1 00:00:00,120 --> 00:00:03,080 S1: So I'm next to a house on Induced Street, West Footscray, 2 00:00:03,080 --> 00:00:07,440 S1: and ten metres away is the data centre. So you 3 00:00:07,440 --> 00:00:11,240 S1: can probably hear some of the construction noise in the background. 4 00:00:11,520 --> 00:00:14,880 S1: It's an absolutely enormous site. We can see six water 5 00:00:14,920 --> 00:00:18,800 S1: cooling towers that are on the completed data centre to 6 00:00:18,840 --> 00:00:22,360 S1: cool down the racks and racks and racks of servers 7 00:00:22,360 --> 00:00:23,560 S1: that are inside that building. 8 00:00:24,440 --> 00:00:27,600 S2: That was investigative reporter Clay Lucas at the site of 9 00:00:27,600 --> 00:00:31,040 S2: what will become a data centre in Melbourne. It will, yes, 10 00:00:31,040 --> 00:00:34,600 S2: store data like emails, but crucially, it will also one 11 00:00:34,600 --> 00:00:38,640 S2: day provide the immense computing power needed to train and 12 00:00:38,640 --> 00:00:42,280 S2: run AI models that are predicted to define the next 13 00:00:42,280 --> 00:00:46,960 S2: century of the economy. I'm Samantha Cylindromorphus and you're listening 14 00:00:47,000 --> 00:00:52,160 S2: to Morning Edition from The Age and the Sydney Morning Herald. Today, 15 00:00:52,200 --> 00:00:56,120 S2: investigative reporter Clay Lucas on the Australians living next door 16 00:00:56,120 --> 00:00:59,880 S2: to these loud, energy sucking centres that some say are 17 00:00:59,920 --> 00:01:03,140 S2: a threat to our environment, and whether our state governments 18 00:01:03,140 --> 00:01:07,979 S2: are letting a rapidly evolving, resource intensive industry expand largely 19 00:01:07,980 --> 00:01:17,500 S2: unchecked until problems become undeniable. Now, Clay, we just heard 20 00:01:17,500 --> 00:01:20,179 S2: you out at that site of a major data center 21 00:01:20,180 --> 00:01:23,220 S2: in the works that's out in West Melbourne. Now, we 22 00:01:23,220 --> 00:01:25,860 S2: know that data centers have been around for a while, right? 23 00:01:25,860 --> 00:01:29,100 S2: But we are seeing a massive explosion of them. So 24 00:01:29,100 --> 00:01:30,020 S2: why is that? 25 00:01:30,500 --> 00:01:33,459 S1: Well, they have been around for decades in one form 26 00:01:33,459 --> 00:01:37,020 S1: or another. You know, we all know the computer server 27 00:01:37,020 --> 00:01:40,780 S1: rooms that were in our offices from the 70s and 28 00:01:40,780 --> 00:01:43,140 S1: 80s and 90s onwards that were, you know, might have 29 00:01:43,140 --> 00:01:46,380 S1: been in the basement or a little side corridor. You know, 30 00:01:46,420 --> 00:01:50,500 S1: those started moving offsite into what got marketed as the cloud, 31 00:01:50,500 --> 00:01:53,660 S1: but it was really just an off site server storage area. 32 00:01:54,300 --> 00:01:59,860 S1: In 2022, though, things exploded when ChatGPT launched their AI system. 33 00:02:00,490 --> 00:02:04,410 S1: And since then, there's just been this incredible uptake of 34 00:02:04,410 --> 00:02:10,010 S1: artificial intelligence by the wider community. And that's driving a 35 00:02:10,010 --> 00:02:15,730 S1: really strong demand for new computing power, both internationally, in Australia, 36 00:02:16,210 --> 00:02:18,690 S1: in America, this has just gone it's sort of out 37 00:02:18,730 --> 00:02:19,290 S1: of control. 38 00:02:19,290 --> 00:02:22,130 S3: Their water and everything on our phones are starting to 39 00:02:22,169 --> 00:02:24,370 S3: use AI without us even consenting to it. 40 00:02:24,410 --> 00:02:27,410 S4: Is it correct to say that you are in a 41 00:02:27,410 --> 00:02:30,730 S4: relationship of sorts with an AI chatbot? 42 00:02:30,970 --> 00:02:33,929 S5: I think that's that's an accurate description. 