25:06,880 --> 00:25:11,160
Speaker 3: is a nonprofit, Signal, which is the encrypted messaging app

484
00:25:11,200 --> 00:25:15,320
Speaker 3: that is also a nonprofit, and even just recently, two

485
00:25:15,440 --> 00:25:18,399
Speaker 3: of the leading AI firms in Silicon Valley. One is

486
00:25:18,440 --> 00:25:21,240
Speaker 3: called Inflection, the other one is Anthropic, the latter of

487
00:25:21,240 --> 00:25:24,280
Speaker 3: which split away from open Ai a couple of years ago.

488
00:25:24,560 --> 00:25:29,280
Speaker 3: Those are also trying to thread this needle between being

489
00:25:29,359 --> 00:25:32,879
Speaker 3: businesses that make money but also building AI that is

490
00:25:32,920 --> 00:25:35,960
Speaker 3: safe and beneficial to humanity, and so they are also

491
00:25:36,119 --> 00:25:39,720
Speaker 3: tinkering with these kind of unusual corporate structures, like being

492
00:25:39,760 --> 00:25:43,879
Speaker 3: a public benefit corporation. I think in the case of Inflection,

493
00:25:44,160 --> 00:25:48,280
Speaker 3: they're structured in such a way where they have to

494
00:25:48,640 --> 00:25:53,000
Speaker 3: prioritize the environment and consumers on the exact same level

495
00:25:53,080 --> 00:25:56,560
Speaker 3: as their investors. So profit is not the number one

496
00:25:56,720 --> 00:25:58,760
Speaker 3: priority for them. It has to be upheld with these

497
00:25:58,760 --> 00:26:02,399
Speaker 3: other things now, and this is clearly very difficult to

498
00:26:02,480 --> 00:26:05,160
Speaker 3: pull off open Ai. It doesn't seem to have worked

499
00:26:05,160 --> 00:26:08,560
Speaker 3: for them. Deep Mind, which is the big AI lab

500
00:26:08,600 --> 00:26:11,080
Speaker 3: that's part of Google, also tried to do this. They

501
00:26:11,160 --> 00:26:14,400
Speaker 3: tried to spin out of Google for several years. They

502
00:26:14,440 --> 00:26:18,040
Speaker 3: tried to create a governance structure called a Global Interest Company.

503
00:26:18,400 --> 00:26:20,919
Speaker 3: They wanted to be like a nonprofit style company. It

504
00:26:21,000 --> 00:26:25,399
Speaker 3: totally failed. Google next it. So there's this history of

505
00:26:25,600 --> 00:26:28,399
Speaker 3: AI builders trying to do this because they know this

506
00:26:28,440 --> 00:26:33,040
Speaker 3: technology is so transformative and they don't necessarily feel comfortable

507
00:26:33,080 --> 00:26:37,600
Speaker 3: even with it being controlled by monopolistic corporations essentially, but

508
00:26:37,640 --> 00:26:39,520
Speaker 3: it's been very difficult to figure out how to make

509
00:26:39,560 --> 00:26:40,000
Speaker 3: it work.

510
00:26:40,359 --> 00:26:41,960
Speaker 2: I think I was sop on. Isn't it the case

511
00:26:41,960 --> 00:26:45,520
Speaker 2: that this model of a nonprofit running a technology product,

512
00:26:45,800 --> 00:26:47,880
Speaker 2: it comes under a lot more strain given the sheer

513
00:26:47,880 --> 00:26:50,479
Speaker 2: amounts of money necessary to pull it off. With aim

514
00:26:50,880 --> 00:26:53,879
Speaker 2: absolutely eating power Microsoft place. I think it was up

515
00:26:53,880 --> 00:26:57,680
Speaker 2: to thirteen billion dollars to have open ai use its

516
00:26:57,720 --> 00:27:00,679
Speaker 2: servers to do the crunching that makes AI possible. So,

517
00:27:01,280 --> 00:27:04,119
Speaker 2: you know, Mozilla is a great example there of an

518
00:27:04,200 --> 00:27:07,199
Speaker 2: organization that runs a browser, but creating and running a

519
00:27:07,200 --> 00:27:09,840
Speaker 2: browser isn't in the same league as having to build

520
00:27:10,520 --> 00:27:12,959
Speaker 2: cutting edge AI. And even in the case of Mozilla,

521
00:27:13,000 --> 00:27:14,919
Speaker 2: you know, much of their funding comes from companies like

522
00:27:14,960 --> 00:27:18,080
Speaker 2: Google and others. So it's always been a bit uneasy

523
00:27:18,160 --> 00:27:22,760
Speaker 2: this tech model with nonprofits, and it's been particularly strained

524
00:27:23,040 --> 00:27:25,400
Speaker 2: when it comes to AI just because of the magnitude

525
00:27:25,440 --> 00:27:27,000
Speaker 2: of what needs to be done well.

526
00:27:27,000 --> 00:27:29,720
Speaker 1: And you also have, you know, a chief executive whose

527
00:27:29,720 --> 00:27:32,720
Speaker 1: mission is to grow the company and increase profitability and

528
00:27:32,800 --> 00:27:36,040
Speaker 1: expand the share of the product that's selling. And then

529
00:27:36,080 --> 00:27:40,000
Speaker 1: a board which is invested with responsibilities for helping to

530
00:27:40,000 --> 00:27:43,680
Speaker 1: achieve that, but also being kind of a self regulatory

531
00:27:43,760 --> 00:27:46,639
Speaker 1: device that is trying to stand in the way of abuses.

532
00:27:46,720 --> 00:27:49,000
Speaker 1: All of it, Parmi was saying, And you know, I

533
00:27:49,000 --> 00:27:52,560
Speaker 1: think this is the historical tension anyway between boards and executives.

534
00:27:52,680 --> 00:27:55,520
Speaker 1: It happens in lots of industries. Boards came into being

535
00:27:56,320 --> 00:28:00,159
Speaker 1: to sort of make sure, especially publicly traded companies, that

536
00:28:00,240 --> 00:28:02,560
Speaker 1: investors interests were being looked after and that you didn't

537
00:28:02,600 --> 00:28:06,120
Speaker 1: have rogue CEOs. And there's a lot of situations even

538
00:28:06,119 --> 00:28:07,960
Speaker 1: outside of tech, where you can have a rogue CEO.

