1
00:00:02,440 --> 00:00:03,600
Speaker 1: All Zone Media.

2
00:00:06,320 --> 00:00:08,680
Speaker 2: Hello and welcome to Better Offline. I'm your host and

3
00:00:08,800 --> 00:00:24,439
Speaker 2: the single most punished man alive ed Zetron. Everyone hates me,

4
00:00:24,880 --> 00:00:27,680
Speaker 2: but the people don't hate the man who's joining me today.

5
00:00:27,720 --> 00:00:30,480
Speaker 2: I've got Paris Marks, the host of Tech Won't Save Us,

6
00:00:30,480 --> 00:00:33,280
Speaker 2: the writer of the Disconnect blog, and the actual host

7
00:00:33,280 --> 00:00:35,440
Speaker 2: of the brand new series is about to tell us

8
00:00:35,440 --> 00:00:37,920
Speaker 2: about data vampires. Paris, thank you for joining me.

9
00:00:38,600 --> 00:00:40,200
Speaker 3: Absolutely, it's so fun to be on the show.

10
00:00:40,960 --> 00:00:43,000
Speaker 2: So tell me about the new show. I'm excited.

11
00:00:43,720 --> 00:00:46,760
Speaker 3: Yeah, it's uh, you know, I think everyone has kind

12
00:00:46,800 --> 00:00:49,720
Speaker 3: of been paying attention a bit more to like the energy,

13
00:00:49,800 --> 00:00:52,000
Speaker 3: use of data center as a cryptocurrency and all this

14
00:00:52,080 --> 00:00:56,040
Speaker 3: kind of stuff, and you know, especially what powers generative AI,

15
00:00:56,480 --> 00:00:58,400
Speaker 3: and you know what's been kind of fueling the hype

16
00:00:58,400 --> 00:01:00,520
Speaker 3: for the past two years. And so in this series,

17
00:01:00,560 --> 00:01:03,040
Speaker 3: I wanted to go into why we're actually building so

18
00:01:03,080 --> 00:01:05,600
Speaker 3: many data centers, What is behind these things, how much

19
00:01:05,680 --> 00:01:08,400
Speaker 3: energy are they actually using, And on the one hand,

20
00:01:08,480 --> 00:01:11,560
Speaker 3: like the commercial impulses to do this, but also like

21
00:01:11,600 --> 00:01:14,640
Speaker 3: the kind of weird ideological reasons that these people want

22
00:01:14,680 --> 00:01:17,240
Speaker 3: to build these massive AI systems in the first place.

23
00:01:17,560 --> 00:01:20,360
Speaker 3: Relating back to all these like things about AGI and

24
00:01:20,480 --> 00:01:23,600
Speaker 3: you know, trying to make computers with human level intelligence

25
00:01:23,640 --> 00:01:25,200
Speaker 3: and all this kind of stuff where they eventually want

26
00:01:25,200 --> 00:01:28,039
Speaker 3: to merge their brains with machines. So it goes into

27
00:01:28,080 --> 00:01:29,800
Speaker 3: like the wide range of all these sorts of things

28
00:01:29,880 --> 00:01:31,920
Speaker 3: over over for episodes.

29
00:01:32,880 --> 00:01:35,880
Speaker 2: So talk to me about the scale of this. How

30
00:01:36,319 --> 00:01:39,000
Speaker 2: many of these data centers are they building? Like, give

31
00:01:39,040 --> 00:01:41,480
Speaker 2: the audience some idea because it from what I know,

32
00:01:41,520 --> 00:01:42,160
Speaker 2: it's not great.

33
00:01:43,120 --> 00:01:45,720
Speaker 3: Yeah, absolutely, and it's quite a lot, right, especially when

34
00:01:45,720 --> 00:01:47,960
Speaker 3: you see it scale up over the past number of years.

35
00:01:48,560 --> 00:01:51,880
Speaker 3: So we've seen this like really significant escalation over the

36
00:01:51,880 --> 00:01:54,680
Speaker 3: past five years, in particular in the construction of these

37
00:01:54,760 --> 00:01:57,360
Speaker 3: data centers. So there was something like five hundred or so,

38
00:01:58,880 --> 00:02:02,280
Speaker 3: you know, about five years ago, and now we're looking

39
00:02:02,360 --> 00:02:04,760
Speaker 3: at more like a thousand that were completed by the

40
00:02:04,880 --> 00:02:07,120
Speaker 3: end of or in the early part of twenty twenty four.

41
00:02:07,400 --> 00:02:09,080
Speaker 3: We have many more than that now, and we know

42
00:02:09,160 --> 00:02:12,919
Speaker 3: that Microsoft, Amazon, and Google in particular are making these

43
00:02:13,000 --> 00:02:16,079
Speaker 3: huge investments to increase the number of data centers that

44
00:02:16,080 --> 00:02:18,600
Speaker 3: they're building in to build them more quickly. And so

45
00:02:19,000 --> 00:02:21,200
Speaker 3: you know, the thing is some people might listen to

46
00:02:21,240 --> 00:02:23,440
Speaker 3: this and say Okay, Yeah, but data centers have been

47
00:02:23,480 --> 00:02:25,040
Speaker 3: around for a long time, and I think that this

48
00:02:25,120 --> 00:02:27,960
Speaker 3: is an important distinction to make, right. Data centers as

49
00:02:28,000 --> 00:02:30,280
Speaker 3: we knew them were like a floor in an office

50
00:02:30,280 --> 00:02:33,720
Speaker 3: building that a company was using for their own purposes,

51
00:02:34,000 --> 00:02:36,760
Speaker 3: or you know, maybe even a specific room in the

52
00:02:36,760 --> 00:02:39,760
Speaker 3: basement or something. But what we're talking about in particular

53
00:02:39,800 --> 00:02:42,400
Speaker 3: with Amazon, Bikersoft, and Google and this move to the

54
00:02:42,400 --> 00:02:45,600
Speaker 3: cloud and this kind of centralized computation is the hyper

55
00:02:45,600 --> 00:02:48,360
Speaker 3: scale data centers that exist on the scale that is

56
00:02:48,480 --> 00:02:51,040
Speaker 3: so much greater than what we used to have before

57
00:02:51,080 --> 00:02:54,280
Speaker 3: the cloud, and how we've really seen that escalate over

58
00:02:54,320 --> 00:02:56,520
Speaker 3: the past few years in particular. So that is the

59
00:02:56,520 --> 00:02:59,120
Speaker 3: specific problem, right. It's not that we have computation, and

60
00:02:59,160 --> 00:03:01,720
Speaker 3: it's not that there is some degree of centralized computation.

61
00:03:01,840 --> 00:03:04,520
Speaker 3: It's the scale that these infrastructures exist. That and then

62
00:03:04,560 --> 00:03:06,760
Speaker 3: the question of why we're actually building all these in

63
00:03:06,800 --> 00:03:07,440
Speaker 3: the first place.

64
00:03:08,080 --> 00:03:10,600
Speaker 2: So why are we building them? What? Like, what are

65
00:03:10,600 --> 00:03:11,840
Speaker 2: they actually putting in there?

66
00:03:13,160 --> 00:03:15,919
Speaker 3: Yeah, And it's an essential point, right, And I think

67
00:03:16,080 --> 00:03:17,799
Speaker 3: there are a few different ways to look at it, right,

68
00:03:18,200 --> 00:03:21,040
Speaker 3: Some computation is always going to be necessary for the

69
00:03:21,080 --> 00:03:23,359
Speaker 3: types of things that we're doing online. You know, there's

70
00:03:23,440 --> 00:03:26,560
Speaker 3: there's no question about that, right. I think it's fair

71
00:03:26,600 --> 00:03:28,880
Speaker 3: to say that we want some degree of Netflix, we

72
00:03:28,919 --> 00:03:31,080
Speaker 3: want to have easy access to our email, you know,

73
00:03:31,120 --> 00:03:33,200
Speaker 3: we want to be able to socialize with one another

74
00:03:33,760 --> 00:03:36,000
Speaker 3: on these various platforms that we use. And I think

75
00:03:36,040 --> 00:03:37,960
Speaker 3: that that is totally legitimate and that we should be

76
00:03:38,000 --> 00:03:40,000
Speaker 3: able to do those sorts of things, right. But then

77
00:03:40,280 --> 00:03:43,480
Speaker 3: you ask, okay, but what is really driving this significant

78
00:03:43,480 --> 00:03:45,840
Speaker 3: growth that we have been seeing over the past while.

79
00:03:46,280 --> 00:03:48,440
Speaker 3: And we see on the one hand, certainly that is

80
00:03:48,520 --> 00:03:51,880
Speaker 3: generative AI. The cryptocurrencies are a little bit separate from

81
00:03:51,920 --> 00:03:54,160
Speaker 3: that because generally that stuff is not being run on

82
00:03:54,240 --> 00:03:56,960
Speaker 3: say like the Amazon, Microsoft and Google servers. Those are

83
00:03:57,080 --> 00:04:00,720
Speaker 3: like specific things I don't even think if they let you, right,

84
00:04:00,840 --> 00:04:03,520
Speaker 3: So these are specific infrastructures being set up by like

85
00:04:03,560 --> 00:04:06,000
Speaker 3: crypto miners and things like that. So it's a little

86
00:04:06,000 --> 00:04:07,760
Speaker 3: bit different. But we can still kind of, you know,

87
00:04:07,840 --> 00:04:10,360
Speaker 3: look at the amount of energy use that's being put

88
00:04:10,400 --> 00:04:12,760
Speaker 3: into those sorts of use cases. But the other piece

89
00:04:12,800 --> 00:04:15,440
Speaker 3: that I think that we tend to you know, not

90
00:04:15,560 --> 00:04:17,560
Speaker 3: think about so much, especially when we talk so much

91
00:04:17,560 --> 00:04:21,680
Speaker 3: about generative AI, is this sort of you know foundation

92
00:04:21,960 --> 00:04:24,080
Speaker 3: that we have built the Internet on over the past

93
00:04:24,160 --> 00:04:26,320
Speaker 3: number of decades. And for a lot of these companies,

94
00:04:26,360 --> 00:04:29,640
Speaker 3: the business model is collect as much data as possible

95
00:04:29,920 --> 00:04:31,960
Speaker 3: so that then we can target ads, and we can

96
00:04:32,000 --> 00:04:34,960
Speaker 3: target product recommendations and all this sort of stuff to you,

97
00:04:35,800 --> 00:04:39,600
Speaker 3: And that like surveillance piece and storage of massive amounts

98
00:04:39,640 --> 00:04:41,320
Speaker 3: of data on all of us and everything that we

99
00:04:41,360 --> 00:04:44,120
Speaker 3: do online is also a big component of it as well.