43 00:02:34,090 --> 00:02:38,330 S6: A teenager's death by suicide, encouraged by artificial intelligence. That's 44 00:02:38,330 --> 00:02:40,290 S6: what the parents of an Orange County teenager. 45 00:02:40,290 --> 00:02:45,210 S7: Systems now detect AI, text, AI, audio, AI images, AI, video. 46 00:02:45,250 --> 00:02:45,810 S7: What is the. 47 00:02:46,690 --> 00:02:49,410 S1: A lot of us are doing AI searches several times 48 00:02:49,410 --> 00:02:52,450 S1: over the course of the day, and we now know 49 00:02:52,450 --> 00:02:55,330 S1: that each of those searches is is powered by a 50 00:02:55,330 --> 00:02:59,090 S1: small computation, done largely in a data center like the 51 00:02:59,090 --> 00:03:03,550 S1: one that we visited, and it becomes clearer and clearer 52 00:03:03,550 --> 00:03:06,950 S1: to all of us that these AI searches don't come 53 00:03:06,950 --> 00:03:11,470 S1: at no cost to the environment, to energy, to water, 54 00:03:11,790 --> 00:03:14,470 S1: and that they bring with them a real impact. 55 00:03:15,270 --> 00:03:16,750 S8: Like I do guess that a lot of the world 56 00:03:16,750 --> 00:03:19,230 S8: gets covered in data centers over time. Do you really? 57 00:03:19,270 --> 00:03:20,990 S8: But I don't know, because maybe we put them in space. 58 00:03:21,710 --> 00:03:25,150 S9: Reports have shown how energy use by AI data centers 59 00:03:25,150 --> 00:03:28,510 S9: that we've been talking about are potentially causing electricity prices 60 00:03:28,510 --> 00:03:29,950 S9: to rise for some Americans. 61 00:03:30,150 --> 00:03:32,790 S10: You can lead a horse to water, but will there 62 00:03:32,790 --> 00:03:36,630 S10: be enough in the trough to run one ChatGPT query? 63 00:03:36,990 --> 00:03:39,710 S10: That is what some people have asked over the past year. 64 00:03:45,910 --> 00:03:47,470 S2: Okay, and I think it'd be fair to say that 65 00:03:47,470 --> 00:03:51,270 S2: many Australians do support the AI revolution, but perhaps they're 66 00:03:51,270 --> 00:03:55,990 S2: not aware that this revolution requires massive and let's just 67 00:03:55,990 --> 00:03:58,470 S2: call it pretty ugly buildings that for some people are 68 00:03:58,510 --> 00:04:01,810 S2: actually on their doorstep because not all of these data 69 00:04:01,850 --> 00:04:04,570 S2: centers are built out in far flung regional sites, right? 70 00:04:04,610 --> 00:04:08,250 S2: Like some of them are popping up in metropolitan suburban areas. Why? 71 00:04:08,250 --> 00:04:09,050 S2: Why is that? 72 00:04:09,370 --> 00:04:12,930 S1: Yeah. So in in Sydney you've got the real focus 73 00:04:12,930 --> 00:04:17,410 S1: around Macquarie Park and North Ryde. Um Lane Cove, there's 74 00:04:17,450 --> 00:04:19,690 S1: a few popping up and there is scores of them 75 00:04:20,089 --> 00:04:22,690 S1: sort of to the west of Parramatta in what I 76 00:04:22,810 --> 00:04:25,930 S1: would guess are old industrial areas in Melbourne, uh, where 77 00:04:25,930 --> 00:04:30,490 S1: I am, it's Brooklyn, West Footscray in the west. Um 78 00:04:30,810 --> 00:04:34,570 S1: derrimut and then further up in the north, Craigieburn and Donnybrook. 79 00:04:35,290 --> 00:04:38,010 S1: The data center operators really want them to be as 80 00:04:38,010 --> 00:04:40,930 S1: close as they can to metropolitan centres, just just for 81 00:04:40,930 --> 00:04:43,690 S1: the speed of responses. I mean, for the sort of 82 00:04:43,730 --> 00:04:46,610 S1: inferencing tasks that AI needs them to do, though, it's 83 00:04:46,610 --> 00:04:49,010 S1: better for them to be close. But the reason data 84 00:04:49,050 --> 00:04:52,450 S1: center operators are coming closer and closer to to the 85 00:04:52,450 --> 00:04:55,410 S1: centre of town, in particularly in old industrial areas, is 86 00:04:55,410 --> 00:04:59,680 S1: that those areas have really great access to power lines, 87 00:04:59,680 --> 00:05:03,160 S1: and they have access to really, really good water sources. 