539
00:28:08,040 --> 00:28:11,760
Speaker 1: But of course, and again, as Party mentioned earlier, we're

540
00:28:11,760 --> 00:28:15,000
Speaker 1: talking about a product and a technology at play here

541
00:28:15,040 --> 00:28:18,960
Speaker 1: that scares people. You know, the board, Parmi continues to

542
00:28:19,040 --> 00:28:21,840
Speaker 1: interest me because, as you've noted in one of your columns,

543
00:28:22,359 --> 00:28:24,840
Speaker 1: open Ai had a chance to get more women on

544
00:28:24,880 --> 00:28:27,680
Speaker 1: its board and it just drove right past that, which

545
00:28:27,720 --> 00:28:30,320
Speaker 1: now has an all male board. One of the people

546
00:28:30,840 --> 00:28:34,679
Speaker 1: on the board that was reconstituted, Helen Turner, interested me

547
00:28:34,800 --> 00:28:37,879
Speaker 1: because Helen Toner is an academic. She has done a

548
00:28:37,880 --> 00:28:41,440
Speaker 1: lot of work herself on AI and the uses of technology,

549
00:28:41,920 --> 00:28:44,680
Speaker 1: and it has been reported that she had worked on

550
00:28:44,720 --> 00:28:49,040
Speaker 1: a paper that raised some questions about even whether or

551
00:28:49,080 --> 00:28:52,600
Speaker 1: not open AI's own products could be problematic, and that

552
00:28:52,680 --> 00:28:55,959
Speaker 1: Sam Altman confronted her about that and they had a dispute.

553
00:28:56,360 --> 00:28:59,600
Speaker 1: She was a pivotal person in his departure based on

554
00:28:59,680 --> 00:29:03,000
Speaker 1: the report we've seen. Tell me about those two things,

555
00:29:03,080 --> 00:29:04,840
Speaker 1: about the fact that we have an all male board

556
00:29:04,840 --> 00:29:06,880
Speaker 1: there now and what happened to Helen Toner.

557
00:29:07,480 --> 00:29:10,520
Speaker 3: Well, I'll start with Toner, and absolutely right. She did

558
00:29:10,680 --> 00:29:12,880
Speaker 3: write that research paper. You can see it. It's public

559
00:29:12,920 --> 00:29:16,360
Speaker 3: and has written in association with Georgetown University, and she's

560
00:29:16,440 --> 00:29:21,000
Speaker 3: scathing about open Ai. She referred to frantic corner cutting

561
00:29:21,240 --> 00:29:24,680
Speaker 3: in the progress of building chat Ept, and then she

562
00:29:24,800 --> 00:29:29,960
Speaker 3: compared open Ai with its rival Anthropic and said that Anthropic,

563
00:29:30,080 --> 00:29:32,280
Speaker 3: which was the group of people at open Ai that

564
00:29:32,320 --> 00:29:35,400
Speaker 3: broke away to make Safer AI, had done a better

565
00:29:35,520 --> 00:29:38,840
Speaker 3: job of more slowly deploying a product, making sure that

566
00:29:38,880 --> 00:29:41,320
Speaker 3: it was safe before they put it out into the wild,

567
00:29:41,840 --> 00:29:45,960
Speaker 3: and sam Altmand apparently was livid about that. Didn't like

568
00:29:46,040 --> 00:29:48,280
Speaker 3: the fact that she was putting that into the public

569
00:29:48,320 --> 00:29:51,640
Speaker 3: domain when you know the open ai is being investigated

570
00:29:51,680 --> 00:29:55,920
Speaker 3: by the Federal Trade Commission for potentially infringing on people's privacy.

571
00:29:56,000 --> 00:29:59,200
Speaker 3: He told her, according to reports, that it was compromising

572
00:29:59,480 --> 00:30:02,120
Speaker 3: the company, and so, yes, there was a lot of

573
00:30:02,160 --> 00:30:04,680
Speaker 3: tension there, but I think tim, I mean, you could

574
00:30:04,720 --> 00:30:06,600
Speaker 3: look at it both ways. Yes, what she was doing

575
00:30:06,720 --> 00:30:10,440
Speaker 3: was compromising open AI's reputation, but she was an academic.

576
00:30:10,520 --> 00:30:12,719
Speaker 3: They knew that when they brought her on the board.

577
00:30:13,200 --> 00:30:15,400
Speaker 3: And as you also mentioned, this is kind of what

578
00:30:15,440 --> 00:30:17,680
Speaker 3: boards are supposed to do, push back a little bit,

579
00:30:18,040 --> 00:30:21,720
Speaker 3: and boards typically have a fiduciary duty to shareholders. This

580
00:30:21,880 --> 00:30:24,880
Speaker 3: was not the case with open Ai. The board had

581
00:30:24,880 --> 00:30:29,840
Speaker 3: a fiduciary duty to the open ai mission to benefit humanity.

582
00:30:29,880 --> 00:30:31,640
Speaker 3: It's kind of, if you think about it, quite a

583
00:30:31,680 --> 00:30:34,640
Speaker 3: bizarre phrase. Feels weird saying that, but that is literally

584
00:30:34,680 --> 00:30:37,680
Speaker 3: how it was written and signed and agreed by everyone.

585
00:30:38,160 --> 00:30:40,800
Speaker 3: But Toner was the one that lost her seat, and

586
00:30:40,840 --> 00:30:42,320
Speaker 3: so was Tasha McCauley.

587
00:30:42,800 --> 00:30:45,160
Speaker 1: It's smacks of Sir Gey Brand and Larry Page do

588
00:30:45,320 --> 00:30:47,960
Speaker 1: no harm in their early Google literation.

589
00:30:48,160 --> 00:30:51,760
Speaker 3: Absolutely yeah. And Google also famously said in the very beginning,

590
00:30:51,800 --> 00:30:54,880
Speaker 3: we don't want advertising, we hate advertising. And now they're

591
00:30:54,920 --> 00:30:56,160
Speaker 3: the world's biggest ad giant.