100
00:04:44,360 --> 00:04:47,400
Speaker 3: That I think it's under considered. And so my provocation

101
00:04:47,520 --> 00:04:49,360
Speaker 3: is like, do we really need to be collecting all

102
00:04:49,360 --> 00:04:51,960
Speaker 3: that data in the first place? That requires so much

103
00:04:52,040 --> 00:04:55,560
Speaker 3: you know, computation but also storage, which is driving the

104
00:04:56,120 --> 00:04:58,520
Speaker 3: creation and the building of so many of these hyperscale

105
00:04:58,560 --> 00:04:59,160
Speaker 3: data centers.

106
00:05:00,000 --> 00:05:02,679
Speaker 2: Building them just for generative AI or as generateve AI

107
00:05:02,880 --> 00:05:04,320
Speaker 2: just kind of an excuse.

108
00:05:05,040 --> 00:05:07,960
Speaker 3: Yeah, I certainly see it as an excuse it, you know,

109
00:05:08,200 --> 00:05:10,200
Speaker 3: if we're really digging into it, it is a bit

110
00:05:10,240 --> 00:05:12,520
Speaker 3: of both, right. On the one hand, these companies are

111
00:05:12,560 --> 00:05:15,160
Speaker 3: trying to build a ton of data centers in order

112
00:05:15,200 --> 00:05:18,279
Speaker 3: to power generative AI, because we both know how computationally

113
00:05:18,360 --> 00:05:21,039
Speaker 3: intensive not just the training of those models is, but

114
00:05:21,120 --> 00:05:23,960
Speaker 3: also then using those products. As these companies are trying

115
00:05:23,960 --> 00:05:26,640
Speaker 3: to roll them out into so much of what we do,

116
00:05:26,680 --> 00:05:28,200
Speaker 3: so many of the services that we use.

117
00:05:28,279 --> 00:05:28,479
Speaker 2: Right.

118
00:05:28,680 --> 00:05:30,320
Speaker 3: But then the other piece of it is that even

119
00:05:30,360 --> 00:05:33,000
Speaker 3: before the generative AI moment like I was talking about,

120
00:05:33,040 --> 00:05:34,919
Speaker 3: over the past five years, we can already see that

121
00:05:34,960 --> 00:05:37,279
Speaker 3: there was this scale up in the number of hyper

122
00:05:37,320 --> 00:05:41,320
Speaker 3: scale data centers being built in particular, and so in

123
00:05:41,360 --> 00:05:44,560
Speaker 3: that way, I see generative AI as an excuse to

124
00:05:44,680 --> 00:05:48,159
Speaker 3: continue building these things that they were already building in

125
00:05:48,160 --> 00:05:50,880
Speaker 3: the first place. And this is like the foundational point here, right,

126
00:05:51,000 --> 00:05:53,919
Speaker 3: is that if you think about a company like Amazon,

127
00:05:53,960 --> 00:05:57,320
Speaker 3: Microsoft and Google that has this massive cloud business, and

128
00:05:57,520 --> 00:06:00,839
Speaker 3: especially Microsoft and Amazon where they're getting so many you know,

129
00:06:00,920 --> 00:06:02,839
Speaker 3: you know where the profits from those businesses are so

130
00:06:02,920 --> 00:06:06,280
Speaker 3: important to their to their wider business model and being

131
00:06:06,279 --> 00:06:09,640
Speaker 3: able to expand into so many different areas of business.

132
00:06:10,640 --> 00:06:13,200
Speaker 3: You know, there is an inherent incentive then, and you know,

133
00:06:13,240 --> 00:06:16,080
Speaker 3: we know how capitalism works. These businesses always need to grow.

134
00:06:16,160 --> 00:06:17,920
Speaker 3: They need to be making more profits in order to

135
00:06:17,960 --> 00:06:21,360
Speaker 3: keep shareholders happy. So if your business is, you know,

136
00:06:21,480 --> 00:06:25,400
Speaker 3: providing centralized computation at scale, you need the amount of

137
00:06:25,440 --> 00:06:29,159
Speaker 3: computation that we all collectively use to continue growing year

138
00:06:29,200 --> 00:06:31,080
Speaker 3: on year, and you know they want it to be

139
00:06:31,080 --> 00:06:33,200
Speaker 3: growing very quickly, and that means you're not just going

140
00:06:33,240 --> 00:06:35,600
Speaker 3: to need more hyper scale data centers, but you need

141
00:06:35,640 --> 00:06:38,960
Speaker 3: to sell people on more computation and to make the

142
00:06:39,000 --> 00:06:43,200
Speaker 3: things that we use more computationally intensive to justify you know,

143
00:06:43,279 --> 00:06:45,599
Speaker 3: this this kind of business incentive that that's driving you.

144
00:06:46,279 --> 00:06:49,200
Speaker 2: So it almost kind of sounds like they have just

145
00:06:49,240 --> 00:06:53,320
Speaker 2: been trying to find computationally expensive or intensive even things

146
00:06:53,440 --> 00:06:55,960
Speaker 2: so that they could build more ways to compute and

147
00:06:56,160 --> 00:06:59,840
Speaker 2: justify these expensive It's kind of sickening, pisses yea off.

148
00:07:00,400 --> 00:07:03,760
Speaker 2: It's like when I think about that, people say to me, Oh,

149
00:07:03,800 --> 00:07:06,279
Speaker 2: you're angry. Why are I don't know why other people

150
00:07:06,320 --> 00:07:08,840
Speaker 2: aren't angry. I feel like you and I are about

151
00:07:08,839 --> 00:07:10,560
Speaker 2: as angry. We're very about this.

152
00:07:11,040 --> 00:07:12,560
Speaker 3: Maybe we express it in different ways.

153
00:07:12,560 --> 00:07:16,000
Speaker 2: Sometimes perhaps agree, yeah, you're a little nicer than I am.

154
00:07:16,080 --> 00:07:20,680
Speaker 2: Perhaps it's just frustrating because I was just on a

155
00:07:20,680 --> 00:07:23,000
Speaker 2: podcast actually just before this, and I was talking to

156
00:07:23,000 --> 00:07:25,000
Speaker 2: them and they were talking about the promise of AI.

157
00:07:26,040 --> 00:07:28,200
Speaker 2: But when you get past the promise, it's really just

158
00:07:28,320 --> 00:07:31,760
Speaker 2: kind of shit, like it's mediaoka, it's and they're building

159
00:07:31,960 --> 00:07:35,840
Speaker 2: all of this infrastructure for nothing, And one theory I

160
00:07:35,880 --> 00:07:38,520
Speaker 2: have is that it's not being used. I don't know

161
00:07:38,560 --> 00:07:41,000
Speaker 2: if you've seen anything that suggests that or what I mean,

162
00:07:41,040 --> 00:07:44,320
Speaker 2: you've likely working off of public documents, I'm guessing. But

163
00:07:45,160 --> 00:07:47,120
Speaker 2: what if this is for nothing? What if they've just

164
00:07:47,120 --> 00:07:50,160
Speaker 2: built a bunch of data centers for no reason.

165
00:07:51,560 --> 00:07:53,720
Speaker 3: Yeah, And that is something that I've talked to a

166
00:07:53,800 --> 00:07:56,600
Speaker 3: number of people about, right, because I'm trying to understand

167
00:07:56,600 --> 00:07:58,760
Speaker 3: that question as well. And I think that there's two

168
00:07:58,840 --> 00:08:02,080
Speaker 3: potential past that this takes. So one of them is

169
00:08:02,080 --> 00:08:04,160
Speaker 3: that maybe they don't end up using some of these

170
00:08:04,240 --> 00:08:06,640
Speaker 3: data centers that they're currently building, or they scale back

171
00:08:06,720 --> 00:08:09,320
Speaker 3: some of the projects that they're currently planning to build.

172
00:08:09,400 --> 00:08:09,600
Speaker 2: Right.

173
00:08:09,880 --> 00:08:12,680
Speaker 3: And so, you know, my kind of example to point

174
00:08:12,680 --> 00:08:16,560
Speaker 3: to for that being a potential pathway is you know,

175
00:08:16,880 --> 00:08:20,800
Speaker 3: remember in the early kind of probably year or eighteen

176
00:08:20,840 --> 00:08:24,200
Speaker 3: months of the pandemic, there was a lot of people

177
00:08:24,240 --> 00:08:26,720
Speaker 3: who were using like Amazon to get more things delivered.

178
00:08:26,760 --> 00:08:29,800
Speaker 3: And as a result, Amazon kind of created this plan

179
00:08:29,880 --> 00:08:32,400
Speaker 3: to build a ton more warehouses than they were actually

180
00:08:32,400 --> 00:08:35,000
Speaker 3: going to need. And as like you know, people went

181
00:08:35,040 --> 00:08:38,120
Speaker 3: back to kind of shopping a bit more normally. As

182
00:08:38,200 --> 00:08:41,640
Speaker 3: you know, the threat or as the lockdowns ended and

183
00:08:41,920 --> 00:08:44,520
Speaker 3: things like that, what you saw is the demand for

184
00:08:44,600 --> 00:08:48,040
Speaker 3: Amazon or the demand that Amazon had projected decreased, and

185
00:08:48,080 --> 00:08:51,319
Speaker 3: so they canceled a bunch of warehouse projects and expansions

186
00:08:51,840 --> 00:08:54,880
Speaker 3: across the United States and their wider warehouse network because

187
00:08:54,880 --> 00:08:56,800
Speaker 3: they felt that they weren't actually going to need all

188
00:08:56,880 --> 00:08:58,880
Speaker 3: that capacity that they thought they were going to. So

189
00:08:59,320 --> 00:09:02,280
Speaker 3: that's one potent path. But I think, you know, because

190
00:09:02,320 --> 00:09:04,040
Speaker 3: of what we were saying, and because of what was

191
00:09:04,280 --> 00:09:06,800
Speaker 3: what I was explaining about, you know, this need to

192
00:09:07,600 --> 00:09:10,480
Speaker 3: grow the amount of computation that we are like collectively using,

193
00:09:11,120 --> 00:09:14,200
Speaker 3: I think that it is less the chance with data

194
00:09:14,200 --> 00:09:17,079
Speaker 3: centers that that is the pathway that we see, and instead,

195
00:09:17,120 --> 00:09:19,800
Speaker 3: I think that the amount the number of data centers

196
00:09:19,840 --> 00:09:23,559
Speaker 3: that these companies are creating, if generative a AI falls off,

197
00:09:23,600 --> 00:09:25,560
Speaker 3: which I think we you know, we both think is

198
00:09:25,640 --> 00:09:28,000
Speaker 3: going to happen ultimately, that there's going to be this

199
00:09:28,000 --> 00:09:31,760
Speaker 3: this collapse, this crash, I think they will ultimately try

200
00:09:31,800 --> 00:09:35,880
Speaker 3: to find some other use case to justify building out

201
00:09:35,880 --> 00:09:38,520
Speaker 3: that computation because I think that that is like the

202
00:09:38,520 --> 00:09:41,520
Speaker 3: inherent project that they want to realize. On the one hand,

203
00:09:41,559 --> 00:09:43,720
Speaker 3: on the commercial side, but also because the people who

204
00:09:43,760 --> 00:09:46,960
Speaker 3: are running these companies really believe that everything that we

205
00:09:47,000 --> 00:09:49,760
Speaker 3: do in our society needs to have computation like inserted

206
00:09:49,800 --> 00:09:52,280
Speaker 3: in there somehow, and that this is the way that

207
00:09:52,320 --> 00:09:53,560
Speaker 3: we build like the future right.