88 00:05:03,160 --> 00:05:05,200 S1: So new water pipes that have been laid in the 89 00:05:05,200 --> 00:05:09,400 S1: last couple of decades. The other less discussed reason is 90 00:05:09,400 --> 00:05:12,520 S1: that to build these data centers, you need a lot 91 00:05:12,520 --> 00:05:16,440 S1: of construction might. And you get all those workers from 92 00:05:16,440 --> 00:05:19,240 S1: metropolitan centers. So if you want to build a data 93 00:05:19,240 --> 00:05:21,800 S1: center quickly, if you're going to do it in regional 94 00:05:22,000 --> 00:05:25,600 S1: New South Wales, regional Victoria, it's going to cost you 95 00:05:25,600 --> 00:05:28,400 S1: a lot more in terms of manpower to to build 96 00:05:28,400 --> 00:05:29,760 S1: those those centers. 97 00:05:30,080 --> 00:05:32,760 S2: Are these data centers actually noisy once they're up and running, 98 00:05:32,760 --> 00:05:35,200 S2: or is it sort of those problems aren't there anymore 99 00:05:35,200 --> 00:05:37,440 S2: once they're actually up and running it just during the 100 00:05:37,440 --> 00:05:38,360 S2: construction phase. 101 00:05:38,760 --> 00:05:41,880 S1: So most of the residents in West Footscray, on the 102 00:05:41,880 --> 00:05:44,480 S1: days when there's not construction going, they say there is 103 00:05:44,520 --> 00:05:47,680 S1: a low background hum, but it's nothing like the construction 104 00:05:47,680 --> 00:05:51,040 S1: and that they think they can live alongside it relatively 105 00:05:51,040 --> 00:05:52,960 S1: comfortably once it's built. 106 00:05:53,440 --> 00:05:56,679 S11: Good. Just one second. Let me just get the dogs 107 00:05:56,680 --> 00:05:57,159 S11: out of the way. 108 00:05:57,820 --> 00:06:00,140 S1: There is these huge water towers on the site. There's 109 00:06:00,140 --> 00:06:02,460 S1: been anytime there's a windy day. One of the residents, 110 00:06:02,500 --> 00:06:05,500 S1: Jacquie Glover, told us that there was suddenly this whistling 111 00:06:05,500 --> 00:06:06,940 S1: noise through the water towers. 112 00:06:06,940 --> 00:06:10,140 S12: This was supposed to be apparently a children's playground and 113 00:06:10,140 --> 00:06:15,020 S12: a little park here, but that never happened. So. But yeah, 114 00:06:15,060 --> 00:06:17,740 S12: from an aesthetic point of view, it's not the prettiest 115 00:06:17,740 --> 00:06:21,740 S12: looking thing. It's quite imposing and people often stop and go, 116 00:06:21,779 --> 00:06:22,300 S12: what is it? 117 00:06:22,580 --> 00:06:25,580 S1: But it's it's probably a lesser issue once it's built 118 00:06:25,580 --> 00:06:28,940 S1: than the construction process. The thing is, though, that the 119 00:06:28,940 --> 00:06:33,820 S1: construction is just like I haven't seen construction on this scale, 120 00:06:34,100 --> 00:06:36,940 S1: other than some of the mega road and rail projects 121 00:06:36,940 --> 00:06:39,260 S1: that Victoria's been doing in the last few years. Like 122 00:06:39,260 --> 00:06:43,020 S1: the West Footscray one is costing $1.5 billion in total. 123 00:06:43,060 --> 00:06:47,460 S1: It's being built probably eventually over ten hectares. This is 124 00:06:47,460 --> 00:06:49,300 S1: a really big site that's going to take a long 125 00:06:49,300 --> 00:06:50,179 S1: time to fill in. 126 00:06:51,700 --> 00:06:55,659 S12: I do worry about the water consumption of these places 127 00:06:55,660 --> 00:06:59,039 S12: and how safe it really is living next to such 128 00:06:59,080 --> 00:07:01,960 S12: a massive building. If there was to be a major 129 00:07:01,960 --> 00:07:03,240 S12: fire or. 130 00:07:03,720 --> 00:07:05,960 S1: So, one of the residents we interviewed was fired. Yusuf 131 00:07:06,120 --> 00:07:09,720 S1: Thad's situation was really not great for him. He lives 132 00:07:09,720 --> 00:07:13,120 S1: opposite the main entrance to the site. The noise is excessive. 