592
00:30:56,240 --> 00:30:59,040
Speaker 1: So there you go, Ah, the children, but we need

593
00:30:59,080 --> 00:31:01,920
Speaker 1: them among us. Whatever the merits are of how the

594
00:31:01,960 --> 00:31:04,440
Speaker 1: board intervened here, and I have sympathy for a lot

595
00:31:04,480 --> 00:31:07,560
Speaker 1: of them, it also had quite the clown show aspects

596
00:31:07,600 --> 00:31:09,880
Speaker 1: to the whole thing. You sort of wondered why this

597
00:31:09,920 --> 00:31:13,880
Speaker 1: could have been choreographed or sorted in a more private way,

598
00:31:14,080 --> 00:31:17,200
Speaker 1: but maybe that was impossible. Here is there another way

599
00:31:17,240 --> 00:31:18,680
Speaker 1: this could have played out at open AI?

600
00:31:19,440 --> 00:31:22,160
Speaker 2: I mean yes, And the way it could have played

601
00:31:22,160 --> 00:31:25,880
Speaker 2: out was the phrase that was bringing your receipts. If

602
00:31:25,880 --> 00:31:28,360
Speaker 2: the board had this feeling that there was issues with

603
00:31:28,440 --> 00:31:32,400
Speaker 2: how Sam Wontman had been communicating with them, it needed examples.

604
00:31:32,480 --> 00:31:35,160
Speaker 2: It needed a way to sort of explain its thinking

605
00:31:35,160 --> 00:31:36,760
Speaker 2: in a way it just hasn't done. It kind of

606
00:31:36,760 --> 00:31:39,000
Speaker 2: reminds me of when you have a sort of argument

607
00:31:39,080 --> 00:31:40,760
Speaker 2: with a spouse and you kind of well, give me

608
00:31:40,760 --> 00:31:42,240
Speaker 2: one example, and you go, well, I can't think of

609
00:31:42,240 --> 00:31:44,560
Speaker 2: one right now, but you never quite get anywhere, do you,

610
00:31:44,560 --> 00:31:47,320
Speaker 2: And nobody gives any concession on either side, and that's

611
00:31:47,360 --> 00:31:49,560
Speaker 2: kind of where we're at with Open AI. It was

612
00:31:49,960 --> 00:31:52,080
Speaker 2: being described as a coup, and I think that's kind

613
00:31:52,120 --> 00:31:54,120
Speaker 2: of accurate, and that there was a sort of swing

614
00:31:54,160 --> 00:31:56,240
Speaker 2: for the king and you've got to have everything in

615
00:31:56,280 --> 00:31:58,600
Speaker 2: a row and if you swing, you better not miss,

616
00:31:58,600 --> 00:32:00,200
Speaker 2: and that's what they did. They swung and they and

617
00:32:00,200 --> 00:32:02,200
Speaker 2: they missed because they didn't have the reason to back

618
00:32:02,200 --> 00:32:04,240
Speaker 2: it up. And you know, as we've now discovered as well,

619
00:32:04,280 --> 00:32:06,960
Speaker 2: they were fighting the forces of the commercial interest and

620
00:32:07,040 --> 00:32:10,200
Speaker 2: open AI, so you could argue they perhaps never stood

621
00:32:10,240 --> 00:32:12,680
Speaker 2: a chance. But I spoke to one the business professor

622
00:32:12,680 --> 00:32:15,440
Speaker 2: recently who made the point that talking about this sort

623
00:32:15,440 --> 00:32:20,440
Speaker 2: of nonprofit management and so forth, we shouldn't necessarily write

624
00:32:20,480 --> 00:32:22,640
Speaker 2: off this model because what this could have just been

625
00:32:22,840 --> 00:32:26,320
Speaker 2: was in competency in this board in particular when it

626
00:32:26,360 --> 00:32:28,920
Speaker 2: came to carrying out this coup or whatever we want

627
00:32:28,960 --> 00:32:29,480
Speaker 2: to describe it.

628
00:32:29,520 --> 00:32:31,560
Speaker 3: I completely agree with that, and I think that's what's

629
00:32:31,600 --> 00:32:34,080
Speaker 3: such a shame about the whole thing, because it does

630
00:32:34,120 --> 00:32:36,240
Speaker 3: sound like they did come across as quite incompetent, and

631
00:32:36,280 --> 00:32:37,719
Speaker 3: if you look at them they didn't have a lot

632
00:32:37,760 --> 00:32:41,040
Speaker 3: of experience on other boards. So if they had just

633
00:32:41,160 --> 00:32:44,000
Speaker 3: executed it a little bit more professionally, then I think

634
00:32:44,040 --> 00:32:47,360
Speaker 3: people could maybe consider this kind of model as potentially working.

635
00:32:47,480 --> 00:32:50,240
Speaker 3: It's not a terrible model. I mean the idea behind

636
00:32:50,280 --> 00:32:52,880
Speaker 3: it is quite noble when you think about it. It

637
00:32:53,000 --> 00:32:53,880
Speaker 3: just couldn't work.

638
00:32:54,200 --> 00:32:56,080
Speaker 2: There's the thing, you know, talking about the paper from

639
00:32:56,080 --> 00:32:58,040
Speaker 2: Helen Toner. I mean, I think it's worth remembering that

640
00:32:58,040 --> 00:32:59,720
Speaker 2: that paper, like you say, it was written, it was

641
00:32:59,680 --> 00:33:02,600
Speaker 2: public and Helen Tona was still on the border Open AI.

642
00:33:02,720 --> 00:33:05,560
Speaker 2: So that was an indication that that dynamic did work right.

643
00:33:05,680 --> 00:33:08,360
Speaker 2: I mean, it created tension, but ultimately the governance was

644
00:33:08,360 --> 00:33:10,640
Speaker 2: still in place. It was just this sort of follow

645
00:33:10,720 --> 00:33:12,360
Speaker 2: up move that caused all this chaos.