208
00:09:54,280 --> 00:09:57,360
Speaker 2: Kind of makes me think of Peloton. No sure, if

209
00:09:57,400 --> 00:09:59,599
Speaker 2: you remember the twenty twenty one stories of Peloton was

210
00:09:59,640 --> 00:10:03,680
Speaker 2: like pos the next trillion dollar company. Everyone was furiously

211
00:10:03,720 --> 00:10:07,160
Speaker 2: excited about Peloton, and then the moment people got allowed

212
00:10:07,200 --> 00:10:10,240
Speaker 2: outside again, it started to kind of fall apart. Well, okay,

213
00:10:10,280 --> 00:10:12,040
Speaker 2: I'm getting my dates wrong because twenty twenty one was

214
00:10:12,040 --> 00:10:14,679
Speaker 2: when it started to fall off. But nevertheless, I have

215
00:10:14,720 --> 00:10:17,640
Speaker 2: to wonder if that could happen here because while you

216
00:10:17,720 --> 00:10:20,640
Speaker 2: say they'll come up with a reason, what is it

217
00:10:20,960 --> 00:10:23,160
Speaker 2: and how does that lead to money? I just add

218
00:10:23,240 --> 00:10:25,440
Speaker 2: Zephyr teach out on and we were talking about this.

219
00:10:25,559 --> 00:10:28,760
Speaker 2: It's like, what if there is no plan? What if

220
00:10:28,760 --> 00:10:32,120
Speaker 2: they don't other than bigger? Because it's based on what

221
00:10:32,240 --> 00:10:36,080
Speaker 2: I've what the series you're doing is about. It seems

222
00:10:36,120 --> 00:10:38,360
Speaker 2: like they're committed to this at least, like this is

223
00:10:38,400 --> 00:10:40,880
Speaker 2: they need to do this to prove that they are cool,

224
00:10:41,200 --> 00:10:45,000
Speaker 2: rather than researching or developing things, it almost feels as

225
00:10:45,040 --> 00:10:46,560
Speaker 2: if they think this is inevitable.

226
00:10:47,480 --> 00:10:49,920
Speaker 3: That's what I feel, right, And part of that comes

227
00:10:49,920 --> 00:10:52,640
Speaker 3: not just from the commercial imperative, but also like these

228
00:10:52,640 --> 00:10:55,240
Speaker 3: broader ideological things that I'm sure that you've discussed on

229
00:10:55,280 --> 00:10:57,520
Speaker 3: your show and that I've been discussing with people as well.

230
00:10:57,559 --> 00:11:00,400
Speaker 3: But when you hear people like say Sam Altman or

231
00:11:00,520 --> 00:11:03,840
Speaker 3: Eric Schmidt or you know, Elon Musk talk about how

232
00:11:03,880 --> 00:11:07,040
Speaker 3: ais are going to become so intelligent, and you know

233
00:11:07,080 --> 00:11:09,559
Speaker 3: Sam Altman specifically saying that he wants to make sure

234
00:11:09,600 --> 00:11:12,040
Speaker 3: that we like I'll merge our brains with machines and

235
00:11:12,320 --> 00:11:15,080
Speaker 3: all these sorts of things, Like I think that suggests

236
00:11:15,120 --> 00:11:17,719
Speaker 3: also like the direction that these people want to go,

237
00:11:18,200 --> 00:11:21,480
Speaker 3: not because you know, not just because of commercial reasons,

238
00:11:21,480 --> 00:11:23,440
Speaker 3: not just because they think they're going to make you know,

239
00:11:23,480 --> 00:11:26,000
Speaker 3: more profits or try to realize some kind of world

240
00:11:26,000 --> 00:11:27,719
Speaker 3: that they want to realize this way, but because they

241
00:11:27,720 --> 00:11:30,160
Speaker 3: think that if this is the direction that we don't

242
00:11:30,200 --> 00:11:34,760
Speaker 3: take as a society, then you know, the future that

243
00:11:34,800 --> 00:11:36,880
Speaker 3: they think is the one that we need to realize

244
00:11:36,920 --> 00:11:39,960
Speaker 3: in order to have this you know, wonderful, glorious future

245
00:11:40,000 --> 00:11:43,200
Speaker 3: where humanity exists forever and we colonize all these planets

246
00:11:43,240 --> 00:11:44,959
Speaker 3: and all this kind of stuff. If we don't take

247
00:11:45,000 --> 00:11:47,920
Speaker 3: this direction, if we don't develop these incredibly powerful ais,

248
00:11:47,960 --> 00:11:50,080
Speaker 3: then in their mind, this isn't going to happen. So

249
00:11:50,120 --> 00:11:53,320
Speaker 3: it feels like, yes, there's the commercial element of this, right,

250
00:11:53,400 --> 00:11:56,080
Speaker 3: They're they're trying to drive investment with this AI boom

251
00:11:56,120 --> 00:11:59,600
Speaker 3: by promising so much about it. But like ideologically, in

252
00:11:59,640 --> 00:12:01,840
Speaker 3: their in their kind of minds and world views, they

253
00:12:01,840 --> 00:12:03,400
Speaker 3: think that this is the direction that we have to

254
00:12:03,400 --> 00:12:06,040
Speaker 3: go or we're basically all doomed, even though in the

255
00:12:06,080 --> 00:12:08,520
Speaker 3: process they're kind of dooming us by destroying.

256
00:12:08,120 --> 00:12:10,880
Speaker 2: At the time, destroying the environment to make sure we

257
00:12:10,920 --> 00:12:25,040
Speaker 2: don't do ourselves. That's the Eric Schmidt thing. Yeah, so

258
00:12:25,840 --> 00:12:28,400
Speaker 2: let's talk costs. Yeah, what is being put into this?

259
00:12:28,440 --> 00:12:30,360
Speaker 2: Because I have bullpop numbers, but maybe you have some

260
00:12:30,440 --> 00:12:32,559
Speaker 2: better ideas. How much is actually going into this?

261
00:12:33,760 --> 00:12:38,480
Speaker 3: Yeah, so it's hundreds of hundreds of billions, let me

262
00:12:38,520 --> 00:12:42,160
Speaker 3: get that, right, if not trillions of dollars? Right? And

263
00:12:42,760 --> 00:12:44,800
Speaker 3: I can't remember the exact figures off the top of

264
00:12:44,800 --> 00:12:48,440
Speaker 3: my head, but over the past year, say from June

265
00:12:48,480 --> 00:12:50,640
Speaker 3: of twenty twenty three to July of twenty twenty four.

266
00:12:50,720 --> 00:12:53,440
Speaker 3: Microsoft has been plowing hundreds of billions of dollars into

267
00:12:53,520 --> 00:12:57,480
Speaker 3: data center expansions around the world. Amazon committed I believe

268
00:12:57,520 --> 00:13:00,439
Speaker 3: it was one point five trillion dollars early this year

269
00:13:00,480 --> 00:13:03,440
Speaker 3: or last year to data center expansion, with about five

270
00:13:03,600 --> 00:13:05,960
Speaker 3: hundred billion of that going out in the short term.

271
00:13:06,559 --> 00:13:09,120
Speaker 3: So these are like massive numbers. We know that Microsoft

272
00:13:09,200 --> 00:13:12,040
Speaker 3: and open Ai have talked about building a one hundred

273
00:13:12,120 --> 00:13:15,960
Speaker 3: billion dollar data center complex that would need nuclear energy

274
00:13:16,000 --> 00:13:19,960
Speaker 3: in order to power it because they be so energy intensive. Yeah, right, exactly.

275
00:13:20,840 --> 00:13:23,040
Speaker 3: So like these are the kinds of numbers that we're

276
00:13:23,080 --> 00:13:25,600
Speaker 3: talking about, and I feel like these are the types

277
00:13:25,600 --> 00:13:28,319
Speaker 3: of numbers that we often hear, like, you know, governments

278
00:13:28,360 --> 00:13:30,920
Speaker 3: talk about when we're when they're making like massive investment

279
00:13:30,960 --> 00:13:34,559
Speaker 3: projects or something, right, not like private companies. But because

280
00:13:34,559 --> 00:13:37,440
Speaker 3: these private companies are so huge, it's like they can

281
00:13:37,480 --> 00:13:40,000
Speaker 3: build out these infrastructures that have, you know, such a

282
00:13:40,080 --> 00:13:43,000
Speaker 3: huge impact that they have impacts not just in communities,

283
00:13:43,040 --> 00:13:46,400
Speaker 3: but like nationwide globally because they are so significant.

284
00:13:47,080 --> 00:13:49,800
Speaker 2: Almost feels like a kind of authoritarianism. It feels like

285
00:13:49,840 --> 00:13:52,600
Speaker 2: they're building out a new governmental system that they control

286
00:13:52,920 --> 00:13:56,040
Speaker 2: that they meeter out which includes I guess it's a

287
00:13:56,240 --> 00:13:58,800
Speaker 2: vampiric element to it as well, where they control the

288
00:13:58,800 --> 00:14:03,440
Speaker 2: power of America, even though power is theoretically a civic good,

289
00:14:04,080 --> 00:14:05,280
Speaker 2: like a civic cutility.

