1
00:00:15,356 --> 00:00:22,516
Speaker 1: Pushkin. Digging up metal from out of the ground is

2
00:00:22,556 --> 00:00:26,116
Speaker 1: a business that is literally thousands of years old, but

3
00:00:26,236 --> 00:00:30,756
Speaker 1: mining suddenly has new importance. The energy transition, going from

4
00:00:30,796 --> 00:00:33,756
Speaker 1: fossil fuels to renewable energy is going to take a

5
00:00:33,836 --> 00:00:38,196
Speaker 1: ridiculous amount of metal, metal like copper and lithium. The

6
00:00:38,316 --> 00:00:40,996
Speaker 1: need is so great and so urgent that we're gonna

7
00:00:40,996 --> 00:00:43,596
Speaker 1: have to come up with new ways to find metal

8
00:00:43,716 --> 00:00:47,116
Speaker 1: buried in the earth. And as it happens, a new

9
00:00:47,236 --> 00:00:50,156
Speaker 1: kind of mining company, a mining company you might call

10
00:00:50,196 --> 00:00:54,556
Speaker 1: an AI driven startup, just made the biggest copper discovery

11
00:00:54,636 --> 00:01:02,196
Speaker 1: in over a decade. It's worth tens of billions of dollars.

12
00:01:04,356 --> 00:01:06,756
Speaker 1: I'm Jacob Goldstein, and this is What's Your Problem, the

13
00:01:06,796 --> 00:01:08,836
Speaker 1: show where I talk to people who are trying to

14
00:01:08,876 --> 00:01:13,236
Speaker 1: make technological progress. My guest today is Kurt House, the

15
00:01:13,316 --> 00:01:17,516
Speaker 1: founder and CEO of Cobald Metals. Kurt's problem is this,

16
00:01:18,276 --> 00:01:22,356
Speaker 1: how do you use AI machine learning data science to

17
00:01:22,436 --> 00:01:26,276
Speaker 1: find the metals we need for the energy transition. As

18
00:01:26,316 --> 00:01:29,076
Speaker 1: you'll hear, my conversation with Kurt goes beyond mining and

19
00:01:29,156 --> 00:01:33,196
Speaker 1: AI to cover Kurt's really compelling way of just thinking

20
00:01:33,196 --> 00:01:37,836
Speaker 1: about making decisions in an uncertain world. We started though,

21
00:01:37,916 --> 00:01:40,636
Speaker 1: by talking about how he came up with the idea.

22
00:01:40,356 --> 00:01:42,516
Speaker 2: For his company.

23
00:01:44,636 --> 00:01:48,916
Speaker 3: So if you go back about eight years ago, my

24
00:01:48,956 --> 00:01:52,756
Speaker 3: co founders and I were looking at the trends in

25
00:01:52,796 --> 00:01:57,116
Speaker 3: the energy transition, seeing the electric vehicle and renewable energy

26
00:01:57,156 --> 00:02:03,916
Speaker 3: sort of revolutions coming, and it's quite easy to convince

27
00:02:03,956 --> 00:02:08,596
Speaker 3: yourself that the material requirements for the energy transition will

28
00:02:08,636 --> 00:02:12,836
Speaker 3: be tremendous. The amount of very specific materials that the

29
00:02:12,876 --> 00:02:18,196
Speaker 3: world needs copper, lithium, cobalt, nickel, graphite, others.

30
00:02:18,636 --> 00:02:22,716
Speaker 1: Uh, this is basically stuff to build, like batteries and wires.

31
00:02:23,116 --> 00:02:23,556
Speaker 2: Tony right.

32
00:02:23,556 --> 00:02:25,996
Speaker 1: This is motors electrification.

33
00:02:26,436 --> 00:02:29,436
Speaker 3: Just an electric motor is a bundle of copper wire

34
00:02:29,556 --> 00:02:34,876
Speaker 3: surrounded by surrounding a permanent magnet. Every battery require, every

35
00:02:34,996 --> 00:02:38,876
Speaker 3: mobile battery requires lithium and nickel and cobalt.

36
00:02:39,316 --> 00:02:40,476
Speaker 2: These are all that These are.

37
00:02:40,516 --> 00:02:43,156
Speaker 3: These are the key materials for which which in some cases,

38
00:02:43,156 --> 00:02:45,116
Speaker 3: like the humanity has been using lots of copper for

39
00:02:45,156 --> 00:02:47,756
Speaker 3: a long time, there's a big copper market, but it

40
00:02:48,316 --> 00:02:51,996
Speaker 3: needs to at least double from a very large base lithium.

41
00:02:52,076 --> 00:02:55,276
Speaker 3: Humanity has not been using much lithium for very long

42
00:02:55,956 --> 00:02:59,596
Speaker 3: and now the lithium market needs to grow by well

43
00:02:59,596 --> 00:03:03,116
Speaker 3: more than a factor of ten uh to to UH

44
00:03:03,236 --> 00:03:09,316
Speaker 3: fully electrify just just the transportation sector. So the the

45
00:03:09,356 --> 00:03:12,996
Speaker 3: sort of macro needs were very obvious. So that's observation one.

46
00:03:13,196 --> 00:03:17,516
Speaker 3: Observation two say, okay, well, maybe the incumbents are really

47
00:03:17,556 --> 00:03:20,636
Speaker 3: good at finding new materials, and as prices rise a

48
00:03:20,676 --> 00:03:22,836
Speaker 3: little bit, they'll find new materials and the market will

49
00:03:22,876 --> 00:03:25,756
Speaker 3: just be well supplied. And that turns out to be

50
00:03:25,796 --> 00:03:28,396
Speaker 3: definitely wrong. And it's actually really easy to verify that

51
00:03:28,436 --> 00:03:33,396
Speaker 3: it's wrong because the large, well resourced mining companies basically

52
00:03:33,436 --> 00:03:37,716
Speaker 3: don't even do exploration. Actually, the big mining companies out

53
00:03:37,756 --> 00:03:40,796
Speaker 3: of they spend sixty sixty five billion dollars a year

54
00:03:40,836 --> 00:03:42,756
Speaker 3: on dividends and share buybacks and less than half a

55
00:03:42,796 --> 00:03:44,356
Speaker 3: billion on exploration activities.

56
00:03:44,556 --> 00:03:47,236
Speaker 2: But that half a billion, that's the.

57
00:03:47,276 --> 00:03:52,116
Speaker 3: Deployment of conventional exploration technologies, right, things that would be

58
00:03:52,276 --> 00:03:57,996
Speaker 3: natural to most geologists from the nineteen sixties or seventies, right, almost,

59
00:03:58,196 --> 00:04:00,356
Speaker 3: so you can round to zero how much money they're

60
00:04:00,396 --> 00:04:03,636
Speaker 3: spending on research and development for new techniques and new

61
00:04:03,636 --> 00:04:05,516
Speaker 3: technologies to improve.

62
00:04:05,196 --> 00:04:06,316
Speaker 2: The exploration process.

63
00:04:06,796 --> 00:04:09,796
Speaker 3: So it was basically those sets of observations, those two

64
00:04:09,796 --> 00:04:10,876
Speaker 3: sets of observations.

65
00:04:11,236 --> 00:04:13,476
Speaker 1: We need metals, and nobody's really looking.

66
00:04:13,516 --> 00:04:15,276
Speaker 3: They're not looking for them, but they're certainly not getting

67
00:04:15,276 --> 00:04:17,636
Speaker 3: better at it, they're getting worse at We call this,

68
00:04:17,956 --> 00:04:21,076
Speaker 3: we call that that that that trend and the increasing

69
00:04:21,156 --> 00:04:25,356
Speaker 3: cost of discovery, we call that e Rooms law of mining.

70
00:04:26,476 --> 00:04:27,196
Speaker 1: Moore's law back.

71
00:04:27,516 --> 00:04:28,516
Speaker 2: Good, Yeah, I'm impressed.

72
00:04:28,596 --> 00:04:33,996
Speaker 1: Yeah, they talk about that in biotech as meaning, I'm

73
00:04:34,156 --> 00:04:38,676
Speaker 1: where whereas microchips get cheaper and better every year, mining

74
00:04:38,756 --> 00:04:41,396
Speaker 1: gets worse and slower and more more expensive.

75
00:04:41,396 --> 00:04:46,076
Speaker 3: And specifically exploration, exploit discovery specifically. Okay, yeah, so those

76
00:04:46,076 --> 00:04:48,316
Speaker 3: were the those were the major needs. So then you say, okay,

77
00:04:48,316 --> 00:04:51,156
Speaker 3: what what can we do, how can we do? What

78
00:04:51,156 --> 00:04:54,836
Speaker 3: can we do differently? How can we help? And And

79
00:04:54,876 --> 00:05:00,996
Speaker 3: the answer is that exploration is fundamentally an information problem,

80
00:05:01,396 --> 00:05:04,196
Speaker 3: fundamentally right, if you we know, for deep physical reasons

81
00:05:04,236 --> 00:05:05,876
Speaker 3: which I can explain in a minute, we know there's

82
00:05:07,236 --> 00:05:12,596
Speaker 3: there's gobs and gob of undiscovered uh Rich deposits out there.

83
00:05:13,036 --> 00:05:15,716
Speaker 3: We don't know where they are. So so the gap

84
00:05:15,756 --> 00:05:18,596
Speaker 3: is the knowledge of where they are. Right If if

85
00:05:19,076 --> 00:05:21,876
Speaker 3: God gave you a perfect model of the Earth's crust,

86
00:05:21,956 --> 00:05:24,396
Speaker 3: right the location and form of every atom.

87
00:05:24,596 --> 00:05:25,636
Speaker 2: You'd be a perfect explorer.

88
00:05:25,676 --> 00:05:28,676
Speaker 3: You'd know where all the where all the high grade

89
00:05:29,036 --> 00:05:30,836
Speaker 3: high concentration anomalies were.

90
00:05:30,956 --> 00:05:32,236
Speaker 2: You'd also be a perfect miner.

91
00:05:32,356 --> 00:05:36,236
Speaker 1: The miner's religious vision, right, is the gift from God

92
00:05:36,276 --> 00:05:37,316
Speaker 1: of perfect information.

93
00:05:37,476 --> 00:05:40,636
Speaker 3: Yes, perfect exactly, But it's not that it's a so

94
00:05:40,676 --> 00:05:41,796
Speaker 3: we don't have that. So we have we have a

95
00:05:41,876 --> 00:05:46,676
Speaker 3: huge amount of uncertainty. But the sort of managing the

96
00:05:46,756 --> 00:05:50,236
Speaker 3: data that you have and then making probabilistic inferences on

97
00:05:50,396 --> 00:05:55,476
Speaker 3: that data, uh is fundamentally an information problem. We look

98
00:05:55,516 --> 00:05:57,996
Speaker 3: at it as this is this is kind of a

99
00:05:58,436 --> 00:06:03,516
Speaker 3: perfect tailored application for data science and modern scientific computing.