646
00:33:13,560 --> 00:33:16,120
Speaker 1: Let's take another quick break, my friends, and then we'll

647
00:33:16,160 --> 00:33:18,400
Speaker 1: come right back and get into our last act here. Today,

648
00:33:24,000 --> 00:33:27,520
Speaker 1: we're back and we're chatting about Sam Altman, Open Ai, chat,

649
00:33:27,560 --> 00:33:31,560
Speaker 1: GPT and the AI Revolution with Parmie Olsen and Dave

650
00:33:31,640 --> 00:33:35,240
Speaker 1: Lee David. It's not clear to me that open AI's

651
00:33:35,320 --> 00:33:39,160
Speaker 1: management problems are solved and they're not operating in a

652
00:33:39,320 --> 00:33:41,520
Speaker 1: protected market as it were. They have a head start

653
00:33:42,040 --> 00:33:44,240
Speaker 1: they have a product that is early out of the

654
00:33:44,280 --> 00:33:47,600
Speaker 1: gates and much beloved, but they have to deal with

655
00:33:47,640 --> 00:33:50,840
Speaker 1: a lot of competitors who are in this market too,

656
00:33:50,920 --> 00:33:54,080
Speaker 1: And do you think that their management problems are in

657
00:33:54,240 --> 00:33:57,680
Speaker 1: rain enough that this episode won't affect their longevity.

658
00:33:58,600 --> 00:34:00,880
Speaker 2: Well, it's interesting. I was the other night and they

659
00:34:00,920 --> 00:34:03,240
Speaker 2: just happened to be a fairly senior person open aiye there,

660
00:34:03,760 --> 00:34:05,520
Speaker 2: and they said to me, you know that prior week

661
00:34:05,560 --> 00:34:07,840
Speaker 2: when all this was up in the air, Yes, it

662
00:34:07,920 --> 00:34:10,480
Speaker 2: was disruptive. They didn't know whether they were going to

663
00:34:10,560 --> 00:34:12,960
Speaker 2: be working for Microsoft in a week or any of

664
00:34:12,960 --> 00:34:15,480
Speaker 2: these other sort of different outcomes that we could have seen.

665
00:34:15,920 --> 00:34:18,280
Speaker 2: But as of sort of Monday and Tuesday, the week after,

666
00:34:18,880 --> 00:34:21,680
Speaker 2: things were basically back to normal and people were coding again,

667
00:34:21,719 --> 00:34:24,280
Speaker 2: people were planning again, and it's almost as if nothing

668
00:34:24,360 --> 00:34:26,040
Speaker 2: is that had ever happened. Now, of course, a big

669
00:34:26,080 --> 00:34:27,839
Speaker 2: thing had happened, and we've yet to see how that

670
00:34:27,880 --> 00:34:32,160
Speaker 2: fully shakes out. But I think given where they were

671
00:34:32,200 --> 00:34:34,799
Speaker 2: at on that Friday evening when this announcement was made

672
00:34:34,840 --> 00:34:37,960
Speaker 2: and the weekend of complete madness, I actually think it's

673
00:34:38,000 --> 00:34:41,000
Speaker 2: been a pretty impressive sort of pulling back into just

674
00:34:41,080 --> 00:34:44,080
Speaker 2: sort of normality, So we'll see whether that has any

675
00:34:44,120 --> 00:34:46,840
Speaker 2: long standing impact. I think you know those other companies,

676
00:34:46,960 --> 00:34:50,400
Speaker 2: Google in particular, part Me mentioned Gemini earlier one of

677
00:34:50,440 --> 00:34:53,480
Speaker 2: their AI projects that's delayed for various reasons. So a

678
00:34:53,520 --> 00:34:55,440
Speaker 2: company like Google's probably thinking, oh, wouldn't it be good

679
00:34:55,440 --> 00:34:57,759
Speaker 2: if open Eye was slowed down a little bit because

680
00:34:57,760 --> 00:34:59,319
Speaker 2: we could do with catching up here. So I think

681
00:34:59,360 --> 00:35:03,440
Speaker 2: competitors were mildly delighted to see that, But I still

682
00:35:03,440 --> 00:35:06,040
Speaker 2: think open eye is going to be considered to be

683
00:35:06,120 --> 00:35:09,400
Speaker 2: the leader here still, and that perhaps wouldn't have been

684
00:35:09,400 --> 00:35:12,560
Speaker 2: the case had they successfully asked sam Oltman Harmy.

685
00:35:12,600 --> 00:35:17,400
Speaker 1: It's been almost exactly a year since chatchept was released publicly.

686
00:35:17,880 --> 00:35:21,560
Speaker 1: What has open ai done since then to protect and

687
00:35:21,640 --> 00:35:25,640
Speaker 1: expand its franchise. I think chatchipt can accept more queries.

688
00:35:25,800 --> 00:35:29,239
Speaker 1: It's not only techt base. Its database is more up

689
00:35:29,320 --> 00:35:33,280
Speaker 1: to date. But as you've noted, also, mistakes have bound

690
00:35:33,840 --> 00:35:37,560
Speaker 1: still when you use the product. So how do you

691
00:35:37,719 --> 00:35:40,319
Speaker 1: see the evolution of the company over the last year?

692
00:35:41,520 --> 00:35:44,520
Speaker 3: Oh, over the last year? I think really the success

693
00:35:44,560 --> 00:35:47,440
Speaker 3: of Chatchipt, first of all, took them by surprise. They

694
00:35:47,480 --> 00:35:50,440
Speaker 3: were taking bets internally about how many people would use

695
00:35:50,440 --> 00:35:52,400
Speaker 3: it within the first week, and the top bet was

696
00:35:52,440 --> 00:35:54,839
Speaker 3: one hundred thousand users and it ended up being more

697
00:35:54,920 --> 00:35:57,799
Speaker 3: like a million. So I think, really the past year

698
00:35:57,840 --> 00:36:00,880
Speaker 3: has just been open ai scrambling to keep up with

699
00:36:01,040 --> 00:36:05,480
Speaker 3: viral success that it absolutely did not anticipate. Having said that,

700
00:36:05,640 --> 00:36:10,040
Speaker 3: it has from a business perspective, done incredibly well to

701
00:36:10,160 --> 00:36:13,399
Speaker 3: keep up with and satisfy the demand for people who

702
00:36:13,400 --> 00:36:16,040
Speaker 3: want to use this, whether you're a consumer who wants

703
00:36:16,040 --> 00:36:19,000
Speaker 3: to spend twenty dollars a month on chat GPT plus,

704
00:36:19,440 --> 00:36:21,600
Speaker 3: or a company who wants to get access to open

705
00:36:21,600 --> 00:36:25,640
Speaker 3: AI's software and technology, or a company that wants to

706
00:36:25,640 --> 00:36:28,560
Speaker 3: access it through Microsoft. And I think that's really where

707
00:36:28,600 --> 00:36:33,960
Speaker 3: open ai has real stability is through that partnership with Microsoft.