290
00:14:06,520 --> 00:14:08,600
Speaker 3: Well, it definitely feels that way, especially when you look

291
00:14:08,640 --> 00:14:11,680
Speaker 3: at like the politics that they're embracing too, right where

292
00:14:11,679 --> 00:14:14,400
Speaker 3: you have so many of these influential people in Silicon

293
00:14:14,480 --> 00:14:18,000
Speaker 3: Valley not just embracing like Donald Trump, but embracing this

294
00:14:18,080 --> 00:14:20,880
Speaker 3: form of far right politics that is, you know, gaining

295
00:14:20,920 --> 00:14:23,000
Speaker 3: steam in a number of countries around the world that

296
00:14:23,160 --> 00:14:27,320
Speaker 3: is explicitly anti democratic. You know, Elon Musk is talking

297
00:14:27,360 --> 00:14:29,480
Speaker 3: a lot a lot about that these days and saying

298
00:14:29,520 --> 00:14:33,000
Speaker 3: that he thinks the Democrats are going to end democracy. Meanwhile,

299
00:14:33,000 --> 00:14:35,880
Speaker 3: like Donald Trump is saying quite explicitly that you know,

300
00:14:35,920 --> 00:14:40,000
Speaker 3: he wants to impinge on democratic rights. And these people

301
00:14:40,040 --> 00:14:44,200
Speaker 3: have no problem like working with anti democratic leaders around

302
00:14:44,240 --> 00:14:46,320
Speaker 3: the world as long as they see it as the

303
00:14:46,360 --> 00:14:50,240
Speaker 3: way to like advance not only their kind of personal

304
00:14:50,280 --> 00:14:52,800
Speaker 3: interests and to make sure their companies do well, but

305
00:14:52,840 --> 00:14:55,040
Speaker 3: also again to try to realize this like type of

306
00:14:55,080 --> 00:14:58,800
Speaker 3: future that they think is desirable. I was just recently

307
00:14:58,800 --> 00:15:01,400
Speaker 3: talking to Julia Black, who's a reporter at the Information

308
00:15:01,440 --> 00:15:03,680
Speaker 3: about this, and she did a profile on Curtis Jarvin,

309
00:15:03,760 --> 00:15:06,520
Speaker 3: who is like, you know this guy who yeah, this

310
00:15:06,560 --> 00:15:08,920
Speaker 3: guy who like calls himself a dark Elf, but you know,

311
00:15:08,960 --> 00:15:10,840
Speaker 3: as one of the co founders of the Dark Enlightenment.

312
00:15:10,880 --> 00:15:14,600
Speaker 3: This like you know, this this like political movement that

313
00:15:14,680 --> 00:15:18,960
Speaker 3: wants to see a monarchy established in the United States,

314
00:15:19,040 --> 00:15:21,400
Speaker 3: that structured more like you know, the CEO of a

315
00:15:21,440 --> 00:15:25,160
Speaker 3: corporation to have you know, the United States ron as

316
00:15:25,200 --> 00:15:27,400
Speaker 3: an authoritarian state. And we know that Peter Teal and

317
00:15:27,400 --> 00:15:29,600
Speaker 3: Mark Andriesen and a ton of these guys like support

318
00:15:29,600 --> 00:15:32,080
Speaker 3: these ideas, so you know, like you can you can

319
00:15:32,080 --> 00:15:34,120
Speaker 3: see that with the data centers, you can see this

320
00:15:34,200 --> 00:15:38,360
Speaker 3: like concerning power grab as these infrastructures expand and they're

321
00:15:38,360 --> 00:15:41,480
Speaker 3: in the hands of these major tech companies. But even

322
00:15:41,520 --> 00:15:43,800
Speaker 3: when you look at like the leaders of these companies themselves,

323
00:15:43,800 --> 00:15:45,680
Speaker 3: you can see that they very much want to make

324
00:15:45,720 --> 00:15:48,000
Speaker 3: sure that, you know, the power of the people through

325
00:15:48,040 --> 00:15:50,960
Speaker 3: democracy is curtailed so that they can do whatever they want.

326
00:15:51,480 --> 00:15:53,520
Speaker 2: And if you want to know more about Curtis Yarvin,

327
00:15:53,840 --> 00:15:56,840
Speaker 2: somehow behind the bus has had ed helms On to

328
00:15:56,960 --> 00:16:01,880
Speaker 2: talk about Curtis Yarvin. Want to hear comedian Ed Helms

329
00:16:01,880 --> 00:16:07,200
Speaker 2: talk about one of the worst people alive, like, definitely

330
00:16:08,400 --> 00:16:11,920
Speaker 2: a bottle popper. When he pops his clogs, there will

331
00:16:11,960 --> 00:16:15,720
Speaker 2: all be celebrating. So has there been any pushback against

332
00:16:15,720 --> 00:16:19,920
Speaker 2: these data centers you've seen, either from governments or people. Yeah.

333
00:16:19,960 --> 00:16:22,800
Speaker 3: Absolutely, And that's part of the reason that I wanted

334
00:16:22,800 --> 00:16:26,160
Speaker 3: to make the series and also like why it came

335
00:16:26,280 --> 00:16:28,600
Speaker 3: much more on my radar because I started hearing these

336
00:16:28,640 --> 00:16:32,680
Speaker 3: stories of pushback happening really around the world, right. And

337
00:16:32,720 --> 00:16:35,520
Speaker 3: it started with just hearing like the story of the

338
00:16:35,600 --> 00:16:38,480
Speaker 3: dolls in Oregon, where these people were trying to find

339
00:16:38,520 --> 00:16:41,240
Speaker 3: out the amount of water that Google's data centers were using.

340
00:16:41,520 --> 00:16:44,640
Speaker 3: And then Google or sorry, then you know, there was

341
00:16:44,680 --> 00:16:47,800
Speaker 3: actually a lawsuit because Google wouldn't share that information and

342
00:16:47,840 --> 00:16:49,360
Speaker 3: it took a year for them to share it with

343
00:16:49,440 --> 00:16:52,200
Speaker 3: the Oregonian, which is the local newspaper, and they found

344
00:16:52,200 --> 00:16:55,360
Speaker 3: that over the past few years, Google's water use, I

345
00:16:55,400 --> 00:16:57,200
Speaker 3: think it was over the past five years, Google's water

346
00:16:57,280 --> 00:17:00,880
Speaker 3: use had tripled, you know, just in that city alone. Yeah,

347
00:17:00,960 --> 00:17:03,640
Speaker 3: so these are really significant developments, right, But then I

348
00:17:03,640 --> 00:17:06,399
Speaker 3: started to hear about how, say in Phoenix or northern

349
00:17:06,440 --> 00:17:08,760
Speaker 3: Virginia or many other parts of the United States, there

350
00:17:08,760 --> 00:17:11,040
Speaker 3: were also growing concerns about data centers. And then I

351
00:17:11,119 --> 00:17:13,760
Speaker 3: started to learn about in Ireland how now twenty one

352
00:17:13,800 --> 00:17:16,560
Speaker 3: percent of all electricity goes to data centers, and that's

353
00:17:16,600 --> 00:17:19,200
Speaker 3: causing concerns about the ability of the power grid to

354
00:17:19,320 --> 00:17:22,679
Speaker 3: keep up, you know, but also about whether you know

355
00:17:22,720 --> 00:17:24,879
Speaker 3: it makes sense to be allowing this many data centers

356
00:17:24,880 --> 00:17:27,040
Speaker 3: to be built. You know, concerns in Spain about the

357
00:17:27,040 --> 00:17:30,600
Speaker 3: amount of water usage in areas that are increasingly facing droughts.

358
00:17:31,160 --> 00:17:34,120
Speaker 3: You know, concerns in France and the Netherlands, but also

359
00:17:34,600 --> 00:17:37,679
Speaker 3: in parts of Asia, in South America, where communities in

360
00:17:37,760 --> 00:17:40,320
Speaker 3: Chile were saying like, if you build this massive Google

361
00:17:40,600 --> 00:17:43,680
Speaker 3: data center, will we still have running water into our homes,

362
00:17:43,720 --> 00:17:46,359
Speaker 3: and Google not being able to say like, yeah, definitely,

363
00:17:46,440 --> 00:17:48,000
Speaker 3: we can make sure that will happen, and so then

364
00:17:48,040 --> 00:17:50,880
Speaker 3: they campaign to try to stop it. Like around the world,

365
00:17:50,960 --> 00:17:54,240
Speaker 3: you're seeing the opposition to these things grow, and I

366
00:17:54,240 --> 00:17:56,120
Speaker 3: would say part of the reason for that, going back

367
00:17:56,119 --> 00:17:58,520
Speaker 3: to what I was saying earlier, is that, you know,

368
00:17:58,520 --> 00:18:00,960
Speaker 3: it's not just the scale of these structures, but it's

369
00:18:00,960 --> 00:18:03,520
Speaker 3: because these companies are trying to build out so many

370
00:18:03,560 --> 00:18:05,880
Speaker 3: more of them around the world, and when they get

371
00:18:05,920 --> 00:18:09,440
Speaker 3: located in these communities, once one is established, they usually

372
00:18:09,520 --> 00:18:11,720
Speaker 3: try to build more around it to cluster them, right,

373
00:18:11,800 --> 00:18:15,160
Speaker 3: because they're usually beneficial reasons why they've located one there

374
00:18:15,160 --> 00:18:17,760
Speaker 3: in the first place. And so then you have these

375
00:18:17,880 --> 00:18:21,879
Speaker 3: these increasingly you know, greater demands being placed on the

376
00:18:21,880 --> 00:18:25,040
Speaker 3: power grids and the water infrastructure of these communities, and

377
00:18:25,119 --> 00:18:27,440
Speaker 3: eventually it reaches the point where where many of them

378
00:18:27,560 --> 00:18:29,919
Speaker 3: start to break and start to say, like, wait, this

379
00:18:30,000 --> 00:18:33,080
Speaker 3: doesn't make sense anymore. You're threatening our own access to

380
00:18:33,200 --> 00:18:37,520
Speaker 3: electricity or to water or whatever. And so, you know,

381
00:18:37,680 --> 00:18:39,920
Speaker 3: that is why I, you know, I've been paying so

382
00:18:40,000 --> 00:18:42,240
Speaker 3: much more attention to this issue because I think it's

383
00:18:42,240 --> 00:18:45,000
Speaker 3: something that we're all going to be hearing much more about.

