1 00:00:00,000 --> 00:00:01,840 Speaker 1: We're about to join the AI race. Were the first 2 00:00:01,880 --> 00:00:03,920 Speaker 1: center given the green light. This is an AI data 3 00:00:04,000 --> 00:00:06,000 Speaker 1: sent to give them the green light in southand data grid 4 00:00:06,000 --> 00:00:08,039 Speaker 1: to the company. They've got resource consent for a three 5 00:00:08,039 --> 00:00:10,479 Speaker 1: billion dollars seventy eight thousand square meter data sent to 6 00:00:10,520 --> 00:00:12,960 Speaker 1: twelve hundred jobs to build it. It'll use two hundred 7 00:00:12,960 --> 00:00:16,119 Speaker 1: and eighty megawatts of electricity. Doctor Olrik Speidel is a 8 00:00:16,200 --> 00:00:18,599 Speaker 1: senior lecturer in computer science at the University of Auckland 9 00:00:18,640 --> 00:00:19,880 Speaker 1: and as well. It's very good morning to. 10 00:00:19,840 --> 00:00:23,000 Speaker 2: You, Gordon, warning Mike, it seems good. 11 00:00:23,040 --> 00:00:25,040 Speaker 1: It seems very good news. Are we behind the eight 12 00:00:25,120 --> 00:00:26,800 Speaker 1: ball a bit? Shouldn't we have a few more of these? 13 00:00:26,840 --> 00:00:28,040 Speaker 1: Given what's happened globally? 14 00:00:30,520 --> 00:00:33,600 Speaker 2: Well two an extents on looking at it at that 15 00:00:33,640 --> 00:00:37,400 Speaker 2: particular project, it's actually quite interesting because you've just mentioned 16 00:00:37,400 --> 00:00:40,120 Speaker 2: a cigarette after off two hundred and eighty megawatts that 17 00:00:40,159 --> 00:00:43,120 Speaker 2: it's going to have in terms of power consumption, and 18 00:00:43,200 --> 00:00:45,080 Speaker 2: if you look at how that fits into the big picture, 19 00:00:45,479 --> 00:00:47,919 Speaker 2: you actually realized that this is not a sort of 20 00:00:47,960 --> 00:00:50,920 Speaker 2: a trivial amount. It's roughly the same amount you know, 21 00:00:51,000 --> 00:00:56,120 Speaker 2: ballpark wise that Wellington takes some power, and so building 22 00:00:56,160 --> 00:00:59,200 Speaker 2: this is going to be you know, obviously, you know, 23 00:00:59,440 --> 00:01:01,840 Speaker 2: the significant chunk of of the generation in the South 24 00:01:01,840 --> 00:01:05,200 Speaker 2: Island because you know, we you know, generate a lot 25 00:01:05,240 --> 00:01:07,280 Speaker 2: more power in the suth Island than we do in 26 00:01:07,280 --> 00:01:09,480 Speaker 2: the North Island. So this is going to add you know, 27 00:01:09,560 --> 00:01:11,400 Speaker 2: quite a bit to the load in the South Island 28 00:01:11,400 --> 00:01:15,120 Speaker 2: when once it's up and running. And yeah, this thousand, 29 00:01:15,160 --> 00:01:17,800 Speaker 2: two hundred jobs for building It sounds really really great. 30 00:01:17,880 --> 00:01:20,399 Speaker 2: But my questions really, once it's up and running, how 31 00:01:20,400 --> 00:01:21,440 Speaker 2: many jobs are they going to be? 32 00:01:21,640 --> 00:01:23,680 Speaker 1: I already know the answer. The answer is fifty. It 33 00:01:23,720 --> 00:01:27,080 Speaker 1: takes nothing to run a data set. The US arguments 34 00:01:27,520 --> 00:01:30,000 Speaker 1: around once these things arrive, and that things arrive in town, 35 00:01:30,080 --> 00:01:32,400 Speaker 1: they suck up the power and the price of power 36 00:01:32,440 --> 00:01:35,240 Speaker 1: goes up. Can we negate that are we smart enough 37 00:01:35,240 --> 00:01:36,880 Speaker 1: to have worked all that out and get around it? 38 00:01:38,000 --> 00:01:40,280 Speaker 2: Well, I'd love to know what exactly has been happening 39 00:01:40,280 --> 00:01:43,840 Speaker 2: in the negotiations there as to you know, what sort 40 00:01:43,840 --> 00:01:47,120 Speaker 2: of pole prices that's been promised, you know, whether any 41 00:01:47,160 --> 00:01:49,040 Speaker 2: sort of the profits are going to be staying in 42 00:01:49,080 --> 00:01:50,600 Speaker 2: the country. I Mean, one of the things it also 43 00:01:50,600 --> 00:01:53,000 Speaker 2: strikes me is that obviously they're going to be running 44 00:01:53,040 --> 00:01:55,640 Speaker 2: off the New Testament Ring network cable that's going to 45 00:01:55,640 --> 00:01:58,840 Speaker 2: come in. Otherwise, the few as we're sort of looking 46 