100
00:06:04,036 --> 00:06:06,636
Speaker 3: It has it's it's it's a little different, it has

101
00:06:06,676 --> 00:06:09,396
Speaker 3: some some sort of unique, really cool attributes to it.

102
00:06:09,436 --> 00:06:12,516
Speaker 3: But it is fundamentally an information problem and fundamentally a

103
00:06:12,556 --> 00:06:16,556
Speaker 3: search problem. And so the thing that could be massively

104
00:06:16,636 --> 00:06:19,916
Speaker 3: different would be a company built from the ground up,

105
00:06:20,876 --> 00:06:24,156
Speaker 3: a sort of a Silicon Valley company built from the

106
00:06:24,156 --> 00:06:29,636
Speaker 3: ground up that combines the best existing knowledge of geoscientists

107
00:06:29,356 --> 00:06:34,556
Speaker 3: with world class data scientists and software engineers coming out

108
00:06:34,596 --> 00:06:39,276
Speaker 3: of the major tech monopolies Google, Apple, Facebook, you name it,

109
00:06:40,316 --> 00:06:43,116
Speaker 3: who have never worked in the metals and mining business before, right.

110
00:06:43,236 --> 00:06:47,756
Speaker 1: So it's fundamentally sort of bringing the tools of data science,

111
00:06:48,316 --> 00:06:52,036
Speaker 1: machine learning AI to bear on geoscience.

112
00:06:52,076 --> 00:06:54,236
Speaker 2: Absolutely, if I'm going to reduce totally totally.

113
00:06:54,316 --> 00:06:57,716
Speaker 1: Yeah, Yeah, it's amazing that nobody got to it before

114
00:06:57,756 --> 00:07:01,076
Speaker 1: you did. There are these giant billion dollar mining companies

115
00:07:01,076 --> 00:07:02,796
Speaker 1: and it was right there for them, but they didn't

116
00:07:02,796 --> 00:07:05,876
Speaker 1: do it. I mean, why didn't somebody do it before you?

117
00:07:06,476 --> 00:07:08,716
Speaker 3: What you will definitely hear is, oh, we use data

118
00:07:08,716 --> 00:07:11,796
Speaker 3: science like we we we use scientist, right, And it's

119
00:07:11,796 --> 00:07:14,396
Speaker 3: like and it's like not totally wrong. But what is

120
00:07:14,436 --> 00:07:19,676
Speaker 3: what is definitely unambiguously different, if not unique to Cobold,

121
00:07:19,836 --> 00:07:22,196
Speaker 3: is that we're we're a full stack explorer.

122
00:07:22,196 --> 00:07:25,196
Speaker 2: We were started, we were started and built.

123
00:07:26,476 --> 00:07:32,436
Speaker 3: On the concept that that applying uh vanguard scientific competing

124
00:07:32,476 --> 00:07:36,836
Speaker 3: techniques to these problems would improve the efficacy and efficiency

125
00:07:36,836 --> 00:07:38,836
Speaker 3: of exploration, right, that that is the goal. So we

126
00:07:38,916 --> 00:07:43,556
Speaker 3: have our technical staff is about sixty percent data scientists

127
00:07:43,596 --> 00:07:46,556
Speaker 3: or software engineers, about forty percent geo scientists, right, so

128
00:07:46,596 --> 00:07:49,476
Speaker 3: we're roughly equal equal numbers across the three disciplines.

129
00:07:50,156 --> 00:07:53,596
Speaker 2: And that's that's completely Uh, that's that's unique.

130
00:07:53,996 --> 00:07:57,876
Speaker 1: Let's let's talk about data, right. I feel like, uh,

131
00:07:58,316 --> 00:08:01,716
Speaker 1: discussions about AI, for me tend to get more interesting

132
00:08:01,796 --> 00:08:04,436
Speaker 1: when we get into data and and it seems like

133
00:08:04,476 --> 00:08:06,316
Speaker 1: that's where a lot of the action is and and

134
00:08:06,716 --> 00:08:08,796
Speaker 1: from what I understand about the story of your company,

135
00:08:08,876 --> 00:08:12,156
Speaker 1: kind of building the data set is a big part

136
00:08:12,156 --> 00:08:13,716
Speaker 1: of the story and a big part of what has

137
00:08:13,756 --> 00:08:16,996
Speaker 1: differentiated you. So you have all these data scientists, what

138
00:08:17,036 --> 00:08:19,916
Speaker 1: they need is data. How do you go about building

139
00:08:19,916 --> 00:08:22,076
Speaker 1: this data set to find these metals?

140
00:08:22,116 --> 00:08:23,436
Speaker 2: Yeah, it's incredibly good question.

141
00:08:23,916 --> 00:08:28,236
Speaker 3: So most of the data we use was collected by

142
00:08:28,236 --> 00:08:33,236
Speaker 3: other people at other times. Humans have been collecting information

143
00:08:33,396 --> 00:08:35,716
Speaker 3: about the physics and the chemistry of the Ear's crust

144
00:08:35,756 --> 00:08:38,836
Speaker 3: for a very very long time. Right, They've been doing

145
00:08:38,836 --> 00:08:41,116
Speaker 3: it for well, I mean, in some sense for millennia,

146
00:08:41,196 --> 00:08:44,076
Speaker 3: but certainly over the last century they've been doing it

147
00:08:44,196 --> 00:08:49,716
Speaker 3: in ever more sophisticated ways, and for reasons I can explain.

148
00:08:50,076 --> 00:08:52,156
Speaker 3: Most almost all of that data is actually in the

149
00:08:52,196 --> 00:08:58,316
Speaker 3: public domain. The problem is it is a utter mess

150
00:08:58,796 --> 00:09:02,716
Speaker 3: It is like an end member hard messy data problem.

151
00:09:03,076 --> 00:09:08,316
Speaker 3: Think of different humans in different decades, speaking different languages

152
00:09:08,356 --> 00:09:09,956
Speaker 3: and different places of the world.

153
00:09:10,876 --> 00:09:12,316
Speaker 2: Collecting different types of.

154
00:09:12,356 --> 00:09:13,836
Speaker 3: Data, and I'll get into the types of data in

155
00:09:13,876 --> 00:09:18,076
Speaker 3: a moment, with different measurement techniques based on the vintage

156
00:09:18,076 --> 00:09:21,196
Speaker 3: of their of the era, and then storing it in

157
00:09:21,316 --> 00:09:25,476
Speaker 3: all manner of storage media, everything from literally hand written

158
00:09:26,476 --> 00:09:29,596
Speaker 3: geologic notes or handwritten drilling notes all the way to

159
00:09:30,036 --> 00:09:34,436
Speaker 3: cloud native data data structures right and everything in between.

160
00:09:35,196 --> 00:09:37,436
Speaker 2: And so it is this incredible mess of data.

161
00:09:37,556 --> 00:09:41,076
Speaker 1: Give me some specific examples. What are specific like did

162
00:09:41,076 --> 00:09:43,796
Speaker 1: you find stuff in a drawer or something sort like,

163
00:09:43,876 --> 00:09:45,396
Speaker 1: give me some specific examples.

164
00:09:45,556 --> 00:09:50,396
Speaker 3: So I'll give you examples of there's geologic libraries archives

165
00:09:50,756 --> 00:09:59,036
Speaker 3: right with carefully carefully constructed geologic maps that were that

166
00:09:59,156 --> 00:10:02,156
Speaker 3: might be one hundred years old, and there they were

167
00:10:02,156 --> 00:10:07,876
Speaker 3: a smart skilled geologist just make it a doing field mapping,

168
00:10:07,916 --> 00:10:12,116
Speaker 3: which basically means deserving and recording the observations of outcrops

169
00:10:12,156 --> 00:10:15,196
Speaker 3: and describing the rock, the rock types and those outcrops

170
00:10:15,516 --> 00:10:19,276
Speaker 3: right and locating them in space. And the Earth's crust

171
00:10:19,356 --> 00:10:22,916
Speaker 3: changes very slowly. So provided that was provided that was

172
00:10:23,076 --> 00:10:27,116
Speaker 3: like a well done one hundred years ago, it's still valid.

173
00:10:27,356 --> 00:10:30,116
Speaker 3: It's just that it's it's it's you know, literally in

174
00:10:30,236 --> 00:10:33,076
Speaker 3: drawers piled on top of each other in you know,

175
00:10:33,236 --> 00:10:36,276
Speaker 3: and uh and.

176
00:10:35,516 --> 00:10:36,836
Speaker 2: And basically not used.

177
00:10:36,876 --> 00:10:39,116
Speaker 3: It would only be used by a very industrious human

178
00:10:39,156 --> 00:10:43,076
Speaker 3: being who spent who spent countless hours sort.

179
00:10:42,916 --> 00:10:44,516
Speaker 2: Of searching through the old archives.

180
00:10:44,596 --> 00:10:47,756
Speaker 3: Right, so we go, so we go to various archives,

181
00:10:47,956 --> 00:10:51,396
Speaker 3: and we make an arrangement to digitize the information at

182
00:10:51,396 --> 00:10:53,676
Speaker 3: our expense, and we give the.

183
00:10:53,476 --> 00:10:56,516
Speaker 2: Owners all full digital copy.

184
00:10:56,956 --> 00:10:59,996
Speaker 3: It's almost always public domain data, and so we we

185
00:11:00,036 --> 00:11:03,076
Speaker 3: have a right to use it, or we negotiate a

186
00:11:03,116 --> 00:11:08,396
Speaker 3: specific use right. So digitizing a geologic map is like

187
00:11:08,676 --> 00:11:10,076
Speaker 3: is the very very beginning.

188
00:11:10,436 --> 00:11:13,556
Speaker 2: Then then you need to extract the information from from the.

189
00:11:13,516 --> 00:11:16,396
Speaker 3: Digital copy of the map, uh and, and you have

190
00:11:16,596 --> 00:11:18,716
Speaker 3: many different types of information there.

191
00:11:18,956 --> 00:11:20,636
Speaker 2: You can have in the paper records.

192
00:11:20,676 --> 00:11:23,876
Speaker 3: You might have you might have chemical assays, so measurements

193
00:11:23,916 --> 00:11:28,876
Speaker 3: of the concentrations of the elemental concentrations of samples taken

194
00:11:28,916 --> 00:11:32,116
Speaker 3: from different locations on the map, and that could be a.

195
00:11:32,076 --> 00:11:33,436
Speaker 2: Part of a part of the record.