708
00:36:34,320 --> 00:36:38,960
Speaker 3: Because the thing about Microsoft is that it's big product

709
00:36:39,160 --> 00:36:43,560
Speaker 3: through which it dispenses open AI's technology is cloud. It's

710
00:36:43,640 --> 00:36:46,239
Speaker 3: known as the Azure platform, and it has something like

711
00:36:46,280 --> 00:36:51,480
Speaker 3: eighteen thousand customers and their big names, you know, like Rolls, Royce, Adobe,

712
00:36:51,520 --> 00:36:54,600
Speaker 3: the Seattle Seahawks, just all sorts of random big companies,

713
00:36:55,000 --> 00:36:58,799
Speaker 3: And the great thing for Microsoft is that these customers

714
00:36:58,840 --> 00:37:02,760
Speaker 3: are locked in to this product because it's actually really

715
00:37:02,800 --> 00:37:07,080
Speaker 3: expensive to extricate all your systems from what one cloud

716
00:37:07,120 --> 00:37:10,360
Speaker 3: provider like Microsoft and move on to another cloud provider

717
00:37:10,440 --> 00:37:14,600
Speaker 3: like Amazon. The customers and IT professionals hate this. They

718
00:37:14,640 --> 00:37:17,480
Speaker 3: really don't like the fact that they can't shop around.

719
00:37:17,640 --> 00:37:20,560
Speaker 3: But this is really good for Microsoft and also very

720
00:37:20,640 --> 00:37:24,200
Speaker 3: very good for open Ai. So this past year kind

721
00:37:24,200 --> 00:37:29,120
Speaker 3: of integrating themselves more and more and serving their top investor, Microsoft,

722
00:37:29,120 --> 00:37:31,200
Speaker 3: who owns forty nine percent of open Ai, has really

723
00:37:31,239 --> 00:37:33,239
Speaker 3: stood them in good stead, and it certainly did over

724
00:37:33,280 --> 00:37:36,360
Speaker 3: this past weekend because part of me wonders would Altman

725
00:37:36,400 --> 00:37:38,960
Speaker 3: have been able to be reinstated if he didn't have

726
00:37:38,960 --> 00:37:40,600
Speaker 3: such an Adella pushing so hard for.

727
00:37:40,680 --> 00:37:43,439
Speaker 1: Him well, and that brings us, Dave to the question

728
00:37:43,480 --> 00:37:47,120
Speaker 1: I wanted to ask you about Microsoft. Given that Sacha Nadella,

729
00:37:47,440 --> 00:37:52,000
Speaker 1: Microsoft's CEO, was very enmeshed in the open Ai upheaval,

730
00:37:52,480 --> 00:37:56,239
Speaker 1: he almost got Sam Altman and many of his employees

731
00:37:56,320 --> 00:37:58,000
Speaker 1: for a song. It probably would have been one of

732
00:37:58,040 --> 00:38:01,680
Speaker 1: the most epic, sort of cheap takeovers in corporate history

733
00:38:01,680 --> 00:38:06,640
Speaker 1: if that had panned out. Nadella has referred to Microsoft

734
00:38:06,640 --> 00:38:10,640
Speaker 1: as being the co pilot company now, like the company

735
00:38:10,640 --> 00:38:13,840
Speaker 1: that makes products that sort of assist you in having

736
00:38:14,160 --> 00:38:18,160
Speaker 1: a better corporate liver a better personal life. And obviously

737
00:38:18,280 --> 00:38:22,680
Speaker 1: chat GPT fits right into that. Microsoft is a player here,

738
00:38:22,719 --> 00:38:26,200
Speaker 1: and it's a heavily influential player both in this drama

739
00:38:26,280 --> 00:38:29,040
Speaker 1: and where I think open ai is going to go correct.

740
00:38:29,880 --> 00:38:31,520
Speaker 2: Yes, I mean it's not just a players, you know,

741
00:38:31,520 --> 00:38:34,360
Speaker 2: as the player externally. I mean there are other investors

742
00:38:34,400 --> 00:38:37,879
Speaker 2: in open ai, but it's only Microsoft that really has

743
00:38:37,920 --> 00:38:40,080
Speaker 2: the power here. I think, you know, Palme was touching

744
00:38:40,080 --> 00:38:42,319
Speaker 2: on a really important point there, and one that I

745
00:38:42,320 --> 00:38:45,760
Speaker 2: think's worth remembering is that the winner in AI isn't

746
00:38:45,840 --> 00:38:49,880
Speaker 2: necessarily going to be the company that has the best

747
00:38:50,200 --> 00:38:52,520
Speaker 2: AI or at least the most sophisticated AI. It's going

748
00:38:52,520 --> 00:38:55,080
Speaker 2: to be the company that can put that AI into

749
00:38:55,120 --> 00:38:59,600
Speaker 2: applications that consumers are either already using or that they

750
00:38:59,640 --> 00:39:01,520
Speaker 2: want to use in future. And so when you have

751
00:39:01,600 --> 00:39:05,200
Speaker 2: Microsoft on board, that is incredibly powerful. Because people have

752
00:39:05,280 --> 00:39:08,239
Speaker 2: been using Word for years, people have been frustrated at