384
00:18:45,240 --> 00:18:47,760
Speaker 3: And I think that the earlier that we do start

385
00:18:47,800 --> 00:18:49,880
Speaker 3: to take notice of these problems and start to push

386
00:18:49,920 --> 00:18:53,119
Speaker 3: back on them, you know, the hopefully the earlier we

387
00:18:53,119 --> 00:18:55,400
Speaker 3: can start to curtail these effects and make sure that

388
00:18:56,320 --> 00:18:58,200
Speaker 3: you know, these companies can't just get away with doing

389
00:18:58,240 --> 00:18:58,880
Speaker 3: whatever they want.

390
00:18:59,440 --> 00:19:03,119
Speaker 2: Have we seen any situations where they have actually stopped

391
00:19:03,160 --> 00:19:04,760
Speaker 2: people accessing power or water?

392
00:19:06,600 --> 00:19:09,639
Speaker 3: Yeah, So in a few places we have seen like

393
00:19:09,720 --> 00:19:12,480
Speaker 3: the grids actually be threatened. So for example, in Ireland

394
00:19:12,480 --> 00:19:15,520
Speaker 3: now the grid operator in the winter often has to

395
00:19:15,560 --> 00:19:18,119
Speaker 3: issue these amber alerts to basically say to people like,

396
00:19:18,920 --> 00:19:23,280
Speaker 3: reduce your energy consumption or we're going to start doing

397
00:19:23,600 --> 00:19:27,960
Speaker 3: rolling blackouts. There were concerns. There's a story that Karen

398
00:19:28,000 --> 00:19:31,119
Speaker 3: Howe wrote in The Atlantic about what was going on

399
00:19:31,160 --> 00:19:34,560
Speaker 3: in Phoenix, and one of the biggest concerns there is,

400
00:19:34,600 --> 00:19:38,880
Speaker 3: of course, is you know, Phoenix's has a lot of desert. Yeah,

401
00:19:39,000 --> 00:19:42,600
Speaker 3: Phoenix experiences drought a lot. And so as more and

402
00:19:42,600 --> 00:19:45,040
Speaker 3: more data centers are being built there because the energy

403
00:19:45,080 --> 00:19:47,359
Speaker 3: it tends to be cheaper and tends to be have

404
00:19:47,400 --> 00:19:50,040
Speaker 3: a higher kind of percentage of renewables in there is

405
00:19:50,080 --> 00:19:54,120
Speaker 3: my understanding that they're going there regardless of the water impacts,

406
00:19:54,119 --> 00:19:57,399
Speaker 3: and so they're growing concerns about people's ability to you know,

407
00:19:58,160 --> 00:20:00,840
Speaker 3: access water when these data centers need much despite the

408
00:20:00,920 --> 00:20:03,760
Speaker 3: drought conditions that so many people are facing. And how

409
00:20:03,800 --> 00:20:06,480
Speaker 3: we know that you know, governments tend to turn off

410
00:20:06,520 --> 00:20:09,919
Speaker 3: water to the populations before the businesses that that just

411
00:20:10,160 --> 00:20:11,040
Speaker 3: on these sorts of things.

412
00:20:11,359 --> 00:20:17,520
Speaker 2: Yeah, it's truly awful like that It's why every time

413
00:20:17,520 --> 00:20:19,159
Speaker 2: I think of stuff like that, I start getting a

414
00:20:19,200 --> 00:20:21,439
Speaker 2: bit crazy, because it's just like you'd think that they

415
00:20:21,440 --> 00:20:24,639
Speaker 2: would say, well, actually, what we want to do is

416
00:20:24,720 --> 00:20:26,879
Speaker 2: turn it off for the business. No, no, no, we

417
00:20:26,960 --> 00:20:30,240
Speaker 2: must make sure the businesses can keep generating AI garfields.

418
00:20:30,280 --> 00:20:34,200
Speaker 2: If we don't have garfields with AK forty seven, well

419
00:20:34,600 --> 00:20:36,520
Speaker 2: what use is being able to drink water?

420
00:20:37,840 --> 00:20:40,359
Speaker 3: Exactly. One of the wild things too, that I was

421
00:20:40,440 --> 00:20:44,879
Speaker 3: learning about in Ireland in particular, is that so you know,

422
00:20:44,960 --> 00:20:46,720
Speaker 3: like I was saying, there's twenty one percent of all

423
00:20:46,720 --> 00:20:49,520
Speaker 3: the electricity now in that country is going to data centers.

424
00:20:49,520 --> 00:20:52,480
Speaker 3: It's projected to be about a third by twenty thirty.

425
00:20:53,280 --> 00:20:56,639
Speaker 3: And so for a while, the government there, or at

426
00:20:56,720 --> 00:20:59,080
Speaker 3: least in certain jurisdictions of the country, kind of stopped

427
00:20:59,080 --> 00:21:02,880
Speaker 3: connecting data centers to the grid, stop making new grid connections,

428
00:21:02,960 --> 00:21:05,480
Speaker 3: because you know, they were like, there's not enough power

429
00:21:05,520 --> 00:21:08,240
Speaker 3: to supply these infrastructures. And so what these data centers

430
00:21:08,240 --> 00:21:13,480
Speaker 3: started doing instead was to build local methane gas generation

431
00:21:13,640 --> 00:21:17,360
Speaker 3: facilities to bring in the gas themselves to the empower

432
00:21:17,440 --> 00:21:20,280
Speaker 3: the data center because they couldn't connect directly to the grid.

433
00:21:20,560 --> 00:21:22,480
Speaker 3: And then of course you you know, you can imagine

434
00:21:22,480 --> 00:21:25,520
Speaker 3: all the additional emissions that that creates with this kind

435
00:21:25,560 --> 00:21:29,199
Speaker 3: of like local gas generation infrastructure, and that's one of

436
00:21:29,200 --> 00:21:33,760
Speaker 3: the things that's contributing to Ireland, you know, not you know,

437
00:21:33,800 --> 00:21:36,000
Speaker 3: being able to meet its climate targets, or that not

438
00:21:36,040 --> 00:21:39,000
Speaker 3: really being in reach. One of the local tds or

439
00:21:39,000 --> 00:21:41,800
Speaker 3: members of Parliament who I spoke to there basically told

440
00:21:41,800 --> 00:21:44,679
Speaker 3: me that, like, there is renewable energy being added to

441
00:21:44,680 --> 00:21:46,840
Speaker 3: the grid in Ireland, but the problem that we face

442
00:21:46,920 --> 00:21:49,720
Speaker 3: is that because the energy demand is increasing so rapidly

443
00:21:50,119 --> 00:21:52,399
Speaker 3: that you know, the renewables just go to powering the

444
00:21:52,480 --> 00:21:54,800
Speaker 3: data centers and we can't turn the fossil fuels off.

445
00:21:54,960 --> 00:21:56,680
Speaker 3: And we're seeing a ton of stories about that in

446
00:21:56,720 --> 00:22:00,000
Speaker 3: the United States as well, where yes, renewables are being

447
00:22:00,040 --> 00:22:02,040
Speaker 3: added to the grid, but the fossil fuels are not

448
00:22:02,080 --> 00:22:04,360
Speaker 3: being turned off next to it, and in some cases

449
00:22:04,680 --> 00:22:08,359
Speaker 3: fossil fuel infrastructure is actually being reactivated. Or there was

450
00:22:08,400 --> 00:22:10,359
Speaker 3: a story and I can't remember what was Bloomberg or

451
00:22:10,400 --> 00:22:13,320
Speaker 3: the Financial Times recently that basically said the United States

452
00:22:13,359 --> 00:22:16,480
Speaker 3: is investing in new fossil fuel infrastructure at the fastest

453
00:22:16,560 --> 00:22:18,400
Speaker 3: rate it has in years, which is.

454
00:22:18,359 --> 00:22:22,720
Speaker 2: Like Northern Virginia the data center rally. That's a very

455
00:22:22,760 --> 00:22:27,720
Speaker 2: depressing thing. I want to die. My question is why

456
00:22:27,720 --> 00:22:29,840
Speaker 2: don't we I know this is perhaps a little bit

457
00:22:30,119 --> 00:22:32,400
Speaker 2: blunt force, but why are we not making them pay

458
00:22:32,560 --> 00:22:35,040
Speaker 2: to upgrade the infrastructure? Why why is the government not

459
00:22:35,119 --> 00:22:37,439
Speaker 2: just being like, you want this shit, go and build

460
00:22:37,480 --> 00:22:40,639
Speaker 2: it for us, give us the money, we'll do it.

461
00:22:40,760 --> 00:22:43,800
Speaker 2: Is that happy? I know we've this nuclear power plants, but.

462
00:22:45,880 --> 00:22:49,399
Speaker 3: In some places that is happening actually, so in the Dolls,

463
00:22:49,440 --> 00:22:52,240
Speaker 3: for example, in Oregon, you know where I think maybe

464
00:22:52,240 --> 00:22:54,840
Speaker 3: most people probably heard the story of that because Google,

465
00:22:55,160 --> 00:22:57,400
Speaker 3: you know, was trying to hold back its water use

466
00:22:58,560 --> 00:23:01,320
Speaker 3: you know, data and whatnot and eventually had to share it.

467
00:23:02,080 --> 00:23:04,280
Speaker 3: But in that case, you know, they have made an

468
00:23:04,320 --> 00:23:07,880
Speaker 3: agreement with the local council in order to upgrade their

469
00:23:07,880 --> 00:23:12,040
Speaker 3: water infrastructure so that they should have more water available. Again,

470
00:23:12,119 --> 00:23:15,480
Speaker 3: that doesn't mean that the community isn't still concerned about

471
00:23:15,480 --> 00:23:17,440
Speaker 3: water access and what's going to.

472
00:23:17,359 --> 00:23:21,080
Speaker 2: Happen the finite. It's not an unlimited amount of it.

473
00:23:21,200 --> 00:23:23,320
Speaker 2: More infrastructure isn't going to help it if we run

474
00:23:23,359 --> 00:23:23,880
Speaker 2: out of it.

475
00:23:24,600 --> 00:23:26,719
Speaker 3: Yeah, that's a pretty fundamental issue, right.

476
00:23:28,440 --> 00:23:30,359
Speaker 2: How do you feel about the nuclear power side, because

477
00:23:30,359 --> 00:23:32,879
Speaker 2: I'm kind of fifty to fifty. I like nuclear power,

478
00:23:33,800 --> 00:23:35,880
Speaker 2: I think, say it back, but it seems like it's

479
00:23:36,359 --> 00:23:39,080
Speaker 2: it seems like they're privatizing it, which is not solving

480
00:23:39,119 --> 00:23:40,160
Speaker 2: the problem.