196
00:11:33,796 --> 00:11:36,036
Speaker 3: And so that's tabular data because it'll say, well, this

197
00:11:36,116 --> 00:11:39,996
Speaker 3: sample sample whatever had had x percent calcium and y

198
00:11:40,036 --> 00:11:43,356
Speaker 3: percent magnesium and z percent silica, and et cetera, et cetera,

199
00:11:43,396 --> 00:11:46,396
Speaker 3: et cetera. That's all valuable information. So that's tabular information

200
00:11:46,436 --> 00:11:50,876
Speaker 3: that then gets extracted by by our systems and populated

201
00:11:50,916 --> 00:11:55,356
Speaker 3: into into what we call our universal schema, which just

202
00:11:55,436 --> 00:11:58,476
Speaker 3: means that every data type is stored in a in

203
00:11:58,516 --> 00:11:59,596
Speaker 3: a consistent format.

204
00:11:59,916 --> 00:12:03,996
Speaker 1: You're standardizing this wildly messy heterogeneous test.

205
00:12:05,716 --> 00:12:07,396
Speaker 3: That's exactly right, And we should talk about more about

206
00:12:07,396 --> 00:12:09,676
Speaker 3: what the data is because it's really fascinating, right, So

207
00:12:09,876 --> 00:12:11,596
Speaker 3: I gave you, I give you two examples. I give

208
00:12:11,636 --> 00:12:14,956
Speaker 3: you the sort of qualitative almost like drawn geologic map,

209
00:12:14,996 --> 00:12:19,236
Speaker 3: which is incredibly useful information, but qualitative and continuous in nature.

210
00:12:19,556 --> 00:12:21,636
Speaker 3: Then there's the sort of tabular data that would be

211
00:12:21,636 --> 00:12:25,276
Speaker 3: any kind of any kind of assay data, measurements of composition.

212
00:12:25,556 --> 00:12:27,556
Speaker 3: But then you have a whole different classes of data,

213
00:12:27,716 --> 00:12:30,076
Speaker 3: like geophysical data, which tells you something.

214
00:12:29,756 --> 00:12:31,956
Speaker 2: About the physics of the Earth's crust.

215
00:12:32,036 --> 00:12:35,796
Speaker 3: So, for example, you probably know that the Earth's gravitational

216
00:12:35,836 --> 00:12:38,596
Speaker 3: field changes from place to place as you move around.

217
00:12:38,916 --> 00:12:42,076
Speaker 3: It changes because you can go up or down in elevation. Well,

218
00:12:42,076 --> 00:12:44,156
Speaker 3: that's easy to adjust for because you know the elevation.

219
00:12:44,276 --> 00:12:46,916
Speaker 3: It also changes because the density of the rocks below

220
00:12:46,996 --> 00:12:50,836
Speaker 3: you change. And so if you're standing over over a

221
00:12:51,436 --> 00:12:56,116
Speaker 3: or body that has twice the density of whose rocks

222
00:12:56,116 --> 00:12:58,516
Speaker 3: are twice as dense as the surrounding rocks that'll pull

223
00:12:58,556 --> 00:12:59,796
Speaker 3: on you slightly more.

224
00:13:00,196 --> 00:13:02,716
Speaker 2: Okay, and you can measure that that I did not know.

225
00:13:02,796 --> 00:13:05,436
Speaker 3: And this is let's go down this rabbit hole actually

226
00:13:05,436 --> 00:13:09,236
Speaker 3: because it's super interesting. Okay, because imagine, so imagine you

227
00:13:09,276 --> 00:13:12,276
Speaker 3: make this measurement. What are you actually measuring. You're measuring

228
00:13:12,276 --> 00:13:14,956
Speaker 3: the force of gravity in a particular location, and you

229
00:13:14,996 --> 00:13:17,276
Speaker 3: can measure Okay, I've adjusted for elevation and the force

230
00:13:17,276 --> 00:13:19,956
Speaker 3: of gravity is a little bit higher here. Okay, that's

231
00:13:19,996 --> 00:13:22,556
Speaker 3: all you actually know at this moment. So what is

232
00:13:22,556 --> 00:13:24,636
Speaker 3: that telling you? Is it telling you you have a

233
00:13:24,796 --> 00:13:28,436
Speaker 3: modestly more dense object like just below the surface, or

234
00:13:28,476 --> 00:13:30,756
Speaker 3: is it telling you you have a massively more dense

235
00:13:30,796 --> 00:13:33,556
Speaker 3: object deeper. Well, it turns out that this is a

236
00:13:33,636 --> 00:13:37,836
Speaker 3: fundamentally degenerate problem or non unique problem would be the

237
00:13:37,876 --> 00:13:38,756
Speaker 3: way to describe it.

238
00:13:38,796 --> 00:13:39,476
Speaker 2: Mathematically.

239
00:13:39,676 --> 00:13:41,836
Speaker 3: There are many ways to solve that problem. There are

240
00:13:41,836 --> 00:13:43,716
Speaker 3: many different configurations.

241
00:13:43,716 --> 00:13:46,076
Speaker 1: You don't know the answer, you know the ant, but.

242
00:13:46,036 --> 00:13:48,356
Speaker 3: You know, what you do know is that there's a

243
00:13:48,836 --> 00:13:52,956
Speaker 3: there's a very large class of in of invalid solutions,

244
00:13:53,396 --> 00:13:56,956
Speaker 3: and then there's a smaller, uh but still very large

245
00:13:56,996 --> 00:13:58,636
Speaker 3: class of valid solutions.

246
00:13:58,836 --> 00:14:00,676
Speaker 1: Ok there's a lot of things that it is not,

247
00:14:00,916 --> 00:14:02,916
Speaker 1: and there's some things that it could be.

248
00:14:02,796 --> 00:14:06,156
Speaker 3: Could exactly incredibly well said, that's that's that's exactly right.

249
00:14:06,356 --> 00:14:08,796
Speaker 3: So here's a there's a really cool application of our

250
00:14:08,876 --> 00:14:12,196
Speaker 3: of our our technology and our approach. So the industry

251
00:14:12,236 --> 00:14:14,836
Speaker 3: standard approach to this is basically, what what does a

252
00:14:15,156 --> 00:14:18,636
Speaker 3: does a normal conventional company do with this gravitational noma. Well,

253
00:14:18,636 --> 00:14:20,556
Speaker 3: they do one of two things. Most of the time,

254
00:14:20,636 --> 00:14:23,396
Speaker 3: maybe ninety percent of the time. They'll just look at

255
00:14:23,396 --> 00:14:25,676
Speaker 3: it and say, okay, here, here is a gravitational anomaly.

256
00:14:25,876 --> 00:14:28,756
Speaker 3: This is hot, it's higher here than here. That's interesting,

257
00:14:28,796 --> 00:14:30,316
Speaker 3: and they just see it on a two D map.

258
00:14:30,356 --> 00:14:33,956
Speaker 2: Okay. So that doesn't do anything for the non uniqueness.

259
00:14:33,996 --> 00:14:35,276
Speaker 2: It just tells you. It just tells you what the

260
00:14:35,316 --> 00:14:37,956
Speaker 2: measurement is, okay. So that now they would just use it.

261
00:14:37,956 --> 00:14:41,156
Speaker 3: They would just say that, well, it's interesting because it's higher, Okay.

262
00:14:41,156 --> 00:14:46,236
Speaker 3: So what Cobold does is very very different and sort

263
00:14:46,276 --> 00:14:49,436
Speaker 3: of impossible to have done even ten years ago, maybe

264
00:14:49,436 --> 00:14:52,316
Speaker 3: even five years ago. What we do is we solve

265
00:14:52,716 --> 00:14:55,676
Speaker 3: a conditioned on a set of geologic hypotheses that we

266
00:14:55,716 --> 00:15:01,916
Speaker 3: find interesting. We solve for the full set of possible subsurfaces.

267
00:15:02,236 --> 00:15:06,316
Speaker 3: So we might actually test like a billion subsurfaces and

268
00:15:06,396 --> 00:15:12,356
Speaker 3: say nine hundred and nine d nine nine nine million,

269
00:15:12,476 --> 00:15:16,116
Speaker 3: nine hundred thousand of those don't match the data we tried.

270
00:15:16,156 --> 00:15:16,836
Speaker 2: They don't match it.

271
00:15:16,876 --> 00:15:19,236
Speaker 3: So there they've been rejected, But we still have one

272
00:15:19,316 --> 00:15:21,476
Speaker 3: hundred thousand now that do. So now we have all

273
00:15:21,556 --> 00:15:25,076
Speaker 3: of the good we've rejected the bad possibilities and we've

274
00:15:25,156 --> 00:15:27,236
Speaker 3: narrowed on the good possibilities. But it's still an incredibly

275
00:15:27,316 --> 00:15:28,036
Speaker 3: large search space.

276
00:15:28,516 --> 00:15:32,236
Speaker 1: Yes, impractical, impractically, you've got to get a lot smaller

277
00:15:32,276 --> 00:15:33,236
Speaker 1: for you to do anything right.

278
00:15:33,436 --> 00:15:36,276
Speaker 3: But there's there's a lot of information in the uncertainty

279
00:15:36,356 --> 00:15:39,716
Speaker 3: that we've now quantified. We've now quantified the uncertainty, and

280
00:15:39,756 --> 00:15:42,596
Speaker 3: so we apply something that we call efficacy of information,

281
00:15:42,996 --> 00:15:45,556
Speaker 3: which is is a phrase that we coined and you

282
00:15:45,556 --> 00:15:47,796
Speaker 3: can read scientific papers about it. It's it's a very

283
00:15:47,836 --> 00:15:50,636
Speaker 3: general and really cool concept, and it's also kind of

284
00:15:50,676 --> 00:15:52,836
Speaker 3: obvious actually, although it's hard to formalize.

285
00:15:52,956 --> 00:15:55,996
Speaker 1: But you look very happy right now. This is an

286
00:15:56,476 --> 00:15:59,596
Speaker 1: audio medium, Like your face is full of delight right now.

287
00:15:59,596 --> 00:16:01,276
Speaker 1: I don't want to interrupt to you. Keep going good.

288
00:16:01,596 --> 00:16:03,876
Speaker 3: Yeah, I'm excited. I love talking about this stuff because

289
00:16:03,876 --> 00:16:07,116
Speaker 3: it's super cool. So we say, okay, the basic idea

290
00:16:07,156 --> 00:16:11,436
Speaker 3: behind EOI is you're going to collect some information next, okay,

291
00:16:11,476 --> 00:16:14,796
Speaker 3: And that's really what every exploration process always is.

292
00:16:15,636 --> 00:16:17,556
Speaker 1: You're going to go try and get more information to

293
00:16:17,596 --> 00:16:19,876
Speaker 1: figure out what is going on in the rocks of

294
00:16:19,916 --> 00:16:20,356
Speaker 1: the surface.