133 00:07:13,320 --> 00:07:16,920 S1: There's there's constantly light being beamed onto the site. He's 134 00:07:16,920 --> 00:07:20,560 S1: had to get new curtains to block out that light, and. 135 00:07:20,560 --> 00:07:24,560 S13: Then also even on the weekends that sometimes that because, 136 00:07:24,600 --> 00:07:29,040 S13: for example, on Saturday they can work from 7 a.m. 137 00:07:29,080 --> 00:07:30,320 S13: and then there's a few times. 138 00:07:30,320 --> 00:07:33,440 S1: Every other weekend they're working on Saturday and Sunday. And 139 00:07:33,440 --> 00:07:35,320 S1: he just wanted to know why he couldn't get any 140 00:07:35,320 --> 00:07:38,480 S1: respite from from the building. And he wasn't at all 141 00:07:38,480 --> 00:07:41,200 S1: opposed to the data center being built there. It was 142 00:07:41,200 --> 00:07:45,320 S1: purely the construction issues. But, you know, they are going 143 00:07:45,360 --> 00:07:47,000 S1: to go on for years and years. So it's not 144 00:07:47,000 --> 00:07:49,280 S1: like it's just a temporary condition for him. 145 00:07:51,320 --> 00:07:51,800 S2: We'll be right. 146 00:07:51,800 --> 00:07:52,240 S4: Back. 147 00:07:58,670 --> 00:07:59,750 S14: And I really want to ask. 148 00:07:59,750 --> 00:08:03,790 S2: You about reports about the massive energy usage and water 149 00:08:03,790 --> 00:08:07,670 S2: consumption that these data centers require. There's one figure quite 150 00:08:07,670 --> 00:08:11,550 S2: astonishing from Sydney Water. It estimates that in 2035, about 151 00:08:11,550 --> 00:08:15,070 S2: a quarter of all Sydney's water will go towards running 152 00:08:15,070 --> 00:08:18,310 S2: the data centres in that state. So does that concern you? 153 00:08:18,830 --> 00:08:22,110 S1: It is a really striking figure with with water consumption 154 00:08:22,110 --> 00:08:27,190 S1: and it is a really serious concern. But the problem 155 00:08:27,190 --> 00:08:29,830 S1: is that there is what is sort of referred to 156 00:08:29,870 --> 00:08:32,430 S1: as phantom demand in the industry, where you get all 157 00:08:32,470 --> 00:08:35,630 S1: of these predictions of how much water will be used 158 00:08:36,190 --> 00:08:38,870 S1: if all of these data centres are built, and if 159 00:08:38,870 --> 00:08:42,230 S1: they were required to be cooled on the hottest days 160 00:08:42,230 --> 00:08:46,430 S1: that Australia offers. And the reality is that's probably not 161 00:08:46,429 --> 00:08:49,950 S1: going to. They won't all be built and we don't 162 00:08:50,230 --> 00:08:51,589 S1: you know, we do get a lot of hot days. 163 00:08:51,590 --> 00:08:56,070 S1: But the water companies have to accommodate the worst case scenario. 164 00:08:56,530 --> 00:09:00,689 S1: and so you're getting slightly hysterical concerns about how much 165 00:09:00,730 --> 00:09:04,450 S1: water will be used. These centers do use huge amounts 166 00:09:04,450 --> 00:09:08,650 S1: of water. At the moment, the technology is progressing so 167 00:09:08,650 --> 00:09:14,170 S1: that the new generation centers have these closed loop supplies 168 00:09:14,170 --> 00:09:16,810 S1: that use a lot less water. The issue there is 169 00:09:16,850 --> 00:09:19,929 S1: that in order to cool them, they use a lot 170 00:09:19,929 --> 00:09:23,490 S1: more power. So the downside of that, they then have 171 00:09:23,490 --> 00:09:26,130 S1: to increase the amount of air conditioning power they use. 172 00:09:26,570 --> 00:09:29,890 S1: And that is huge. They will then say, well, we're 173 00:09:29,929 --> 00:09:33,490 S1: offsetting that with renewables. The issue there is that there's 174 00:09:33,490 --> 00:09:36,530 S1: not enough renewable power to to to power all these 175 00:09:36,530 --> 00:09:41,210 S1: data centers. They might sign up agreements with renewable power operators, 176 00:09:41,210 --> 00:09:44,090 S1: but if there's not enough renewables there to power them, 177 00:09:44,090 --> 00:09:46,090 S1: then they're just going to go back to the grid 178 00:09:46,090 --> 00:09:50,090 S1: and end up running off coal. The concerns are real. 179 00:09:50,330 --> 00:09:53,570 S1: The water question is perhaps less alarming than the amount 180 00:09:53,570 --> 00:09:56,950 S1: of power that they are capable of sucking out of 181 00:09:56,950 --> 00:09:57,830 S1: the energy system. 182 00:09:57,830 --> 00:09:59,470 S14: Well, I really wanted to ask you about that. Okay, 183 00:09:59,510 --> 00:09:59,830 S14: so it. 184 00:09:59,830 --> 00:10:02,949 S2: Seems like they use an incredible amount of water, but 185 00:10:02,950 --> 00:10:05,030 S2: it's not like we're going to be left without enough 186 00:10:05,030 --> 00:10:08,390 S2: drinking water. So that's that's reassuring. But with regards to 187 00:10:08,429 --> 00:10:12,830 S2: the power that they require, will these data centers require 188 00:10:12,830 --> 00:10:15,910 S2: so much that say, we then don't have enough energy 189 00:10:15,910 --> 00:10:19,110 S2: for other functions that we need? Because one of our 190 00:10:19,110 --> 00:10:22,950 S2: colleagues was writing about this issue in New South Wales, 191 00:10:23,230 --> 00:10:24,830 S2: and he wrote that if every data center in the 192 00:10:24,830 --> 00:10:28,069 S2: New South Wales planning portal is built, their combined maximum 193 00:10:28,070 --> 00:10:31,990 S2: power demands in western Sydney will climb to about 4.4GW 194 00:10:31,990 --> 00:10:35,350 S2: in a decade, which is equivalent to the average electricity 195 00:10:35,350 --> 00:10:38,870 S2: load of more than 10 million households. And at their peak, 196 00:10:38,870 --> 00:10:41,190 S2: they'll demand almost four times as much power as the 197 00:10:41,190 --> 00:10:43,190 S2: rest of the city. So as a layperson like me 198 00:10:43,230 --> 00:10:47,790 S2: reads that and goes, that doesn't sound potentially sustainable or 199 00:10:47,790 --> 00:10:50,790 S2: potentially that sounds problematic. What would you say, Clay? 200 00:10:51,429 --> 00:10:54,429 S1: I think this comes back to the question of how 201 00:10:54,429 --> 00:10:57,260 S1: many of these centers ultimately get built? I've spent a 202 00:10:57,300 --> 00:11:00,940 S1: lot of time talking to the CEO of Data Centers Australia, 203 00:11:00,980 --> 00:11:04,140 S1: Belinda Dennett, who's been part of the rollout of data 204 00:11:04,140 --> 00:11:08,020 S1: centers for years and years now. She points to a 205 00:11:08,059 --> 00:11:10,540 S1: lot of research that shows it's six out of seven 206 00:11:11,140 --> 00:11:14,420 S1: projected data centers don't get built, which means there is 207 00:11:14,420 --> 00:11:19,660 S1: this phantom demand for power. So the electricity operators have 208 00:11:19,660 --> 00:11:23,580 S1: to plan for the worst possible scenario. But really in Australia, 209 00:11:23,580 --> 00:11:26,860 S1: we have a strong enough system of governance. That means 210 00:11:26,860 --> 00:11:29,740 S1: we're never going to get to a situation where electricity 211 00:11:29,740 --> 00:11:34,060 S1: companies will just blithely give away energy that will be 212 00:11:34,100 --> 00:11:37,459 S1: needed to do other things. The reason this, the topic 213 00:11:37,460 --> 00:11:41,260 S1: of data centers has become such a fever pitch in 214 00:11:41,260 --> 00:11:44,180 S1: America in particular, is that and have the system of 215 00:11:44,179 --> 00:11:48,620 S1: governance that we've got here that that, um, manages the 216 