481
00:23:40,600 --> 00:23:43,520
Speaker 3: Yeah, I would say I'm probably more on the skeptical

482
00:23:43,520 --> 00:23:46,120
Speaker 3: side of nuclear power, and that's for a few reasons.

483
00:23:46,680 --> 00:23:49,360
Speaker 3: Where nuclear power exists, I think it doesn't make any

484
00:23:49,400 --> 00:23:52,159
Speaker 3: sense to turn it off, right because that's going to

485
00:23:52,160 --> 00:23:55,040
Speaker 3: be far better than any kind of fossil fuel generation

486
00:23:55,160 --> 00:23:57,000
Speaker 3: that we're doing, and that should be kind of one

487
00:23:57,040 --> 00:23:59,359
Speaker 3: of the last things that we actually target for replacement

488
00:23:59,400 --> 00:24:02,520
Speaker 3: with renewed or whatnot. Like, I don't totally agree with say,

489
00:24:02,520 --> 00:24:05,240
Speaker 3: Germany turning off it's nuclear energy and going back to coal.

490
00:24:05,320 --> 00:24:08,200
Speaker 3: I think that that is a mistake. But I'm more

491
00:24:08,240 --> 00:24:12,320
Speaker 3: skeptical of investing in nuclear at this point and treating

492
00:24:12,359 --> 00:24:15,760
Speaker 3: it as you know, like a climate response, because we

493
00:24:15,920 --> 00:24:20,240
Speaker 3: know that nuclear energy is not only very expensive, but

494
00:24:20,359 --> 00:24:22,800
Speaker 3: takes so much time to like set up a new

495
00:24:23,080 --> 00:24:25,440
Speaker 3: nuclear plant, And it feels like at this point, when

496
00:24:25,440 --> 00:24:29,840
Speaker 3: we've seen the cost of installing solar and wind energy

497
00:24:29,880 --> 00:24:32,520
Speaker 3: declined so much in recent years, that it feels much

498
00:24:32,560 --> 00:24:35,359
Speaker 3: more kind of not just cost effective, but much more

499
00:24:35,480 --> 00:24:38,120
Speaker 3: rapid to just invest it in, you know, building out

500
00:24:38,160 --> 00:24:40,440
Speaker 3: large gale renewables instead.

501
00:24:41,000 --> 00:24:43,480
Speaker 2: Renewables don't generate power quite as fast.

502
00:24:44,680 --> 00:24:47,760
Speaker 3: Well possibly, you know you can you can set up

503
00:24:47,800 --> 00:24:50,399
Speaker 3: the battery storage facilities and things like that to store

504
00:24:50,440 --> 00:24:53,480
Speaker 3: things for the times when it's not generating. Yeah, that's

505
00:24:53,560 --> 00:25:08,800
Speaker 3: kind of the way that that I see it.

506
00:25:08,800 --> 00:25:10,919
Speaker 2: It's so funny. I think something's just come to me

507
00:25:11,040 --> 00:25:13,440
Speaker 2: with this as well. That really pisses me off, which

508
00:25:13,440 --> 00:25:16,080
Speaker 2: is I know, unusual for me, the sense that you

509
00:25:16,160 --> 00:25:18,480
Speaker 2: just mentioned battery storage. What if they put all the

510
00:25:18,520 --> 00:25:21,320
Speaker 2: money into investing into stuff like that, What if all

511
00:25:21,359 --> 00:25:23,520
Speaker 2: of this money could go into inventing them that would

512
00:25:23,520 --> 00:25:25,960
Speaker 2: be my make Casey brought this up recently. It's like

513
00:25:26,000 --> 00:25:28,280
Speaker 2: one of the very clear places money could go that

514
00:25:28,320 --> 00:25:32,000
Speaker 2: would be very good. Like if we had massive battery

515
00:25:32,040 --> 00:25:35,000
Speaker 2: storage for power, this would solve many problems and actually

516
00:25:35,000 --> 00:25:38,879
Speaker 2: probably create new things we could do, especially in double conscious,

517
00:25:38,920 --> 00:25:40,879
Speaker 2: we could dug genuinely do amazing things in the world.

518
00:25:41,040 --> 00:25:43,560
Speaker 2: Even describing it now, I feel more excited about this

519
00:25:43,600 --> 00:25:47,080
Speaker 2: than Generative AI. But it almost feels like they're kind

520
00:25:47,080 --> 00:25:49,679
Speaker 2: of lazy that they don't want to solve the actual

521
00:25:49,680 --> 00:25:52,240
Speaker 2: problems to get to the point that they just want

522
00:25:52,280 --> 00:25:54,320
Speaker 2: to build more and keep doing the bullshit they've been

523
00:25:54,359 --> 00:25:55,000
Speaker 2: doing for years.

524
00:25:56,080 --> 00:25:58,800
Speaker 3: I feel like part of it is profitability as well. Right, Like,

525
00:25:59,119 --> 00:26:01,800
Speaker 3: when you think about investing in like so called tech

526
00:26:01,920 --> 00:26:05,280
Speaker 3: or generative AI or what have you. Whenever we have

527
00:26:05,359 --> 00:26:08,159
Speaker 3: investors thinking about tech businesses, we're thinking about these kind

528
00:26:08,200 --> 00:26:10,639
Speaker 3: of really rapid takeoffs in the amount of money that

529
00:26:10,640 --> 00:26:12,639
Speaker 3: they're going to make. That the chance for these like

530
00:26:12,800 --> 00:26:16,320
Speaker 3: really significant payoffs is the money if the company really works.

531
00:26:16,760 --> 00:26:19,480
Speaker 3: And so you often have these companies trading at multiples

532
00:26:19,480 --> 00:26:21,840
Speaker 3: that are far above say, what a traditional company would

533
00:26:21,840 --> 00:26:24,600
Speaker 3: trade at when it goes public. Right Whereas if you

534
00:26:24,640 --> 00:26:27,919
Speaker 3: think of like a more traditional type of company, the

535
00:26:27,960 --> 00:26:30,040
Speaker 3: possibilities with the chances that you're going to get this

536
00:26:30,080 --> 00:26:32,200
Speaker 3: massive payoff are far lower, and so there's less of

537
00:26:32,200 --> 00:26:34,959
Speaker 3: an incentive to put your money into that type of place,

538
00:26:35,080 --> 00:26:38,080
Speaker 3: when say, some sort of tech business is going to

539
00:26:38,119 --> 00:26:41,199
Speaker 3: have you know, this much greater chance of having this

540
00:26:41,280 --> 00:26:43,439
Speaker 3: huge payoff. And so I think that that is like

541
00:26:43,520 --> 00:26:46,320
Speaker 3: one place where you know, obviously we live in a

542
00:26:46,359 --> 00:26:49,639
Speaker 3: capitalist system where our incentives are kind of misaligned, And

543
00:26:49,880 --> 00:26:52,000
Speaker 3: it's one of the things that I find quite silly,

544
00:26:52,000 --> 00:26:54,399
Speaker 3: where like, you know, the United States is having this

545
00:26:54,440 --> 00:26:56,960
Speaker 3: big feud with China now and very concerned about the

546
00:26:57,480 --> 00:27:01,199
Speaker 3: you know, the ability for Chinese tech companies and whatnot

547
00:27:01,240 --> 00:27:05,080
Speaker 3: to compete with US companies on the global stage automakers

548
00:27:05,080 --> 00:27:07,440
Speaker 3: and things as well. But one of the reasons that

549
00:27:07,520 --> 00:27:09,760
Speaker 3: solar energy is so cheap and one of the reasons

550
00:27:09,760 --> 00:27:13,840
Speaker 3: that we've had this like significant expansion in evs. You know,

551
00:27:13,880 --> 00:27:16,520
Speaker 3: we often point to Elon Musk for example, is like,

552
00:27:16,840 --> 00:27:18,840
Speaker 3: you know, the one who deserves all the credit for that.

553
00:27:19,280 --> 00:27:21,760
Speaker 3: But the Chinese have really been successful in bringing down

554
00:27:21,840 --> 00:27:24,639
Speaker 3: the costs of those types of technologies solar panels and

555
00:27:24,680 --> 00:27:28,239
Speaker 3: batteries and things like that in particular. And yeah, you know,

556
00:27:28,520 --> 00:27:30,080
Speaker 3: I think that we should be trying to like build

557
00:27:30,119 --> 00:27:32,200
Speaker 3: on that rather than just trying to like exclude the

558
00:27:32,280 --> 00:27:34,560
Speaker 3: cheaper stuff from our markets.

559
00:27:35,040 --> 00:27:38,440
Speaker 2: Well, fundamentally disagree Paris, because I don't want any Chinese

560
00:27:38,440 --> 00:27:43,200
Speaker 2: companies in like stealing my data, tracking Americans, using that

561
00:27:43,680 --> 00:27:47,200
Speaker 2: to monetize them somehow and manipulating them using that that's

562
00:27:47,280 --> 00:27:51,399
Speaker 2: for American businesses. We keep all surveillance capitalism in the

563
00:27:51,560 --> 00:27:54,000
Speaker 2: US of A. It's just pisses me off as well,

564
00:27:54,040 --> 00:27:58,000
Speaker 2: because I look, I'm not getting into geopolitics, but it

565
00:27:58,040 --> 00:28:01,840
Speaker 2: feels like some of the AI boom is even driven

566
00:28:01,880 --> 00:28:04,840
Speaker 2: by that cinephobia, the sense that if we don't build it,

567
00:28:04,880 --> 00:28:08,080
Speaker 2: the Chinese will build their AI and their garfields will

568
00:28:08,080 --> 00:28:12,000
Speaker 2: be even bustier than ours. And it's just frustrating because

569
00:28:12,520 --> 00:28:14,520
Speaker 2: I don't know about working with the Chinese. I'm not

570
00:28:14,520 --> 00:28:16,560
Speaker 2: going to get into that. But also it feels like

571
00:28:16,600 --> 00:28:18,720
Speaker 2: a dumber, stupid world. What we have another country that's

572
00:28:18,760 --> 00:28:21,919
Speaker 2: building things fast, probably in ways that we might not

573
00:28:22,160 --> 00:28:25,480
Speaker 2: want to do labor wise, I don't know. But nevertheless,

574
00:28:25,480 --> 00:28:29,280
Speaker 2: it feels like we're actually not all of this rapid expansion.

575
00:28:29,359 --> 00:28:33,040
Speaker 2: All of this shit we're supposedly building doesn't actually seem

576
00:28:33,080 --> 00:28:38,080
Speaker 2: to be innovation. It just is capitalist sprawl.