295
00:16:20,396 --> 00:16:24,876
Speaker 3: So the question you obviously want to want to ask

296
00:16:25,076 --> 00:16:28,716
Speaker 3: is what next piece of information will I will tell

297
00:16:28,716 --> 00:16:33,316
Speaker 3: me the most given sort of uh per unit, per

298
00:16:33,396 --> 00:16:35,636
Speaker 3: unit of dollar that I expend, what am I going

299
00:16:35,716 --> 00:16:36,556
Speaker 3: to learn the most from?

300
00:16:36,996 --> 00:16:39,516
Speaker 1: What has what has the highest return on investments?

301
00:16:39,556 --> 00:16:39,716
Speaker 2: Yeah?

302
00:16:39,756 --> 00:16:42,276
Speaker 3: And from a from a knowledge an information perspective, what

303
00:16:42,316 --> 00:16:44,876
Speaker 3: am I going to learn the most from? And so

304
00:16:44,916 --> 00:16:46,436
Speaker 3: here's a way to think about that. It is the

305
00:16:46,436 --> 00:16:49,436
Speaker 3: piece of information that decreases your uncertainty the most. And

306
00:16:49,476 --> 00:16:52,076
Speaker 3: because we solved, we solved the we have the one

307
00:16:52,116 --> 00:16:56,316
Speaker 3: hundred thousand possible answers, right. Yeah, The piece of information

308
00:16:56,396 --> 00:16:59,756
Speaker 3: that will decrease our uncertainty the most is in fact,

309
00:17:00,276 --> 00:17:04,556
Speaker 3: the piece of information that tests the most number that

310
00:17:04,636 --> 00:17:08,236
Speaker 3: falsifies the highest number of those of those one hundred thousand. So,

311
00:17:08,516 --> 00:17:11,236
Speaker 3: for instance, say we're gonna we're gonna drill in different

312
00:17:11,276 --> 00:17:13,956
Speaker 3: in different directions. Okay, yeah, So we're gonna drill and

313
00:17:13,956 --> 00:17:17,516
Speaker 3: we're gonna intersect the sort of various predictions of concentration.

314
00:17:17,796 --> 00:17:19,596
Speaker 3: If we can do one drill hole and it would

315
00:17:19,596 --> 00:17:22,756
Speaker 3: test one hundred of the one hundred thousands, yeah yeah.

316
00:17:22,596 --> 00:17:23,516
Speaker 2: Okay, that's only.

317
00:17:23,596 --> 00:17:26,316
Speaker 3: That's only, we're only testing zero point one percent of

318
00:17:26,356 --> 00:17:27,316
Speaker 3: the possible answers.

319
00:17:27,476 --> 00:17:29,796
Speaker 1: Don't don't drill there, Yeah, don't do that.

320
00:17:29,836 --> 00:17:30,836
Speaker 2: We're gonna learn very little.

321
00:17:30,876 --> 00:17:33,396
Speaker 3: We're gonna end up with basically the same amount of uncertainty.

322
00:17:33,676 --> 00:17:35,876
Speaker 3: But if we could drill a different hole, a different

323
00:17:35,916 --> 00:17:39,316
Speaker 3: core that tests fifty thousand, say half.

324
00:17:39,076 --> 00:17:41,556
Speaker 2: Of them, we massively reduce our search space.

325
00:17:41,876 --> 00:17:46,716
Speaker 3: We we we falsify fifty fifty percent of the possible answers, right,

326
00:17:46,956 --> 00:17:49,596
Speaker 3: and sometimes we can actually falsify like eighty and ninety

327
00:17:49,636 --> 00:17:52,396
Speaker 3: percent of the possible answers. So we massively reduce our

328
00:17:52,436 --> 00:17:56,876
Speaker 3: search space. And we fundamentally it's the most most possible

329
00:17:56,876 --> 00:17:59,676
Speaker 3: information you can get per unit dollar. So every time

330
00:18:00,036 --> 00:18:03,276
Speaker 3: we go to collect any information, we try we try

331
00:18:03,316 --> 00:18:07,956
Speaker 3: to form a formally quantify the uncertainty, and then calculate

332
00:18:07,996 --> 00:18:10,676
Speaker 3: this EOI cont of ter, which is which is the

333
00:18:11,196 --> 00:18:13,636
Speaker 3: piece of information that we have the greatest expectation will

334
00:18:13,676 --> 00:18:17,036
Speaker 3: reduce our uncertainty the most compelling.

335
00:18:17,636 --> 00:18:18,756
Speaker 2: I'm glad you think so.

336
00:18:18,836 --> 00:18:20,396
Speaker 1: It would be nice to be able to do that

337
00:18:20,476 --> 00:18:22,596
Speaker 1: in life. More generally, it's.

338
00:18:22,476 --> 00:18:25,396
Speaker 3: Super super hard, and it comes. The really hard part

339
00:18:25,596 --> 00:18:30,556
Speaker 3: is quantifying the uncertainty correctly. Once you have that quantified correctly,

340
00:18:30,876 --> 00:18:33,796
Speaker 3: then calculating the EOI, like if you know for sure

341
00:18:34,116 --> 00:18:38,916
Speaker 3: you correctly quantified the uncertainty, optimize the calculating the EI

342
00:18:38,996 --> 00:18:40,396
Speaker 3: is kind of an engineering optimization.

343
00:18:40,436 --> 00:18:41,836
Speaker 2: It's kind of it's kind of straightforward.

344
00:18:42,236 --> 00:18:45,756
Speaker 1: In this instance, quantifying the uncertainty is basically how many

345
00:18:46,036 --> 00:18:49,076
Speaker 1: how many ways could the rock under the surface.

346
00:18:48,796 --> 00:18:50,916
Speaker 2: Be exactly Yeah, that's that's exactly right.

347
00:18:51,076 --> 00:18:53,596
Speaker 1: So you've been talking about sort of gathering this very

348
00:18:53,716 --> 00:18:57,796
Speaker 1: old school data and making it useful to you and

349
00:18:57,836 --> 00:19:00,156
Speaker 1: then what you do with it. There is another piece

350
00:19:00,316 --> 00:19:03,756
Speaker 1: of your data gathering operation, or another set of pieces

351
00:19:04,276 --> 00:19:07,316
Speaker 1: that are more high tech and that involve going out

352
00:19:07,316 --> 00:19:09,276
Speaker 1: into the world and getting new data that the then

353
00:19:09,396 --> 00:19:12,076
Speaker 1: already exist. And some of those are kind of fun

354
00:19:12,476 --> 00:19:15,116
Speaker 1: and so I want to talk about that a little bit.

355
00:19:15,396 --> 00:19:17,076
Speaker 1: Tell me about detecting muons.

356
00:19:17,916 --> 00:19:19,836
Speaker 3: Yeah, so this is this is this is kind of

357
00:19:19,836 --> 00:19:24,436
Speaker 3: one of our frontier of R and D projects within

358
00:19:24,476 --> 00:19:24,956
Speaker 3: the company.

359
00:19:25,036 --> 00:19:27,796
Speaker 1: Something we would the opposite of a hundred year old map.

360
00:19:27,876 --> 00:19:31,596
Speaker 3: Yeah, exactly exactly, And we have a lot of a

361
00:19:31,636 --> 00:19:33,716
Speaker 3: lot of physicists at the company, so they love the stuff.

362
00:19:33,876 --> 00:19:36,356
Speaker 3: So what is a muan. Let's start with that and

363
00:19:36,356 --> 00:19:38,676
Speaker 3: then I'll get to the why why they can be useful.

364
00:19:39,116 --> 00:19:41,836
Speaker 3: So in the so cosmic grays are hitting are hitting

365
00:19:42,276 --> 00:19:45,796
Speaker 3: air molecules in the in the upper atmosphere all the time,

366
00:19:46,236 --> 00:19:49,876
Speaker 3: and when when they collide, sometimes they produce they produce

367
00:19:50,036 --> 00:19:52,316
Speaker 3: muons in the in the in the reaction. So it's

368
00:19:52,356 --> 00:19:55,916
Speaker 3: a it's a it's a sub atomic uh particle, a muon,

369
00:19:56,236 --> 00:19:59,716
Speaker 3: and it's it travels very very very fast. It's a

370
00:19:59,796 --> 00:20:02,916
Speaker 3: sort of relativistic particle, travels near the near the speed

371
00:20:02,916 --> 00:20:03,156
Speaker 3: of light.

372
00:20:03,476 --> 00:20:05,316
Speaker 2: So right now muons.

373
00:20:04,916 --> 00:20:08,316
Speaker 3: Are showering through us, you and you in your studio

374
00:20:08,356 --> 00:20:10,476
Speaker 3: and me and my home. Newons are coming through us,

375
00:20:10,516 --> 00:20:12,276
Speaker 3: and they I think about like if you put your

376
00:20:12,276 --> 00:20:15,516
Speaker 3: hand flat, you can expect about one muon per second

377
00:20:15,716 --> 00:20:17,756
Speaker 3: to be going through you through your hand. You don't notice.

378
00:20:17,836 --> 00:20:20,556
Speaker 3: They mostly go right through you. Uh, it doesn't cause

379
00:20:20,556 --> 00:20:25,036
Speaker 3: you any harm. They do interact with electrons, and it

380
00:20:25,076 --> 00:20:28,356
Speaker 3: turns out that when they every time they interact with electron,

381
00:20:28,476 --> 00:20:29,436
Speaker 3: they they.

382
00:20:29,276 --> 00:20:30,316
Speaker 2: Slow down a little bit.

383
00:20:31,196 --> 00:20:31,476
Speaker 1: Uh.

384
00:20:31,516 --> 00:20:33,836
Speaker 3: And when they eventually when when they slow down, eventually

385
00:20:33,876 --> 00:20:36,236
Speaker 3: they slow down enough that they decay into other things.

386
00:20:36,236 --> 00:20:38,236
Speaker 2: Okay, so they they disappear.

387
00:20:38,556 --> 00:20:41,276
Speaker 3: So if you if you are measured, Let's say you

388
00:20:41,276 --> 00:20:43,156
Speaker 3: have a muon detector and it's sitting at the surface

389
00:20:43,396 --> 00:20:46,276
Speaker 3: and you're listening. You know, you like listen to its detection.

390
00:20:46,396 --> 00:20:50,116
Speaker 3: So it's like click click, click click. That's just telling you, okay,

391
00:20:50,156 --> 00:20:52,116
Speaker 3: mwan's going through means going through me's going through right.

392
00:20:52,156 --> 00:20:52,436
Speaker 2: Okay.