753
00:39:08,320 --> 00:39:11,239
Speaker 2: PowerPoint for years, Excel, you know all these things that

754
00:39:11,239 --> 00:39:13,520
Speaker 2: people do, you know, Search changes, and people laugh about

755
00:39:13,560 --> 00:39:16,759
Speaker 2: Microsoft and Clippy, the little the little mascot that used

756
00:39:16,760 --> 00:39:18,880
Speaker 2: to have and that sense you you know where we

757
00:39:18,960 --> 00:39:24,200
Speaker 2: sort of heading again forgotten all about Clipping. I love

758
00:39:24,239 --> 00:39:28,080
Speaker 2: Clip was he was underrated. But this is where we're

759
00:39:28,120 --> 00:39:30,200
Speaker 2: going again. And I think that's why this deal is

760
00:39:30,239 --> 00:39:33,319
Speaker 2: so valuable to both open I and to Microsoft, and

761
00:39:33,360 --> 00:39:37,920
Speaker 2: it's why Satching Thedella leapt to action to step in

762
00:39:38,080 --> 00:39:41,040
Speaker 2: to stop his investment in ape and ai going up

763
00:39:41,040 --> 00:39:44,520
Speaker 2: in flames. That when Microsoft said, you know, we will

764
00:39:44,560 --> 00:39:47,680
Speaker 2: take on and we will salary match as many open

765
00:39:47,719 --> 00:39:50,839
Speaker 2: ai employees are interesting in joining us. I mean, that's

766
00:39:50,880 --> 00:39:52,680
Speaker 2: a huge thing that was still I saw one estimate

767
00:39:52,680 --> 00:39:54,400
Speaker 2: that was going to cost around a billion dollars to

768
00:39:54,400 --> 00:39:58,919
Speaker 2: do that. Now that's arguably getting a eighty billion dollar

769
00:39:58,960 --> 00:40:01,920
Speaker 2: company a huge sort of black fried a deal. But

770
00:40:02,880 --> 00:40:04,560
Speaker 2: that's a huge amount to add to your Headcunt. They

771
00:40:04,560 --> 00:40:06,080
Speaker 2: weren't even sure where they were going to put them.

772
00:40:06,080 --> 00:40:08,239
Speaker 2: They were clearing out desks at an old office that

773
00:40:08,360 --> 00:40:10,080
Speaker 2: used to be used for LinkedIn, which one of the

774
00:40:10,360 --> 00:40:13,080
Speaker 2: companies that Microsoft owns, and so this would have all

775
00:40:13,120 --> 00:40:16,520
Speaker 2: been very very very very difficult for Microsoft. What they've

776
00:40:16,520 --> 00:40:19,360
Speaker 2: come to instead, I think is their sort of preferred option.

777
00:40:19,480 --> 00:40:22,759
Speaker 2: Sam Oltman back in the job, and also on this

778
00:40:22,880 --> 00:40:27,280
Speaker 2: new board that's being made up. They have an observer seat,

779
00:40:27,360 --> 00:40:29,759
Speaker 2: so they can't vote, can't have a say in the decisions,

780
00:40:29,920 --> 00:40:32,880
Speaker 2: but can observe the board now, I asked, at the

781
00:40:32,880 --> 00:40:35,400
Speaker 2: time we're recording this, we're still learning about what exactly

782
00:40:35,480 --> 00:40:37,759
Speaker 2: that means. I asked Microsoft. They weren't sharing any more

783
00:40:37,760 --> 00:40:40,520
Speaker 2: about what exactly they'd be able to observe and what

784
00:40:40,600 --> 00:40:43,120
Speaker 2: imput they could have or indeed who at Microsoft might

785
00:40:43,120 --> 00:40:46,320
Speaker 2: take that position. But Microsoft has gone from a situation

786
00:40:46,400 --> 00:40:49,200
Speaker 2: where it had a sort of very much outsider perspective

787
00:40:49,560 --> 00:40:52,879
Speaker 2: on open Ai, even though it invested more than ten

788
00:40:52,920 --> 00:40:55,839
Speaker 2: billion dollars, to having a situation now where it's got

789
00:40:55,840 --> 00:40:58,520
Speaker 2: a much closer relationship and we'll get a heads up

790
00:40:58,560 --> 00:41:01,000
Speaker 2: on any crazy decisions like this future. One of the

791
00:41:01,000 --> 00:41:04,400
Speaker 2: things that was gobsmacking about how the Sam Altman firing

792
00:41:04,480 --> 00:41:07,000
Speaker 2: went down was that Microsoft learned just about the same

793
00:41:07,040 --> 00:41:09,279
Speaker 2: time I did when the blog post went up. There

794
00:41:09,320 --> 00:41:13,200
Speaker 2: it is, and that given the reliance that Microsoft has

795
00:41:13,280 --> 00:41:16,680
Speaker 2: on open Ai and its importance to Microsoft's strategy.

796
00:41:16,719 --> 00:41:18,520
Speaker 1: I mean, it's a major investorable thing.

797
00:41:19,560 --> 00:41:23,520
Speaker 2: Yes, absolutely, and typically your huge investors, part of that

798
00:41:23,600 --> 00:41:26,239
Speaker 2: investment would mean getting a board seat, getting two board

799
00:41:26,280 --> 00:41:28,879
Speaker 2: seats maybe, but such as the popularity of open ai

800
00:41:29,080 --> 00:41:31,319
Speaker 2: and also were stressing. Is one of the reasons why

801
00:41:31,360 --> 00:41:34,440
Speaker 2: Microsoft was something of an outside is because it worried

802
00:41:34,880 --> 00:41:38,279
Speaker 2: that being too close to OpenAI would raise some level

803
00:41:38,320 --> 00:41:41,960
Speaker 2: of scrutiny from competition regulators, sort of wondering, you know,

804
00:41:42,000 --> 00:41:44,160
Speaker 2: it is Microsoft going to flex its power to stop

805
00:41:44,200 --> 00:41:46,759
Speaker 2: others getting access to good AI. So there's a lot

806
00:41:46,760 --> 00:41:49,840
Speaker 2: of factors at play, but in sure, I think Microsoft

807
00:41:49,840 --> 00:41:53,240
Speaker 2: has found a scenario that suits it very very nicely.