577
00:28:39,360 --> 00:28:41,920
Speaker 3: Yeah, I definitely agree. Right, it feels like Silicon Valley

578
00:28:42,000 --> 00:28:44,640
Speaker 3: left innovation behind quite a while ago. Like I think

579
00:28:44,680 --> 00:28:47,640
Speaker 3: that you can very genuinely say that in the early

580
00:28:47,720 --> 00:28:50,640
Speaker 3: days of the Internet there was innovation going on, right,

581
00:28:51,240 --> 00:28:53,800
Speaker 3: whether it was in you know, software development, but also

582
00:28:53,840 --> 00:28:56,360
Speaker 3: in hardware development too, right, you know, the emergence of

583
00:28:56,400 --> 00:28:58,960
Speaker 3: the mobile phone, and you know, we can even say

584
00:28:58,960 --> 00:29:01,440
Speaker 3: the iPad and those sorts of things, but it feels

585
00:29:01,440 --> 00:29:06,800
Speaker 3: like now, you know, those types of developments, those innovations

586
00:29:06,840 --> 00:29:09,600
Speaker 3: have matured, and it does feel like the industry is

587
00:29:09,880 --> 00:29:12,080
Speaker 3: you know, kind of sort of trying to grope for

588
00:29:12,160 --> 00:29:15,720
Speaker 3: whatever might come next, but really failing in doing that

589
00:29:15,880 --> 00:29:19,600
Speaker 3: because you know, their profits and their whole businesses are

590
00:29:19,640 --> 00:29:22,320
Speaker 3: tied up in what is currently successful at the moment,

591
00:29:22,600 --> 00:29:24,440
Speaker 3: and I think a lot of them don't want to

592
00:29:24,480 --> 00:29:27,360
Speaker 3: disrupt what is working for them and want to protect

593
00:29:27,600 --> 00:29:30,920
Speaker 3: these what are now basically legacy businesses that they have

594
00:29:31,000 --> 00:29:33,680
Speaker 3: built up and that they're now kind of soaking for cash, right,

595
00:29:34,960 --> 00:29:36,920
Speaker 3: And so I don't think that we should be looking

596
00:29:36,960 --> 00:29:39,840
Speaker 3: to the Googles or Apples or Amazons of the world

597
00:29:39,840 --> 00:29:42,720
Speaker 3: for innovation and for the path forward for what is

598
00:29:42,760 --> 00:29:45,240
Speaker 3: going to come next. And I think that even like

599
00:29:45,360 --> 00:29:47,800
Speaker 3: if you think back to like the Internet era and

600
00:29:48,080 --> 00:29:50,800
Speaker 3: things that came before, you know, often what a lot

601
00:29:50,800 --> 00:29:53,760
Speaker 3: of people have observed, you know, Mariana Mazocato has this

602
00:29:53,800 --> 00:29:56,479
Speaker 3: great book called I Think It's the Entrepreneurial State that

603
00:29:56,520 --> 00:29:58,600
Speaker 3: goes into this as well. How a lot of the

604
00:29:58,640 --> 00:30:01,160
Speaker 3: things that these companies were able to launch and able

605
00:30:01,200 --> 00:30:03,560
Speaker 3: to make so much money from were ultimately things that

606
00:30:03,600 --> 00:30:07,000
Speaker 3: were being developed in the public sector and that they

607
00:30:07,000 --> 00:30:09,520
Speaker 3: were able to privatize and make a lot of money

608
00:30:09,520 --> 00:30:09,760
Speaker 3: off of.

609
00:30:09,840 --> 00:30:10,000
Speaker 1: Right.

610
00:30:10,040 --> 00:30:13,240
Speaker 3: Obviously, the Internet was a public innovation before it was privatized,

611
00:30:13,280 --> 00:30:15,880
Speaker 3: and all these companies could use it and commercialize it

612
00:30:15,880 --> 00:30:18,800
Speaker 3: and whatnot. Right, And I feel like something has broken

613
00:30:18,840 --> 00:30:22,120
Speaker 3: down there as well, where so much public research seems

614
00:30:22,160 --> 00:30:24,520
Speaker 3: to be focused on a very early stage and like

615
00:30:24,880 --> 00:30:28,280
Speaker 3: what is the commercialized potential of this, rather than just saying,

616
00:30:28,320 --> 00:30:30,680
Speaker 3: forget about commercialization for a while and let's just like

617
00:30:30,760 --> 00:30:32,720
Speaker 3: work on these things and see if it goes anywhere.

618
00:30:33,240 --> 00:30:35,320
Speaker 3: And it just feels like in general, like you know,

619
00:30:35,400 --> 00:30:39,720
Speaker 3: our deep commitment to capitalism in our society has broken down.

620
00:30:40,040 --> 00:30:43,440
Speaker 3: Whereas before, like it used to be measured a bit

621
00:30:44,200 --> 00:30:46,920
Speaker 3: because there needed to be like certain deliverables for people

622
00:30:46,920 --> 00:30:49,760
Speaker 3: and you know, a different kind of an expectation. But

623
00:30:49,800 --> 00:30:51,600
Speaker 3: at this point we've just gone so full tilt on,

624
00:30:51,720 --> 00:30:53,680
Speaker 3: like whatever makes profit is what we need to do,

625
00:30:54,040 --> 00:30:56,200
Speaker 3: that we've lost the things that ultimately contribute to that

626
00:30:56,240 --> 00:30:58,240
Speaker 3: in the long term rather than just focusing on the

627
00:30:58,240 --> 00:30:58,880
Speaker 3: short term.

628
00:30:59,320 --> 00:31:02,160
Speaker 2: And it makes me think about like giving things time

629
00:31:02,240 --> 00:31:06,000
Speaker 2: to build, because generative AI as an idea, you have

630
00:31:06,040 --> 00:31:08,160
Speaker 2: to wonder if they left it alone for another five

631
00:31:08,200 --> 00:31:11,520
Speaker 2: to ten years without doing this, whether it might have

632
00:31:11,600 --> 00:31:14,800
Speaker 2: actually been good if there was potential for this, because

633
00:31:14,840 --> 00:31:17,520
Speaker 2: I think with long gone with large language models on

634
00:31:17,600 --> 00:31:19,640
Speaker 2: device stuff I think is cool. But nevertheless we're not

635
00:31:19,640 --> 00:31:22,800
Speaker 2: talking about that. But it's kind of we instead of

636
00:31:23,680 --> 00:31:27,040
Speaker 2: investing in public things or nonprofits that actually build the

637
00:31:27,080 --> 00:31:31,080
Speaker 2: building blocks that make innovation happen, we've allowed companies like

638
00:31:31,160 --> 00:31:34,520
Speaker 2: Qualcom to vacuum up various codecs and standards to the

639
00:31:34,520 --> 00:31:36,680
Speaker 2: point that most of them are owned by one company,

640
00:31:37,240 --> 00:31:39,960
Speaker 2: and then we pile our We don't really have anyone

641
00:31:40,000 --> 00:31:43,239
Speaker 2: in power anymore, so who understands what the fuck's going on?

642
00:31:43,600 --> 00:31:46,440
Speaker 2: So we pile all that cash into something that kind

643
00:31:46,440 --> 00:31:49,000
Speaker 2: of looks like the future because I think about the iPhone.

644
00:31:49,120 --> 00:31:51,000
Speaker 2: I was talking to someone about Jim Cavello and the

645
00:31:51,040 --> 00:31:54,480
Speaker 2: Generative AI paper from Goldman earlier, and he was saying

646
00:31:54,520 --> 00:31:57,440
Speaker 2: how one of the big things that made the smartphone

647
00:31:57,480 --> 00:31:59,600
Speaker 2: revolution happen, one of the things that they knew when

648
00:31:59,640 --> 00:32:03,520
Speaker 2: this happened, and we'd bring in was small GPS was

649
00:32:03,520 --> 00:32:06,720
Speaker 2: the ability to have device level GPS and of course

650
00:32:06,760 --> 00:32:08,960
Speaker 2: the chips that support and then someone would need to

651
00:32:08,960 --> 00:32:11,280
Speaker 2: build the software layer, which is where Apple came in,

652
00:32:11,320 --> 00:32:14,880
Speaker 2: and then Android to some extent, and we don't have

653
00:32:14,960 --> 00:32:18,560
Speaker 2: that building block. We have one thing. We have generative AI.

654
00:32:18,600 --> 00:32:21,680
Speaker 2: We have transformer based models, and we're going to put

655
00:32:21,720 --> 00:32:24,960
Speaker 2: all the money in that in the hopes that no

656
00:32:25,000 --> 00:32:27,400
Speaker 2: one's even thinking of what the next devices might look.

657
00:32:27,440 --> 00:32:29,880
Speaker 2: I can't stop thinking about batteries now, because that really

658
00:32:29,920 --> 00:32:33,160
Speaker 2: is it. It's if we had a battery that could

659
00:32:33,160 --> 00:32:35,000
Speaker 2: power I don't know, I'm being a bit wanky here

660
00:32:35,080 --> 00:32:37,520
Speaker 2: a city. Actually I'm not being wanky. This is less

661
00:32:37,520 --> 00:32:40,400
Speaker 2: wanky than what Sam Mormon says every day. That feels

662
00:32:40,480 --> 00:32:43,760
Speaker 2: like or incredibly small batteries which are powerful that would

663
00:32:43,880 --> 00:32:47,560
Speaker 2: enable all sorts of things. EDGAI is exciting, but they've

664
00:32:47,560 --> 00:32:51,000
Speaker 2: been working on it's been around ten years. None of

665
00:32:51,000 --> 00:32:55,080
Speaker 2: the people in power seem to actually be aware of

666
00:32:55,120 --> 00:32:58,200
Speaker 2: how good things are built, which I guess explains the

667
00:32:58,280 --> 00:33:00,280
Speaker 2: data center expansion, because if you think of it like

668
00:33:00,280 --> 00:33:02,680
Speaker 2: a dumb fuck, if you're like, huh, how do we

669
00:33:02,720 --> 00:33:05,000
Speaker 2: make money? We got those data centers right, Well, if

670
00:33:05,000 --> 00:33:08,240
Speaker 2: we build more of them. Yeah, what if they're really big,

671
00:33:09,320 --> 00:33:12,479
Speaker 2: then the money will come out. I'm just are you worried.

672
00:33:12,560 --> 00:33:15,560
Speaker 2: I'm a bit worried about the entire tech economy at

673
00:33:15,560 --> 00:33:17,280
Speaker 2: this point. That's put the environment, but like.