393
00:20:53,076 --> 00:20:56,756
Speaker 3: Now I drill a borehole and I start lowering the

394
00:20:56,836 --> 00:20:59,916
Speaker 3: muon detector into the borehole, and you know the rate

395
00:20:59,956 --> 00:21:00,716
Speaker 3: it was at the surface.

396
00:21:00,756 --> 00:21:02,316
Speaker 2: Then as I get to say one hundred meters, now

397
00:21:02,356 --> 00:21:02,876
Speaker 2: it'll be like.

398
00:21:02,996 --> 00:21:07,396
Speaker 3: Click, click click, And then as I go to like

399
00:21:07,396 --> 00:21:11,476
Speaker 3: five hundred meters, it'll be like click.

400
00:21:12,636 --> 00:21:12,916
Speaker 2: Okay.

401
00:21:13,196 --> 00:21:15,196
Speaker 3: What's happening is you're getting fewer and fewer muons are

402
00:21:15,236 --> 00:21:17,836
Speaker 3: hitting that location. And the reason is because you've got

403
00:21:17,836 --> 00:21:20,516
Speaker 3: so many more, so many more atoms between you and

404
00:21:20,596 --> 00:21:22,436
Speaker 3: the at and the top of the atmosphere that the

405
00:21:22,516 --> 00:21:25,356
Speaker 3: muon that the muons aren't surviving, they're they're hit.

406
00:21:25,356 --> 00:21:25,916
Speaker 1: They're hitting the.

407
00:21:25,996 --> 00:21:26,796
Speaker 2: R're hitting the rock.

408
00:21:27,756 --> 00:21:30,076
Speaker 3: And so now you think of think of the journey

409
00:21:30,116 --> 00:21:33,476
Speaker 3: of a specific muon as it's going through the rock. Okay,

410
00:21:33,596 --> 00:21:38,556
Speaker 3: the the muon that interacts with the fewest electrons is

411
00:21:38,556 --> 00:21:40,836
Speaker 3: the most likely to hit you, and the interro and

412
00:21:40,876 --> 00:21:43,876
Speaker 3: the muon that interacts with the most atoms will say

413
00:21:44,396 --> 00:21:46,756
Speaker 3: is is less likely to hit you because it's likely

414
00:21:46,836 --> 00:21:49,836
Speaker 3: to to decay, because it's likely to lose its energy.

415
00:21:49,876 --> 00:21:51,756
Speaker 3: You know, think of a think of a of a

416
00:21:51,876 --> 00:21:53,916
Speaker 3: of a ball bouncing hitting a bunch of other bals.

417
00:21:53,956 --> 00:21:54,156
Speaker 2: Right.

418
00:21:54,556 --> 00:21:57,196
Speaker 3: Yeah, So, now if you're sitting in the you're sitting

419
00:21:57,316 --> 00:22:00,196
Speaker 3: at the muon detector that's been lowered into into a

420
00:22:00,236 --> 00:22:02,556
Speaker 3: location underground, and you're looking up in kind of a

421
00:22:02,596 --> 00:22:07,716
Speaker 3: cone and you can look in three dimensions. Yeah, you're

422
00:22:07,716 --> 00:22:10,516
Speaker 3: seeing muons from stay say, they're coming from the right

423
00:22:10,596 --> 00:22:13,556
Speaker 3: a lot, but they're not coming from the left very much.

424
00:22:13,796 --> 00:22:14,996
Speaker 1: Yes, Yeah, that's.

425
00:22:14,796 --> 00:22:17,716
Speaker 3: Telling you something about the number of atoms, the relative

426
00:22:17,756 --> 00:22:20,076
Speaker 3: number of atoms to the left, which tells you about

427
00:22:20,116 --> 00:22:20,676
Speaker 3: the density.

428
00:22:21,076 --> 00:22:22,476
Speaker 2: That tells you the rocks to the.

429
00:22:22,476 --> 00:22:24,956
Speaker 3: Up and above you, into the left of you right

430
00:22:25,476 --> 00:22:28,236
Speaker 3: are denser than the rocks up into the right of you.

431
00:22:28,796 --> 00:22:31,876
Speaker 3: And denser might mean an ore body, it might mean

432
00:22:32,036 --> 00:22:34,876
Speaker 3: a high concentration set of metal in the rock.

433
00:22:35,236 --> 00:22:35,596
Speaker 2: Uh.

434
00:22:35,676 --> 00:22:38,876
Speaker 3: And so that is it allows us to probe in

435
00:22:38,956 --> 00:22:42,516
Speaker 3: really sophisticated ways the density of the Earth in the

436
00:22:42,556 --> 00:22:45,716
Speaker 3: same way we were talking about the gravitational force changes

437
00:22:45,716 --> 00:22:46,196
Speaker 3: around the Earth.

438
00:22:46,196 --> 00:22:48,676
Speaker 2: It's the same thing, but it's a much higher precision measurement.

439
00:22:49,076 --> 00:22:49,156
Speaker 1: UH.

440
00:22:49,276 --> 00:22:52,156
Speaker 2: So we've designed our our own novel muon detector.

441
00:22:52,196 --> 00:22:55,876
Speaker 3: We did it in collaboration with Occidental College and it's working,

442
00:22:56,356 --> 00:22:59,676
Speaker 3: and it's it's in a pilot hole, uh, collecting muons.

443
00:22:59,836 --> 00:23:01,596
Speaker 3: And then we have a bunch of new ideas about

444
00:23:01,596 --> 00:23:05,356
Speaker 3: how to use this this you this uh you kind

445
00:23:05,356 --> 00:23:07,516
Speaker 3: of new data data type. There's there's a few other

446
00:23:07,516 --> 00:23:10,996
Speaker 3: companies doing this, but it's very new, very new concept.

447
00:23:14,036 --> 00:23:17,996
Speaker 1: After the break from Fury to Practice, Kurt talks about

448
00:23:18,116 --> 00:23:34,076
Speaker 1: Cobold's huge copper discovery in Zambia. Let's talk about Zambia.

449
00:23:34,436 --> 00:23:37,316
Speaker 1: I want to talk about Zambia because it suggests that

450
00:23:37,436 --> 00:23:44,236
Speaker 1: your hypothesis for the company is worked right, Yeah, to.

451
00:23:44,676 --> 00:23:45,396
Speaker 2: A large degree.

452
00:23:45,476 --> 00:23:45,636
Speaker 1: Yes.

453
00:23:45,756 --> 00:23:48,436
Speaker 2: Let me first sEH, why are we in Zambia in

454
00:23:48,476 --> 00:23:49,036
Speaker 2: the first place.

455
00:23:49,556 --> 00:23:55,236
Speaker 3: So we look around the world and we evaluate jurisdictions

456
00:23:55,276 --> 00:24:00,876
Speaker 3: on four dimensions. Okay, our physical prospectivity or how we

457
00:24:00,996 --> 00:24:03,396
Speaker 3: perceive the physical prospect the likelihood that there's going to

458
00:24:03,436 --> 00:24:05,596
Speaker 3: be something new to discover a particular location.

459
00:24:06,476 --> 00:24:10,076
Speaker 2: Yeah, that's thing one. Thing two is the rule of law.

460
00:24:10,436 --> 00:24:12,556
Speaker 3: Right if we if we make it discovery, we have

461
00:24:12,596 --> 00:24:13,316
Speaker 3: a property right.

462
00:24:13,356 --> 00:24:17,036
Speaker 2: How robust is that property? Right? That's thing two. Thing

463
00:24:17,156 --> 00:24:19,556
Speaker 2: three is access to markets infrastructure.

464
00:24:19,596 --> 00:24:22,276
Speaker 3: Right. If you find something in the middle of Antarctica,

465
00:24:22,836 --> 00:24:23,956
Speaker 3: you're not going to be able to get it into

466
00:24:23,956 --> 00:24:26,276
Speaker 3: the market, no matter how great it is. And then

467
00:24:26,556 --> 00:24:30,636
Speaker 3: thing thing four is how much resistance or slash support

468
00:24:30,636 --> 00:24:32,956
Speaker 3: will there be to building a new industrial project in

469
00:24:32,956 --> 00:24:33,476
Speaker 3: that location?

470
00:24:33,916 --> 00:24:34,076
Speaker 2: Right?

471
00:24:34,156 --> 00:24:36,436
Speaker 3: If you find something in Palo Alto, no one's going

472
00:24:36,516 --> 00:24:38,796
Speaker 3: to allow you to build it, right, So it just doesn't.

473
00:24:38,836 --> 00:24:41,396
Speaker 1: You can't even build an apartment building there, much less

474
00:24:41,396 --> 00:24:43,196
Speaker 1: a lithium exactly exactly.

475
00:24:43,236 --> 00:24:45,156
Speaker 3: So those are the four dimensions we look at, and

476
00:24:45,196 --> 00:24:48,076
Speaker 3: we looked around the world early on Zambia rose to

477
00:24:48,116 --> 00:24:50,436
Speaker 3: the very top on all four of those. It's a

478
00:24:50,476 --> 00:24:55,156
Speaker 3: fantastic jurisdiction. It's the most consistent and stable democracy in

479
00:24:55,196 --> 00:24:59,676
Speaker 3: Southern Africa. The physical prospectivity is tremendous because there's been

480
00:24:59,676 --> 00:25:02,756
Speaker 3: mining there for one hundred years and so well you

481
00:25:02,996 --> 00:25:04,836
Speaker 3: might look at that and say, well, sure it was

482
00:25:04,876 --> 00:25:07,196
Speaker 3: a good place to look a hundred years ago, but

483
00:25:07,716 --> 00:25:08,756
Speaker 3: isn't it all picked over?

484
00:25:09,156 --> 00:25:10,436
Speaker 2: And the answer is definitively not.

485
00:25:11,396 --> 00:25:15,356
Speaker 3: This is easy to verify because the uh, basically all

486
00:25:15,436 --> 00:25:18,036
Speaker 3: the all the deposits that were mined over the last

487
00:25:18,076 --> 00:25:21,676
Speaker 3: one hundred years were actually sticking out of the surface.

488
00:25:21,756 --> 00:25:24,236
Speaker 3: They were they were known about a century ago, and

489
00:25:24,276 --> 00:25:28,116
Speaker 3: there's been almost no exploration into the deep parts of

490
00:25:28,156 --> 00:25:31,476
Speaker 3: the basins. Uh, the what we'd call blind exploration. Right,

491
00:25:31,756 --> 00:25:34,076
Speaker 3: this is not directly directly evident, like you can see

492
00:25:34,076 --> 00:25:36,676
Speaker 3: it at the surface. There's there's been almost none. It

493
00:25:36,716 --> 00:25:39,196
Speaker 3: was this kind of perfect perfect location in that sense.