808
00:41:53,280 --> 00:41:54,800
Speaker 3: Indeed, Yeah, And I would just add that if it

809
00:41:54,880 --> 00:41:57,440
Speaker 3: had been a subsidiary of Microsoft, if Sam Altman had

810
00:41:57,440 --> 00:41:59,759
Speaker 3: been part of this advanced AI group, that's not what

811
00:42:00,280 --> 00:42:04,880
Speaker 3: Inodella wanted at all, because that would mean Microsoft takes

812
00:42:04,920 --> 00:42:07,920
Speaker 3: on all of the reputational risk, all of the legal

813
00:42:08,040 --> 00:42:11,880
Speaker 3: risk for all the crazy stuff that open ai regularly does,

814
00:42:12,320 --> 00:42:15,239
Speaker 3: like putting chatchypt out into the world. Microsoft would never

815
00:42:15,360 --> 00:42:18,960
Speaker 3: do that because copywriter issues for a start, and crazy

816
00:42:18,960 --> 00:42:21,319
Speaker 3: things that this bot could say. That's why big tech

817
00:42:21,360 --> 00:42:24,360
Speaker 3: companies don't do the kinds of things that open ai does.

818
00:42:24,760 --> 00:42:27,440
Speaker 3: So it's much better to be an investor take on

819
00:42:27,640 --> 00:42:31,120
Speaker 3: all the glow of open ai and none of the liability.

820
00:42:30,880 --> 00:42:32,440
Speaker 2: Which saw a sign off that didn't we part me

821
00:42:32,480 --> 00:42:34,319
Speaker 2: a few years ago, which I'm sure you remember, when

822
00:42:34,600 --> 00:42:37,280
Speaker 2: Microsoft had an AI bot that it kind of released

823
00:42:37,320 --> 00:42:40,560
Speaker 2: onto Twitter. I believe it was called Tay the Box. Yeah,

824
00:42:40,600 --> 00:42:44,040
Speaker 2: and almost immediately people realized that if they said certain

825
00:42:44,080 --> 00:42:45,719
Speaker 2: things to it, it would repeat them back to them,

826
00:42:45,719 --> 00:42:48,400
Speaker 2: and it became sweary, it became racist, and it was

827
00:42:48,440 --> 00:42:51,239
Speaker 2: just this is just a tiny, tiny experiment, but it

828
00:42:51,360 --> 00:42:54,560
Speaker 2: caused Microsoft no end of reputational issues and things to

829
00:42:54,560 --> 00:42:56,880
Speaker 2: be accountable for. So you can totally see why they

830
00:42:57,000 --> 00:43:00,000
Speaker 2: want to be outside to some degree.

831
00:43:00,080 --> 00:43:01,920
Speaker 3: Saw people still cringe at that memory.

832
00:43:02,480 --> 00:43:06,719
Speaker 1: And I think Microsoft still licking its federal antitrust investigation

833
00:43:06,840 --> 00:43:10,200
Speaker 1: wounds of the nineteen nineties. That makes it a little

834
00:43:10,440 --> 00:43:12,960
Speaker 1: as as I put it earlier, absolutely has it meant

835
00:43:13,000 --> 00:43:15,240
Speaker 1: to appear to be the titan in this nascent.

836
00:43:15,760 --> 00:43:17,800
Speaker 3: And then these who were dealing with that in the nineties,

837
00:43:17,800 --> 00:43:20,920
Speaker 3: like Brad Smith the Legal Council are still there, like

838
00:43:20,960 --> 00:43:25,440
Speaker 3: they all remember this stuff. So yeah, and satchin Adella two, I.

839
00:43:25,400 --> 00:43:28,440
Speaker 1: Want to ask you each about lessons that you've learned

840
00:43:28,440 --> 00:43:31,400
Speaker 1: from this dramatic little moment in Silicon Valley. I like

841
00:43:31,440 --> 00:43:32,920
Speaker 1: to ask people that at the end of the show.

842
00:43:33,320 --> 00:43:35,759
Speaker 1: So let's start with you, Parmie. What is something you've

843
00:43:35,840 --> 00:43:39,800
Speaker 1: learned watching this recent debacle that you didn't know before

844
00:43:39,840 --> 00:43:40,400
Speaker 1: it happened.

845
00:43:41,440 --> 00:43:45,000
Speaker 3: Hmmm, it's a good question. I guess I would just

846
00:43:45,040 --> 00:43:49,000
Speaker 3: go back to my answer previously about the fact that

847
00:43:49,320 --> 00:43:53,600
Speaker 3: a board like this could actually topple someone like Sam Altman.

848
00:43:53,880 --> 00:43:56,160
Speaker 3: But I suppose the other thing is that this isn't

849
00:43:56,160 --> 00:43:59,360
Speaker 3: so much something that I've learned, but just a suspicion.

850
00:43:59,400 --> 00:44:03,520
Speaker 3: The cynic me has been confirmed that Silicon Valley will

851
00:44:03,520 --> 00:44:06,480
Speaker 3: Silicon Valley, And even if you set up a board

852
00:44:07,040 --> 00:44:10,840
Speaker 3: and a governance structure that looks good to everyone, that

853
00:44:10,960 --> 00:44:15,000
Speaker 3: looks like it's there to help humanity, it will still

854
00:44:15,080 --> 00:44:18,360
Speaker 3: ultimately serve the purpose of its investors. And that is

855
00:44:18,400 --> 00:44:22,400
Speaker 3: precisely what happened after this dramatic weekend where Sam was

856
00:44:22,440 --> 00:44:25,960
Speaker 3: ousted and brought right back in. So a little bit

857
00:44:25,960 --> 00:44:29,840
Speaker 3: of hypocrisy there, and also just Silicon Valley being Silicon Valley.

858
00:44:30,680 --> 00:44:33,879
Speaker 1: Dave, I don't know, can you still learn things? Both

859
00:44:33,880 --> 00:44:36,000
Speaker 1: of you know so much that I sometimes wonder.