674
00:33:18,160 --> 00:33:20,680
Speaker 3: I'm worried about it all, you know, I'm less worried

675
00:33:20,680 --> 00:33:22,720
Speaker 3: about the tech economy in the sense of like will

676
00:33:22,720 --> 00:33:24,720
Speaker 3: they be profitable enough, will they make their money? Like

677
00:33:24,720 --> 00:33:26,840
Speaker 3: I don't really care about that. That's not like it's

678
00:33:26,880 --> 00:33:28,840
Speaker 3: something that's important to me, and I know that's the

679
00:33:28,840 --> 00:33:31,400
Speaker 3: same for you. But it's like I do worry about

680
00:33:31,400 --> 00:33:34,640
Speaker 3: where it's going because they're broad broader, like societal impacts

681
00:33:34,680 --> 00:33:36,760
Speaker 3: to everything that they do. Right on the one hand,

682
00:33:36,760 --> 00:33:39,160
Speaker 3: because of like the expectation that we need to adopt

683
00:33:39,160 --> 00:33:41,320
Speaker 3: so many of these things, but also because our governments

684
00:33:41,320 --> 00:33:43,840
Speaker 3: so much are supporting and you know, willing to push

685
00:33:43,880 --> 00:33:46,920
Speaker 3: out and not regulate effectively whatever it is that they

686
00:33:46,960 --> 00:33:49,640
Speaker 3: do until it's too late. And I feel like, you

687
00:33:49,680 --> 00:33:52,160
Speaker 3: know what you're talking about with generative AI even you know,

688
00:33:52,160 --> 00:33:54,080
Speaker 3: and I'm not going to claim to be, like, you know,

689
00:33:54,160 --> 00:33:57,240
Speaker 3: the most knowledgeable person in the world about the technical

690
00:33:57,280 --> 00:33:59,960
Speaker 3: angle of that, but when you look at what, say,

691
00:34:00,080 --> 00:34:01,880
Speaker 3: open AI was trying to do, and what all these

692
00:34:01,920 --> 00:34:04,000
Speaker 3: other companies have kind of chased after is they were

693
00:34:04,040 --> 00:34:07,680
Speaker 3: trying to build like the general foundation model that could

694
00:34:07,720 --> 00:34:12,160
Speaker 3: do virtually everything right, and you speak to AI researchers

695
00:34:12,239 --> 00:34:15,480
Speaker 3: or you know, I've spoken to some AI researchers who say, like, yes,

696
00:34:15,560 --> 00:34:18,759
Speaker 3: that is very energy intensive, that is going to take

697
00:34:18,800 --> 00:34:21,080
Speaker 3: a lot of power, energy data in order to.

698
00:34:21,120 --> 00:34:21,680
Speaker 2: Make it work.

699
00:34:21,880 --> 00:34:25,960
Speaker 3: But you can like train very tailored, much smaller models

700
00:34:26,120 --> 00:34:29,799
Speaker 3: that are not nearly as computationally or energy intensive as

701
00:34:30,160 --> 00:34:32,600
Speaker 3: the direction that these companies have chosen to go in.

702
00:34:32,880 --> 00:34:35,040
Speaker 3: But we are not doing that, you know. And I

703
00:34:35,040 --> 00:34:37,760
Speaker 3: think for a few reasons, mainly because on the one hand,

704
00:34:37,960 --> 00:34:40,640
Speaker 3: there is like the expectation of scale or the desire

705
00:34:40,719 --> 00:34:43,600
Speaker 3: of scale that comes with these general foundation models. I

706
00:34:43,640 --> 00:34:45,719
Speaker 3: think that there is the other side of it where

707
00:34:46,480 --> 00:34:49,200
Speaker 3: it's very beneficial to the tech companies that exist, you know,

708
00:34:49,239 --> 00:34:53,200
Speaker 3: the cloud giants, because if you're competing on this scale

709
00:34:53,239 --> 00:34:56,360
Speaker 3: of general foundation models that need so much computation in

710
00:34:56,440 --> 00:34:58,840
Speaker 3: order to train that, it's very difficult for smaller companies

711
00:34:58,840 --> 00:35:01,920
Speaker 3: to like compete on level. But then I think it

712
00:35:02,000 --> 00:35:04,880
Speaker 3: also plays into that ideological angle of it that I

713
00:35:04,920 --> 00:35:07,440
Speaker 3: was talking about earlier, where you have these people who

714
00:35:07,480 --> 00:35:10,560
Speaker 3: are leading the tech industry who fundamentally believe that we

715
00:35:10,680 --> 00:35:14,120
Speaker 3: need to build like you know, AI with human intelligence

716
00:35:14,120 --> 00:35:16,839
Speaker 3: and eventually merge our brains with machines. And so if

717
00:35:16,840 --> 00:35:19,920
Speaker 3: we're not building these massive general models, then we're not

718
00:35:19,960 --> 00:35:23,640
Speaker 3: getting closer in their minds to achieving this, which I

719
00:35:23,680 --> 00:35:26,160
Speaker 3: think is ultimately something that's never going to get built

720
00:35:26,160 --> 00:35:28,040
Speaker 3: because I don't think it makes sense. Actually, I think

721
00:35:28,040 --> 00:35:30,759
Speaker 3: it's just science fiction, but I think that is like

722
00:35:30,880 --> 00:35:32,319
Speaker 3: part of what's driving these people too.

723
00:35:33,560 --> 00:35:36,239
Speaker 2: So to finish us off, So you've got two episodes

724
00:35:36,280 --> 00:35:38,720
Speaker 2: out right, now, what have we got to look forward

725
00:35:38,719 --> 00:35:39,880
Speaker 2: to with the rest of the series?

726
00:35:39,920 --> 00:35:43,040
Speaker 3: Even yeah, absolutely, you know. The first one looked into

727
00:35:43,040 --> 00:35:45,240
Speaker 3: like what these data centers are, where they're coming from.

728
00:35:45,360 --> 00:35:48,160
Speaker 3: The second one looked into what this community opposition that

729
00:35:48,200 --> 00:35:51,279
Speaker 3: we're seeing looks like. The third one that will you

730
00:35:51,320 --> 00:35:53,719
Speaker 3: know come out soon, looks at the climate impacts of

731
00:35:53,800 --> 00:35:56,080
Speaker 3: generative AI and the way that generative AI is helping

732
00:35:56,120 --> 00:35:59,320
Speaker 3: to fuel this further construction of hyper scale data centers.

733
00:35:59,719 --> 00:36:02,640
Speaker 3: And the fourth one looks into the broader impacts of

734
00:36:02,680 --> 00:36:04,680
Speaker 3: this whether we can think about a different way of

735
00:36:04,680 --> 00:36:09,000
Speaker 3: approaching this problem and these ideological you know issues and

736
00:36:09,080 --> 00:36:12,280
Speaker 3: predispositions and and you know things that these tach billionaires

737
00:36:12,320 --> 00:36:14,759
Speaker 3: are wound up with that they are trying to push

738
00:36:14,760 --> 00:36:16,880
Speaker 3: on the rest of the world, and how that relates

739
00:36:16,880 --> 00:36:19,399
Speaker 3: to this anti democratic politics that so many of them

740
00:36:19,680 --> 00:36:20,360
Speaker 3: are adopting.

741
00:36:21,640 --> 00:36:24,080
Speaker 2: Well, Paris, it's been such a pleasure avenue on Where

742
00:36:24,080 --> 00:36:24,840
Speaker 2: can people find you?

743
00:36:26,040 --> 00:36:28,200
Speaker 3: Absolutely, you know, they can find the podcast tech won't

744
00:36:28,239 --> 00:36:30,440
Speaker 3: save us on any podcast app where they like to listen.

745
00:36:30,680 --> 00:36:32,759
Speaker 3: I'm on you know, most of the social media, the

746
00:36:32,800 --> 00:36:34,440
Speaker 3: tech based ones at least where you can find me

747
00:36:34,600 --> 00:36:36,920
Speaker 3: at Paris marks Well.

748
00:36:36,760 --> 00:36:38,680
Speaker 2: Thank you so much for listening, Paris. Thank you for

749
00:36:38,760 --> 00:36:41,120
Speaker 2: joining You've been listening to Better Offline. I'm the most

750
00:36:41,120 --> 00:36:44,200
Speaker 2: punishment alive at Zitron. You'll now get exactly the same

751
00:36:44,239 --> 00:36:47,799
Speaker 2: message as this, but slightly different, and you'll get mad

752
00:36:47,840 --> 00:36:49,600
Speaker 2: at me, and you'll email me because it's too say

753
00:36:49,719 --> 00:37:00,239
Speaker 2: me anyway, you're gonna hear it now. Thank you for

754
00:37:00,320 --> 00:37:02,920
Speaker 2: listening to Better Offline. The editor and composer of the

755
00:37:02,920 --> 00:37:06,040
Speaker 2: Better Offline theme song is Matasowski. You can check out

756
00:37:06,040 --> 00:37:09,719
Speaker 2: more of his music and audio projects at Matasowski dot com.

757
00:37:09,880 --> 00:37:13,400
Speaker 2: M A T T O S O W s KI

758
00:37:13,600 --> 00:37:16,479
Speaker 2: dot com. You can email me at easy at better

759
00:37:16,520 --> 00:37:18,960
Speaker 2: offline dot com or visit better offline dot com to

760
00:37:19,000 --> 00:37:21,840
Speaker 2: find more podcast links and of course my newsletter. I

761
00:37:21,880 --> 00:37:24,600
Speaker 2: also really recommend you go to chat dot where's youreaed

762
00:37:24,680 --> 00:37:26,920
Speaker 2: dot at to visit the discord, and go to our

763
00:37:27,000 --> 00:37:30,319
Speaker 2: slash Better Offline to check out our reddit. Thank you

764
00:37:30,360 --> 00:37:31,320
Speaker 2: so much for listening.

765
00:37:32,160 --> 00:37:34,840
Speaker 1: Better Offline is a production of cool Zone Media. For

766
00:37:34,960 --> 00:37:38,120
Speaker 1: more from cool Zone Media, visit our website cool Zonemedia

767
00:37:38,200 --> 00:37:41,040
Speaker 1: dot com, or check us out on the iHeartRadio app,

768
00:37:41,080 --> 00:37:43,760
Speaker 1: Apple Podcasts, or wherever you get your podcasts.

769
00:38:00,080 --> 00:38:00,120
Speaker 2: M