494
00:25:39,596 --> 00:25:43,316
Speaker 3: It's right on the Central African copper belt, which provides

495
00:25:43,396 --> 00:25:45,756
Speaker 3: a significant majority of the world's copper, so it's easy

496
00:25:45,756 --> 00:25:48,196
Speaker 3: to get it to market, right, and it's a legacy

497
00:25:48,196 --> 00:25:51,516
Speaker 3: mining country. That's that's very supportive of development, right, and

498
00:25:51,756 --> 00:25:53,956
Speaker 3: so it's like perfect it was. It rose to the

499
00:25:53,956 --> 00:25:56,196
Speaker 3: top across the board and we'd love it, and we

500
00:25:56,316 --> 00:25:59,196
Speaker 3: love our Zambian colleagues, uh, and we think it's just

501
00:25:59,356 --> 00:26:01,796
Speaker 3: it's one of the best jurisdictions in the world for us,

502
00:26:01,796 --> 00:26:02,516
Speaker 3: for us to operate.

503
00:26:03,076 --> 00:26:05,236
Speaker 2: So that that's why we were there in the first instance.

504
00:26:05,476 --> 00:26:09,596
Speaker 3: And we we we we started we started exploring in

505
00:26:09,676 --> 00:26:16,116
Speaker 3: twenty twenty there in very modest fashion, loosely, and we

506
00:26:17,596 --> 00:26:20,276
Speaker 3: but we were exploring in areas right in the heart

507
00:26:20,276 --> 00:26:22,996
Speaker 3: of in the heart of active mining basins because again

508
00:26:23,316 --> 00:26:26,276
Speaker 3: there were sort of active minds, but if you went

509
00:26:26,476 --> 00:26:28,756
Speaker 3: out into the deeper parts of the basin, deeper than

510
00:26:28,796 --> 00:26:31,436
Speaker 3: five hundred six hundred meters, there were areas that just

511
00:26:31,476 --> 00:26:33,076
Speaker 3: had never been probed or explored.

512
00:26:33,476 --> 00:26:36,076
Speaker 1: Yeah, So like briefly, what did you find and how

513
00:26:36,116 --> 00:26:36,836
Speaker 1: did you find it?

514
00:26:37,196 --> 00:26:39,836
Speaker 3: Looking at all the data we had, some of our

515
00:26:39,876 --> 00:26:45,436
Speaker 3: geoscientists had really really clever ideas about how the mineralogy

516
00:26:45,596 --> 00:26:48,796
Speaker 3: was changing and how we actually might have very very

517
00:26:48,836 --> 00:26:52,916
Speaker 3: distinct mineralogy, so we might have areas where it looks

518
00:26:53,036 --> 00:26:56,956
Speaker 3: like it's all this kind of one distribution, one sort

519
00:26:56,956 --> 00:26:59,636
Speaker 3: of set of statistics, but actually, as you cross this boundary,

520
00:26:59,876 --> 00:27:03,716
Speaker 3: it's a totally different geochemical system. Different sets of geochemical

521
00:27:03,716 --> 00:27:06,236
Speaker 3: reactions occurred. So if you're able to draw that boundary

522
00:27:06,236 --> 00:27:09,316
Speaker 3: and then only go and explore within that comple CAD

523
00:27:09,396 --> 00:27:13,196
Speaker 3: three dimensional boundary, then you'd consistently have high grade and

524
00:27:13,236 --> 00:27:13,836
Speaker 3: thick right.

525
00:27:13,876 --> 00:27:15,716
Speaker 2: That was that was the sort of assertion.

526
00:27:15,476 --> 00:27:18,516
Speaker 3: What you have in a in the in the reservoir,

527
00:27:18,556 --> 00:27:21,316
Speaker 3: we have nine in the legacy data. Ninety five percent

528
00:27:21,356 --> 00:27:24,476
Speaker 3: of the data is this low, you know, kind of

529
00:27:24,676 --> 00:27:27,636
Speaker 3: modest grade thin stuff. And then there's five percent of

530
00:27:27,676 --> 00:27:29,956
Speaker 3: the data is this higher grade stuff. And if we

531
00:27:30,036 --> 00:27:32,676
Speaker 3: and if we drill there and within that boundary and

532
00:27:32,756 --> 00:27:34,556
Speaker 3: only drill there, it's gonna be.

533
00:27:34,556 --> 00:27:37,676
Speaker 2: Good whole, good hohole, good hold, good hoole, good hole. Right,

534
00:27:37,756 --> 00:27:38,756
Speaker 2: And that's and now.

535
00:27:38,636 --> 00:27:42,116
Speaker 1: That's the hypothesis. Yes, you test the hypothesis and does

536
00:27:42,116 --> 00:27:43,156
Speaker 1: it happen, right.

537
00:27:43,116 --> 00:27:45,316
Speaker 3: Yeah, exactly, So you found it, Yes, and we proved it,

538
00:27:45,356 --> 00:27:47,916
Speaker 3: and it's and it's there, and it's it's it's very

539
00:27:48,076 --> 00:27:51,076
Speaker 3: it's it's gone from marginally economic to very economic. For

540
00:27:51,116 --> 00:27:54,356
Speaker 3: the same unit of rock that we move, we we

541
00:27:54,356 --> 00:27:57,636
Speaker 3: we sell ten times as much copper as the average

542
00:27:57,676 --> 00:27:58,516
Speaker 3: copper mind today.

543
00:27:58,916 --> 00:28:02,476
Speaker 1: Yes, okay, so you found it correct. The hypothesis was true,

544
00:28:02,956 --> 00:28:07,076
Speaker 1: the system worked. You found it correct. Now you are

545
00:28:07,236 --> 00:28:11,276
Speaker 1: like gonna become a why different company. It seems to

546
00:28:11,316 --> 00:28:14,196
Speaker 1: me right, Like you are in Silicon Valley, you are

547
00:28:14,236 --> 00:28:16,836
Speaker 1: working with data scientists, you have a technical background. You

548
00:28:16,916 --> 00:28:20,676
Speaker 1: have been running essentially a high tech startup. You are

549
00:28:20,716 --> 00:28:23,596
Speaker 1: about to be running a mining company. Where like your

550
00:28:23,636 --> 00:28:26,196
Speaker 1: problems are not just being very clever and hiring the

551
00:28:26,276 --> 00:28:29,756
Speaker 1: right AI people. They're like, you know, getting the US

552
00:28:29,916 --> 00:28:34,516
Speaker 1: government to finance a railroad in Zambia and making sure

553
00:28:34,596 --> 00:28:39,156
Speaker 1: that the Zambians like you. Like that seems entirely different

554
00:28:39,316 --> 00:28:40,876
Speaker 1: than what you have been doing so far.

555
00:28:42,796 --> 00:28:47,756
Speaker 3: Partially true, there's it's it's a little more continuous than

556
00:28:47,836 --> 00:28:48,636
Speaker 3: you might have implied.

557
00:28:48,836 --> 00:28:51,996
Speaker 1: Yeah, I'm just trying to sight but but it's not.

558
00:28:52,556 --> 00:28:56,476
Speaker 1: It's not a regular startup anymore, correct, Like maybe it

559
00:28:56,556 --> 00:28:59,476
Speaker 1: never was. But like it's a super different job. It's

560
00:28:59,516 --> 00:29:02,476
Speaker 1: a super different skill set. I mean, I'm sure to

561
00:29:02,476 --> 00:29:04,676
Speaker 1: some extent you've been dealing with this, but it's really

562
00:29:04,756 --> 00:29:06,756
Speaker 1: different than what we have been talking about. It's a

563
00:29:06,836 --> 00:29:10,396
Speaker 1: whole different universe of problems and hard things to deal

564
00:29:10,476 --> 00:29:11,876
Speaker 1: with and a very different domain.

565
00:29:12,036 --> 00:29:14,396
Speaker 3: Sure, right, No, you're you're, you're these are these are

566
00:29:14,396 --> 00:29:16,676
Speaker 3: excellent questions. So you can think about think about the

567
00:29:16,676 --> 00:29:21,596
Speaker 3: company now as there is the discovery machine. The discovery

568
00:29:21,636 --> 00:29:24,116
Speaker 3: machine is everything we've been talking about, right, the and

569
00:29:24,236 --> 00:29:27,676
Speaker 3: the discovery the unit economics of exploration are great, right

570
00:29:27,716 --> 00:29:31,436
Speaker 3: because the you can make a hundred times your money

571
00:29:31,676 --> 00:29:34,916
Speaker 3: or even more on proving the deposit because it's worth

572
00:29:34,916 --> 00:29:36,956
Speaker 3: so much once it's clearly clearly economic.

573
00:29:37,196 --> 00:29:39,796
Speaker 1: Right, so you could just sell the rights in some fashion. Right,

574
00:29:39,796 --> 00:29:41,676
Speaker 1: you don't have to be a mining company. You could

575
00:29:41,676 --> 00:29:42,756
Speaker 1: be a discovery company.

576
00:29:42,796 --> 00:29:46,196
Speaker 3: Correct, And it's correct, so so so the most important

577
00:29:46,236 --> 00:29:47,876
Speaker 3: part of them, the heart and soul of the company

578
00:29:47,956 --> 00:29:51,716
Speaker 3: is the discovery machine. Now we have we have at

579
00:29:51,796 --> 00:29:55,076
Speaker 3: least one deposit that is sort of unambiguously going to

580
00:29:55,116 --> 00:29:55,676
Speaker 3: be a mine.

581
00:29:56,316 --> 00:29:56,716
Speaker 1: Uh.

582
00:29:56,796 --> 00:29:59,396
Speaker 3: And and the question is what happens from here with

583
00:29:59,476 --> 00:30:02,516
Speaker 3: that mind and the odds are very high that we're

584
00:30:02,556 --> 00:30:06,556
Speaker 3: going to bring in a partner uh with complimentary capabilities

585
00:30:06,876 --> 00:30:09,756
Speaker 3: to sort of help bring it to production and bring

586
00:30:09,836 --> 00:30:10,396
Speaker 3: in a partner.

587
00:30:10,476 --> 00:30:10,676
Speaker 1: You know.

588
00:30:10,836 --> 00:30:12,636
Speaker 2: It s a simple way to think of this is

589
00:30:13,356 --> 00:30:15,036
Speaker 2: we own eighty percent of it.

590
00:30:15,076 --> 00:30:19,196
Speaker 3: Now the Zambian Parastatal Mining company owns twenty percent of it.

591
00:30:19,196 --> 00:30:21,556
Speaker 3: It's worth a certain amount. You could imagine what it's

592
00:30:21,836 --> 00:30:24,676
Speaker 3: someone would pay to own it entirely, and.