860
00:44:35,960 --> 00:44:37,920
Speaker 2: Learning every day I've learned Listeners to parme me on

861
00:44:37,920 --> 00:44:39,799
Speaker 2: this podcast and I'm glad you let her go first.

862
00:44:39,880 --> 00:44:41,680
Speaker 2: I can have a moment to think about this, because

863
00:44:41,760 --> 00:44:44,520
Speaker 2: it's interesting because on some levels it feels utterly predictable

864
00:44:44,680 --> 00:44:47,200
Speaker 2: this happened. And I think one thing I've learned or

865
00:44:47,320 --> 00:44:51,200
Speaker 2: was surprised by, was I think we always knew that

866
00:44:51,800 --> 00:44:55,680
Speaker 2: at some point there would be some friction around the

867
00:44:55,800 --> 00:44:59,160
Speaker 2: direction of one of these companies. I'm surprised it's come

868
00:44:59,239 --> 00:45:01,600
Speaker 2: so soon for open ai I thought we'd be more

869
00:45:01,640 --> 00:45:05,840
Speaker 2: sort of tangible threats or fall out or crisis that

870
00:45:05,880 --> 00:45:07,520
Speaker 2: we'd all sort of know about that would lead to

871
00:45:07,520 --> 00:45:09,640
Speaker 2: a moment like this. I'm surprised it's come sort of

872
00:45:09,920 --> 00:45:11,960
Speaker 2: from what seems to be more just sort of human

873
00:45:12,040 --> 00:45:15,239
Speaker 2: personality clashes than anything else. So that surprised me. The

874
00:45:15,280 --> 00:45:18,560
Speaker 2: frigidity is surprising. But I think that the lesson, if

875
00:45:18,600 --> 00:45:21,759
Speaker 2: we can take one so far, and I think it's yeah,

876
00:45:21,760 --> 00:45:24,120
Speaker 2: we're still learning the lessons. I think the lesson is

877
00:45:24,120 --> 00:45:26,279
Speaker 2: that open aiy is a tech company. It's a tech

878
00:45:26,280 --> 00:45:28,920
Speaker 2: company like Facebook, to tech company like Google's the tech company.

879
00:45:28,920 --> 00:45:31,799
Speaker 2: And I think I'm going to constantly remind myself of

880
00:45:31,840 --> 00:45:36,359
Speaker 2: this sort of this week and remember that ultimately it

881
00:45:36,520 --> 00:45:39,800
Speaker 2: was the needs of the tech industry and the money

882
00:45:39,800 --> 00:45:42,960
Speaker 2: of the tech industry that essentially had the final say.

883
00:45:43,480 --> 00:45:45,640
Speaker 2: And so I think open AI. Although we can talk

884
00:45:45,680 --> 00:45:49,640
Speaker 2: about governance and nonprofit status, I think ultimately it's still

885
00:45:49,680 --> 00:45:51,959
Speaker 2: a tech company and we should cover it as such

886
00:45:52,000 --> 00:45:53,719
Speaker 2: and think about it as such. So that's my lesson

887
00:45:53,760 --> 00:45:54,000
Speaker 2: from this.

888
00:45:54,080 --> 00:45:57,879
Speaker 1: I think we are out of time. Prime and Dave,

889
00:45:57,960 --> 00:45:59,600
Speaker 1: thank you so much for coming on today.

890
00:46:00,040 --> 00:46:00,480
Speaker 3: Thank you.

891
00:46:00,880 --> 00:46:01,120
Speaker 2: Thanks.

892
00:46:01,200 --> 00:46:03,880
Speaker 1: Tim parme Olsen and Dave Lee are both columnists with

893
00:46:03,880 --> 00:46:06,360
Speaker 1: Bloomberg Opinion. You can find their work on the Bloomberg

894
00:46:06,400 --> 00:46:11,320
Speaker 1: Opinion website and on the Bloomberg terminal. Here at crash Course,

895
00:46:11,440 --> 00:46:15,440
Speaker 1: we believe that collisions can be messy, impressive, challenging, surprising,

896
00:46:15,760 --> 00:46:20,040
Speaker 1: and always instructive. In today's crash Course, I learned that

897
00:46:20,160 --> 00:46:22,799
Speaker 1: no matter how talented or how smart you are, no

898
00:46:22,840 --> 00:46:25,400
Speaker 1: matter what industry you're working in, even if it's in

899
00:46:25,440 --> 00:46:29,400
Speaker 1: Silicon Valley, humans can be humans and they can do crazy,

900
00:46:29,840 --> 00:46:33,279
Speaker 1: crazy things. What did you learn? We'd love to hear

901
00:46:33,320 --> 00:46:36,000
Speaker 1: from you. You can tweet at the Bloomberg Opinion handle

902
00:46:36,160 --> 00:46:40,360
Speaker 1: at Opinion or me at Tim O'Brien using the hashtag

903
00:46:40,440 --> 00:46:43,880
Speaker 1: Bloomberg Crash Course. You can also subscribe to our show

904
00:46:44,000 --> 00:46:47,120
Speaker 1: wherever you're listening right now, and please leave us a review.

905
00:46:47,560 --> 00:46:50,880
Speaker 1: It helps more people find the show. This episode was

906
00:46:50,920 --> 00:46:54,960
Speaker 1: produced by the indispensable and decidedly non crazy Adam Azarakus

907
00:46:55,280 --> 00:46:58,960
Speaker 1: and me. Our supervising producer is Magnus Hendrickson, and we

908
00:46:59,000 --> 00:47:02,239
Speaker 1: had editing help from the Sage Bauman, Jeff Grocock, Mike

909
00:47:02,360 --> 00:47:06,680
Speaker 1: Nitze and Christine Vanden Bilart. Blake Maples does our sound

910
00:47:06,719 --> 00:47:10,880
Speaker 1: engineering and our original theme song was composed by Luis Gara.

911
00:47:11,400 --> 00:47:14,320
Speaker 1: I'm Tim O'Brien. We'll be back next week with another

912
00:47:14,360 --> 00:47:15,000
Speaker 1: crash course