00:01:58,880 --> 00:02:01,240 Speaker 2: at this as actually being in an I an AI 47 00:02:01,280 --> 00:02:03,480 Speaker 2: factory for the rest of New Zealand or something like this. 48 00:02:03,600 --> 00:02:06,080 Speaker 2: The the you know, the cables that go down south 49 00:02:06,120 --> 00:02:09,840 Speaker 2: from Auckland, and you know also, you know, anything that 50 00:02:09,960 --> 00:02:13,360 Speaker 2: sort of is international connectivity in New Zealand otherwise that 51 00:02:13,480 --> 00:02:16,639 Speaker 2: all sits in the North Island. And so building a 52 00:02:16,760 --> 00:02:19,520 Speaker 2: data center is putting things at the very very end 53 00:02:19,520 --> 00:02:21,960 Speaker 2: of a very long chain of cables that are all 54 00:02:22,000 --> 00:02:25,079 Speaker 2: sort of you know, reasonably susceptible to things like the 55 00:02:25,080 --> 00:02:28,320 Speaker 2: earthquake accidents for volcanic eruption sympots and pieces like that 56 00:02:28,320 --> 00:02:31,919 Speaker 2: that might give in the way. So you know, just 57 00:02:32,000 --> 00:02:34,480 Speaker 2: you know, from the connectivity perspective self and is not 58 00:02:34,520 --> 00:02:37,120 Speaker 2: normally a place we would be building a data center. 59 00:02:37,360 --> 00:02:41,680 Speaker 2: From the power perspective, it is. So my big question is, really, 60 00:02:42,240 --> 00:02:44,400 Speaker 2: you know, how many more of those can we afford 61 00:02:44,480 --> 00:02:47,440 Speaker 2: because you know, this, this this one thing that one's 62 00:02:47,520 --> 00:02:51,600 Speaker 2: going to really already shift literally the power balance in 63 00:02:51,639 --> 00:02:53,720 Speaker 2: the South Island quite a bit, because I mean this 64 00:02:53,800 --> 00:02:56,240 Speaker 2: is this is you know, you know, in terms of 65 00:02:56,639 --> 00:02:58,840 Speaker 2: a total power useless is about seventy percent of what 66 00:02:58,919 --> 00:03:03,040 Speaker 2: christ Church takes. So and then you can't simply will 67 00:03:03,080 --> 00:03:06,040 Speaker 2: in nearly at extra park consumers without running. I was 68 00:03:06,080 --> 00:03:07,480 Speaker 2: generating capacity at some point. 69 00:03:07,600 --> 00:03:08,720 Speaker 1: Yeah, it's interesting, isn't it? 70 00:03:08,720 --> 00:03:08,960 Speaker 2: All right? 71 00:03:08,960 --> 00:03:11,160 Speaker 1: Listen, appreciate your time very much, OLORX Spiedel, who's the 72 00:03:11,200 --> 00:03:14,200 Speaker 1: spidal who's the senior lecture at computer science? What is 73 00:03:14,240 --> 00:03:16,880 Speaker 1: interesting I'm reading in Australia this morning is they're running 74 00:03:16,880 --> 00:03:20,200 Speaker 1: into trouble around the world and consenting because there is 75 00:03:20,240 --> 00:03:22,200 Speaker 1: trouble every time a data center comes down. I'm not 76 00:03:22,200 --> 00:03:24,080 Speaker 1: against data centers, and it does go back to our 77 00:03:24,120 --> 00:03:26,080 Speaker 1: conversation we've had with the Prime Minister many times when 78 00:03:26,120 --> 00:03:27,639 Speaker 1: I've said, if we want to be the host of 79 00:03:27,760 --> 00:03:30,200 Speaker 1: data centers, how do we provide the power? And I 80 00:03:30,200 --> 00:03:32,480 Speaker 1: don't think we've even answered that question yet. But nevertheless, 81 00:03:32,520 --> 00:03:34,840 Speaker 1: they're running into a lot of trouble, particularly in America, 82 00:03:34,880 --> 00:03:37,920 Speaker 1: getting consent because towns have gone, oh yay data centers. 83 00:03:38,440 --> 00:03:40,560 Speaker 1: Then they arrive and then they the pricess I say, 84 00:03:40,600 --> 00:03:43,680 Speaker 1: of power goes through the roof. So consenting is becoming 85 00:03:43,760 --> 00:03:46,480 Speaker 1: a problem. So the story this morning, according to Elon 86 00:03:46,600 --> 00:03:48,600 Speaker 1: Musk and the Australian media is that they're looking to 87 00:03:48,640 --> 00:03:55,000 Speaker 1: build data centers in space. But there that's Elon. For 88 00:03:55,120 --> 00:03:58,160 Speaker 1: more from the mic Asking Breakfast, listen live to news 89 00:03:58,200 --> 00:04:01,680 Speaker 1: talks it'd be from six am, or follow the podcast 90 00:04:01,720 --> 00:04:02,600 Speaker 1: on iHeartRadio