593
00:30:24,636 --> 00:30:27,316
Speaker 1: That order of magnitude is billions of dollars.

594
00:30:27,356 --> 00:30:29,636
Speaker 2: It's like, right, yeah, yeah, yep.

595
00:30:30,236 --> 00:30:33,556
Speaker 3: And so someone could come in and pay and contribute

596
00:30:33,596 --> 00:30:36,156
Speaker 3: to capital and capabilities to bring it into a mine

597
00:30:36,196 --> 00:30:38,756
Speaker 3: and we would still have and we can principle spend

598
00:30:38,756 --> 00:30:41,316
Speaker 3: no more money and we still have a large share

599
00:30:41,396 --> 00:30:43,156
Speaker 3: of all the future cash flows.

600
00:30:43,196 --> 00:30:45,436
Speaker 1: It's worth a lot to have the steak you have

601
00:30:45,596 --> 00:30:47,516
Speaker 1: in this thing. Yeah, you have a set of choices

602
00:30:47,556 --> 00:30:49,756
Speaker 1: about what to do with it and how much sort

603
00:30:49,756 --> 00:30:52,236
Speaker 1: of money to take and how much of your interest

604
00:30:52,316 --> 00:30:52,716
Speaker 1: to sell.

605
00:30:52,836 --> 00:30:54,316
Speaker 2: Correct, That's that's exactly right.

606
00:30:54,396 --> 00:30:57,156
Speaker 1: So is the answer you're trying to figure out how

607
00:30:57,236 --> 00:31:00,476
Speaker 1: much you're gonna be a mining company versus a discovery company.

608
00:31:00,596 --> 00:31:02,556
Speaker 3: I mean, it's that yeah, And we know, we know

609
00:31:02,716 --> 00:31:05,676
Speaker 3: for sure that the most important thing is to not

610
00:31:05,676 --> 00:31:09,316
Speaker 3: not weaken the discovery engine, right, that's the most most

611
00:31:09,316 --> 00:31:11,116
Speaker 3: of them. And there's a lot of culture around the

612
00:31:11,156 --> 00:31:13,076
Speaker 3: discovery engine. There's a lot of attention. That's the most

613
00:31:13,076 --> 00:31:16,676
Speaker 3: important thing. We also know that it's it's very very

614
00:31:16,716 --> 00:31:22,676
Speaker 3: important that this that the value potential gets realized in

615
00:31:22,956 --> 00:31:24,236
Speaker 3: the Zambian deposit.

616
00:31:24,596 --> 00:31:27,036
Speaker 1: Is there about to be some deal if this show

617
00:31:27,076 --> 00:31:29,156
Speaker 1: comes out in two weeks?

618
00:31:29,676 --> 00:31:31,956
Speaker 3: Actually no, and I can I can confirm that we're

619
00:31:31,996 --> 00:31:33,916
Speaker 3: not actually looking for a partner for a couple of years.

620
00:31:34,436 --> 00:31:37,396
Speaker 3: Actually we're not. We won't formally partner with anyone for

621
00:31:37,436 --> 00:31:39,396
Speaker 3: a couple of years. And the reason they're actually really obvious.

622
00:31:39,596 --> 00:31:42,596
Speaker 3: It's just that the the what what we have discovered

623
00:31:42,636 --> 00:31:46,316
Speaker 3: sits on about four square kilometers and there's another one

624
00:31:46,396 --> 00:31:49,276
Speaker 3: hundred and fifty square kilometers on the license that we

625
00:31:49,396 --> 00:31:53,756
Speaker 3: own that are totally untouched, totally unexplored, right, completely unexplored.

626
00:31:53,796 --> 00:31:55,996
Speaker 3: So we are going to fully explore that whole area

627
00:31:56,396 --> 00:31:59,196
Speaker 3: and totally know what we have before we formally do

628
00:31:59,516 --> 00:32:02,996
Speaker 3: any partnership. But we're also building the capabilities to take

629
00:32:03,236 --> 00:32:06,556
Speaker 3: to take the project as far as we we want to.

630
00:32:06,636 --> 00:32:08,596
Speaker 2: Right, so we're hired. We recently hired.

631
00:32:08,956 --> 00:32:12,316
Speaker 3: We stood up an amazing Zambian leadership team of people

632
00:32:12,396 --> 00:32:20,076
Speaker 3: of project developers and engineers, metallurgists, et cetera, hydrologists to

633
00:32:20,116 --> 00:32:23,076
Speaker 3: continue to do the engineering optimization of what a mind

634
00:32:23,116 --> 00:32:25,956
Speaker 3: will look like in this location, right, And so we're

635
00:32:25,956 --> 00:32:27,596
Speaker 3: doing all of that because that that has to be

636
00:32:27,596 --> 00:32:30,396
Speaker 3: done anyway. It's it adds a ton of value to

637
00:32:30,476 --> 00:32:36,076
Speaker 3: figure out exactly how you'll optimize the operations and and

638
00:32:36,116 --> 00:32:37,716
Speaker 3: it just moves the project forward.

639
00:32:37,996 --> 00:32:42,236
Speaker 2: And our our goal, our our stated intention.

640
00:32:42,156 --> 00:32:45,196
Speaker 3: Is to start construction on the mine within two years,

641
00:32:45,636 --> 00:32:47,076
Speaker 3: uh and and to have.

642
00:32:47,076 --> 00:32:49,236
Speaker 2: It in production in the early part of the next decade.

643
00:32:49,916 --> 00:32:52,836
Speaker 1: So I know that we need a lot of copper,

644
00:32:52,956 --> 00:32:56,756
Speaker 1: and it's great that you just found more. It is

645
00:32:56,836 --> 00:33:00,516
Speaker 1: also the case that minds have often been bad for

646
00:33:00,556 --> 00:33:02,596
Speaker 1: the places where the minds were and for the people

647
00:33:02,636 --> 00:33:05,596
Speaker 1: who worked in the minds, Like how do you deal

648
00:33:05,636 --> 00:33:08,076
Speaker 1: with that, you know, harming people and the world.

649
00:33:08,436 --> 00:33:11,636
Speaker 3: What's super ex about this particular deposit, and in general

650
00:33:11,716 --> 00:33:14,756
Speaker 3: the deposits we're looking for is super high grade, right,

651
00:33:14,796 --> 00:33:16,956
Speaker 3: ten times higher grade than the average.

652
00:33:16,596 --> 00:33:19,956
Speaker 2: Copper mine around the world. That means ten times less

653
00:33:19,956 --> 00:33:22,356
Speaker 2: waste for the same amount of copper. Huh Okay. It's

654
00:33:22,396 --> 00:33:25,276
Speaker 2: also an underground mine as opposed to an open pit mine.

655
00:33:25,556 --> 00:33:28,516
Speaker 2: So when you consider the overburden that open pit minds

656
00:33:28,556 --> 00:33:30,836
Speaker 2: big holes in the ground would otherwise have it actually

657
00:33:30,876 --> 00:33:34,676
Speaker 2: ends up being about thirty times less waste, and we're

658
00:33:34,716 --> 00:33:37,356
Speaker 2: going to take almost all of that waste. And as

659
00:33:37,396 --> 00:33:41,596
Speaker 2: we mind, as we excavate the locations underground, we take

660
00:33:42,156 --> 00:33:44,156
Speaker 2: we take the waste and we put it back in

661
00:33:44,436 --> 00:33:47,116
Speaker 2: and we backfill is what it's called. We we we

662
00:33:47,156 --> 00:33:48,396
Speaker 2: stuff it back into the area.

663
00:33:48,436 --> 00:33:50,676
Speaker 3: So at any given time, there's only a modest volume,

664
00:33:50,756 --> 00:33:55,196
Speaker 3: you know, modest cavity that's open. And then in terms

665
00:33:55,236 --> 00:34:00,156
Speaker 3: of the we are passionate and obsessed with skills transfer

666
00:34:00,396 --> 00:34:02,196
Speaker 3: and so it's the reason that we are really building

667
00:34:02,236 --> 00:34:05,636
Speaker 3: a Zambian mining company to develop this project. Our CEO,

668
00:34:05,916 --> 00:34:12,276
Speaker 3: vi Kay Mackay is uh the CEO of Cobold Africa.

669
00:34:13,356 --> 00:34:16,716
Speaker 3: We have ninety percent of the employees in country, our

670
00:34:16,836 --> 00:34:20,956
Speaker 3: Zambian chief metallurgist, chief mind engineer, chief project director, right,

671
00:34:22,556 --> 00:34:27,636
Speaker 3: a psych geologist. All of them are our Zambian and

672
00:34:27,716 --> 00:34:33,716
Speaker 3: they're extraordinary and we're investing tremendous amounts in helping them

673
00:34:34,316 --> 00:34:38,436
Speaker 3: get be the best professionals they can be because we're

674
00:34:38,436 --> 00:34:40,956
Speaker 3: going to be there for fifty years, right, and we

675
00:34:41,036 --> 00:34:42,436
Speaker 3: want to be there for fifty years.

676
00:34:43,156 --> 00:34:47,556
Speaker 1: So what's the next big discovery?

677
00:34:47,796 --> 00:34:55,596
Speaker 2: So I'm predicted I will predict nickel and Canada, lithium

678
00:34:55,796 --> 00:35:05,796
Speaker 2: in Australia, and another and another copper discovery in Zambia.

679
00:35:07,316 --> 00:35:11,916
Speaker 1: I appreciate the specificity that I love falsifiable prediction.

680
00:35:12,356 --> 00:35:15,196
Speaker 2: God bless you. So something I make.

681
00:35:15,316 --> 00:35:18,716
Speaker 3: I make wagers all the time, weird, dis weird wagers.

682
00:35:18,916 --> 00:35:21,156
Speaker 3: But it's not it's not because I'm a gambling man.

683
00:35:21,196 --> 00:35:23,196
Speaker 3: I've never played a spin of roulette in my life, right,

684
00:35:23,236 --> 00:35:24,956
Speaker 3: But what I love to do is people make vague

685
00:35:24,996 --> 00:35:26,996
Speaker 3: predictions about the future, and I try to I try

686
00:35:27,036 --> 00:35:29,356
Speaker 3: to pin them down on something that's clearly testable, and

687
00:35:29,396 --> 00:35:32,156
Speaker 3: then we and then we usually bet, like a nice

688
00:35:32,156 --> 00:35:34,036
Speaker 3: bottle of wine or a dinner that we enjoy together.

689
00:35:34,076 --> 00:35:36,556
Speaker 3: So it's you know, it's it's fun. But the loser

690
00:35:36,596 --> 00:35:39,076
Speaker 3: pay is obviously uh and yeah, So this is so

691
00:35:39,116 --> 00:35:43,356
Speaker 3: we have the in Fact company. A big part of

692
00:35:43,356 --> 00:35:45,836
Speaker 3: our company culture is what we call the culture of falsification. Right,

693
00:35:45,836 --> 00:35:48,276
Speaker 3: So when you go out to test hypothesis, your job

694
00:35:48,356 --> 00:35:51,356
Speaker 3: is not to collect information to confirm that hypothesis, because

695
00:35:51,396 --> 00:35:53,196
Speaker 3: you can always do that. You could always paint a

696
00:35:53,236 --> 00:35:56,316
Speaker 3: new story. Right, that's inductive reasoning, it's invalid, yeah, right,

697
00:35:56,356 --> 00:35:58,676
Speaker 3: What your job is to go out is to tell

698
00:35:58,716 --> 00:36:00,236
Speaker 3: me how you're gonna test it, how you're gonna prove

699
00:36:00,236 --> 00:36:02,316
Speaker 3: it wrong, and go and go falsify it.

700
00:36:02,316 --> 00:36:03,996
Speaker 2: And you either one of two things happens.

701
00:36:04,036 --> 00:36:06,356
Speaker 3: You either successfully falsify it, in which we move on

702
00:36:06,396 --> 00:36:09,756
Speaker 3: and we celebrate that, we celebrate falsification, or you fail

703
00:36:09,796 --> 00:36:12,556
Speaker 3: to falsify it, which means, okay, how do okay, so

704
00:36:12,556 --> 00:36:13,396
Speaker 3: it's not dead yet?

705
00:36:13,716 --> 00:36:15,076
Speaker 2: How do we how do we now? How do we

706
00:36:15,076 --> 00:36:15,396
Speaker 2: test it?

707
00:36:15,396 --> 00:36:15,516
Speaker 1: Now?

708
00:36:15,556 --> 00:36:17,276
Speaker 2: How do we test it again? Right? Death?

709
00:36:18,516 --> 00:36:21,636
Speaker 1: Uh? And what's the most efficient way? What's the highest

710
00:36:21,716 --> 00:36:22,556
Speaker 1: return we can get?

711
00:36:22,716 --> 00:36:24,956
Speaker 2: God? Question, what's the EOI? We got it.

712
00:36:28,076 --> 00:36:41,916
Speaker 1: We'll be back in a minute with the lightning round. Okay,

713
00:36:41,916 --> 00:36:44,876
Speaker 1: we're gonna finish with the lightning round. And so I

714
00:36:44,876 --> 00:36:48,396
Speaker 1: got to ask, what is one weird bet you made.

715
00:36:49,716 --> 00:36:50,956
Speaker 2: Since we just had the Olympics.

716
00:36:50,996 --> 00:36:55,476
Speaker 3: I bet that I bet that Usain Bolt's nine point

717
00:36:55,556 --> 00:36:58,276
Speaker 3: five eight second world record in the one hundred meter

718
00:36:58,396 --> 00:37:00,316
Speaker 3: will still be the world record in the year twenty

719
00:37:00,316 --> 00:37:02,516
Speaker 3: thirty six, by the end of the year twenty thirty six.

720
00:37:02,636 --> 00:37:05,036
Speaker 1: That's a long bet. You're playing the long day.

721
00:37:05,156 --> 00:37:08,036
Speaker 3: Yeah, that's a long standing record. He said it in

722
00:37:08,076 --> 00:37:08,876
Speaker 3: two thousand and nine.

723
00:37:09,196 --> 00:37:11,436
Speaker 1: It's crazy for a record to last that long.

724
00:37:11,556 --> 00:37:13,316
Speaker 2: Yeah, would you like to would you like to make

725
00:37:13,356 --> 00:37:14,116
Speaker 2: that wager with me?

726
00:37:14,356 --> 00:37:16,836
Speaker 1: I don't have enough information. I know enough to know

727
00:37:16,876 --> 00:37:19,716
Speaker 1: that I'm ignorant. But what made you make the bet? Uh?

728
00:37:20,076 --> 00:37:23,356
Speaker 2: It's a dramatic outlier time. It's a total outlier. Yeah.

729
00:37:23,356 --> 00:37:26,236
Speaker 3: And he has the four fastest time, so nobody you

730
00:37:26,276 --> 00:37:28,956
Speaker 3: know that the next fastest human is is the fifth

731
00:37:29,036 --> 00:37:29,796
Speaker 3: fifth fastest time.

732
00:37:29,836 --> 00:37:32,596
Speaker 1: It's like he is an outlier and that time is

733
00:37:32,636 --> 00:37:33,956
Speaker 1: an outlier for him.

734
00:37:34,076 --> 00:37:35,236
Speaker 2: Correct? U huh correct?

735
00:37:38,156 --> 00:37:39,036
Speaker 1: What's a cobalt?

736
00:37:40,036 --> 00:37:44,476
Speaker 2: Oh? Good question. So it's a it's a creature from

737
00:37:44,556 --> 00:37:48,916
Speaker 2: German mythology that lives underground, kind of like a goblin

738
00:37:49,076 --> 00:37:53,316
Speaker 2: like creature. Uh, lives underground and controls the mineral wealth

739
00:37:53,356 --> 00:37:53,836
Speaker 2: of the earth.

740
00:37:55,276 --> 00:37:55,556
Speaker 1: Huh.

741
00:37:56,436 --> 00:38:00,356
Speaker 3: And it's also the namesake for the word cobalt, uh.

742
00:37:59,716 --> 00:38:01,396
Speaker 2: For the for the for the metal cobalt.

743
00:38:02,316 --> 00:38:05,036
Speaker 1: Right. I mean, as I understand it, people used to

744
00:38:05,036 --> 00:38:07,796
Speaker 1: think cobalt was bad, right, and then now we're like, oh,

745
00:38:07,876 --> 00:38:09,676
Speaker 1: actually cobalt is good. Good. Uh.

746
00:38:10,436 --> 00:38:13,436
Speaker 3: It looks a lot like nickel sulfides when it's a

747
00:38:13,556 --> 00:38:17,076
Speaker 3: cobalt arsenide and arsenic is toxic, so it would it

748
00:38:17,116 --> 00:38:19,516
Speaker 3: would poison you know, poison miners, and they called it

749
00:38:19,516 --> 00:38:20,316
Speaker 3: the goblin metal.

750
00:38:21,196 --> 00:38:23,396
Speaker 1: What's one thing I should do if I go to Zambia?

751
00:38:24,076 --> 00:38:27,196
Speaker 3: Oh that's I love that question. Well, you can't miss

752
00:38:27,236 --> 00:38:29,716
Speaker 3: most tuna. I call it, we call it most tuna.

753
00:38:29,916 --> 00:38:33,356
Speaker 3: That's the traditional name. It means the smoke that thunders.

754
00:38:33,396 --> 00:38:37,476
Speaker 3: You will know it as Victoria falls. It is completely spectacular.

755
00:38:38,796 --> 00:38:41,796
Speaker 3: Niagara Falls is amazing. It blows it away, totally blows

756
00:38:41,836 --> 00:38:42,116
Speaker 3: it away.

757
00:38:42,156 --> 00:38:44,396
Speaker 2: You just can't. You just can't go to Zambia and

758
00:38:44,436 --> 00:38:46,436
Speaker 2: miss and miss most Tuna.

759
00:38:48,756 --> 00:38:52,316
Speaker 1: We've talked a lot about, you know, trying to predict

760
00:38:52,316 --> 00:38:56,116
Speaker 1: things and trying to quantify uncertainty in the context of

761
00:38:56,756 --> 00:39:04,076
Speaker 1: your company. Do you think that way outside of work? Uh?

762
00:39:05,116 --> 00:39:07,916
Speaker 3: Well, yeah, I guess the way I make, you know,

763
00:39:08,036 --> 00:39:10,956
Speaker 3: wager on things, right, is a form of.

764
00:39:12,516 --> 00:39:14,236
Speaker 2: It's certainly the way I think of.

765
00:39:14,756 --> 00:39:17,516
Speaker 3: Like, I try to be scientific in every aspect of

766
00:39:17,556 --> 00:39:21,876
Speaker 3: my life, and I say that what science is not

767
00:39:22,876 --> 00:39:26,956
Speaker 3: is empiricism, right. It is not looking at the data

768
00:39:27,356 --> 00:39:30,596
Speaker 3: and drawing the inevitable conclusion, because there's no such thing

769
00:39:31,036 --> 00:39:33,036
Speaker 3: you can look at with any set of data. You

770
00:39:33,076 --> 00:39:36,476
Speaker 3: can tell you can fit many, many, many hypotheses that

771
00:39:36,636 --> 00:39:38,956
Speaker 3: explain the data. Right, that's always this is this is

772
00:39:38,956 --> 00:39:42,036
Speaker 3: the non unique thing where it's it's always true. Basically

773
00:39:42,076 --> 00:39:45,796
Speaker 3: it's always true. And so science really is about myth making.

774
00:39:45,916 --> 00:39:48,476
Speaker 3: It's about it's about making a making up a myth

775
00:39:48,836 --> 00:39:52,356
Speaker 3: that explains the data. That's your hypothesis. The difference between

776
00:39:52,396 --> 00:39:55,556
Speaker 3: science and religion is that we test our myths. That's

777
00:39:55,596 --> 00:39:58,916
Speaker 3: the difference, right, That's that's of good science, right.

778
00:39:58,756 --> 00:40:01,996
Speaker 1: And like in your heart, you should want to disprove it, right,

779
00:40:02,076 --> 00:40:05,636
Speaker 1: Like if you're really the best scientist, you should want

780
00:40:05,676 --> 00:40:07,116
Speaker 1: to prove yourself wrong.

781
00:40:07,316 --> 00:40:10,356
Speaker 3: Yes, that's this the thing, as scientists can say that

782
00:40:10,716 --> 00:40:13,396
Speaker 3: means you learn something you really only learn when you

783
00:40:13,516 --> 00:40:14,276
Speaker 3: realize you were wrong.

784
00:40:17,756 --> 00:40:21,156
Speaker 1: Kurthouse is the co founder and CEO of Cobold Medals.

785
00:40:21,956 --> 00:40:25,196
Speaker 1: Today's show was produced by Gabriel Hunter Cheng. It was

786
00:40:25,476 --> 00:40:28,956
Speaker 1: edited by Lyddy jeene Kott and engineered by Sarah Bruguier.

787
00:40:29,436 --> 00:40:32,956
Speaker 1: You can email us at problem at Pushkin dot FM.

788
00:40:33,156 --> 00:40:35,476
Speaker 1: I'm Jacob Goldstein and we'll be back next week with

789
00:40:35,556 --> 00:40:48,316
Speaker 1: another episode of What's Your Problem