00:11:48,620 --> 00:11:52,699 S1: system reasonably well. So you're getting these really pitched battles 217 00:11:52,700 --> 00:11:57,040 S1: between people fighting over energy and residents seeing their bills 218 00:11:57,040 --> 00:12:01,359 S1: go up by up to 200%. That won't happen here. 219 00:12:01,800 --> 00:12:04,640 S1: The bigger question is whether the data centers can be 220 00:12:04,640 --> 00:12:08,719 S1: run on renewables, and at the moment, they're growing so 221 00:12:08,720 --> 00:12:11,840 S1: quickly that there's no way renewable power is is coming 222 00:12:11,840 --> 00:12:14,080 S1: online quick enough to deal with the pace of growth 223 00:12:14,120 --> 00:12:15,880 S1: of of new data centers. 224 00:12:16,280 --> 00:12:18,160 S2: And so, Clay, a big theme in all of this is, 225 00:12:18,200 --> 00:12:21,840 S2: of course, the pace with which data centers are being built. Obviously, 226 00:12:21,840 --> 00:12:24,760 S2: the fast pace with which AI is evolving. So are 227 00:12:24,760 --> 00:12:27,360 S2: people just feeling a bit out of control here? You know, 228 00:12:27,640 --> 00:12:31,319 S2: have our governments or other relevant bodies actually adequately assessed 229 00:12:31,480 --> 00:12:32,720 S2: the risks that are involved? 230 00:12:32,960 --> 00:12:36,079 S1: I think what's been happening is the AI companies and 231 00:12:36,080 --> 00:12:39,560 S1: the data center operators are really well organized, and they 232 00:12:39,559 --> 00:12:42,880 S1: have been getting in the ears of state and federal governments, 233 00:12:42,880 --> 00:12:47,080 S1: particularly the Victorian and New South Wales governments, and telling 234 00:12:47,080 --> 00:12:50,880 S1: them what their needs are. And those governments are so 235 00:12:50,880 --> 00:12:54,620 S1: desperate for jobs that they've been just taking their word 236 00:12:54,620 --> 00:12:57,260 S1: for it and letting them run to the point where 237 00:12:57,260 --> 00:13:03,579 S1: in Victoria had our economic growth and jobs minister saying that, um, 238 00:13:04,220 --> 00:13:07,140 S1: AI should be put in a put on a leash 239 00:13:07,140 --> 00:13:10,500 S1: and allowed to run, and it leads to a situation 240 00:13:10,500 --> 00:13:16,020 S1: where these just enormous corporations with billions and billions of 241 00:13:16,020 --> 00:13:20,260 S1: dollars are dictating how our cities are built based on 242 00:13:20,300 --> 00:13:23,060 S1: their needs, rather than the people who are affected. Like 243 00:13:23,059 --> 00:13:25,500 S1: this case study we've come up, come up with in 244 00:13:25,500 --> 00:13:29,260 S1: West Footscray of having a data centre right on their street. 245 00:13:30,059 --> 00:13:33,380 S1: You know, they're not getting a say until they suddenly 246 00:13:33,420 --> 00:13:37,179 S1: start seeing the massive problems. They're seeing their their footpaths 247 00:13:37,179 --> 00:13:41,740 S1: dug up, their roads shut. There's noise going 24 over seven. 248 00:13:43,580 --> 00:13:46,420 S1: I think politicians have just seen the jobs that are 249 00:13:46,420 --> 00:13:50,860 S1: on offer from these enormous projects, and all of the 250 00:13:51,620 --> 00:13:55,970 S1: money that will flow into the state and they're only 251 00:13:55,970 --> 00:13:59,290 S1: just starting to realize that they are going to face 252 00:13:59,290 --> 00:14:03,130 S1: serious pushback from residents who are concerned about not only 253 00:14:03,130 --> 00:14:05,610 S1: the impacts on them, but the effects on the environment 254 00:14:05,610 --> 00:14:07,449 S1: and the effects on the wider city. 255 00:14:15,890 --> 00:14:19,890 S15: Today's episode of The Morning Edition was produced by Julia Carcasole. 256 00:14:19,930 --> 00:14:23,210 S15: Our executive producer is Tammy Mills. To listen to our 257 00:14:23,210 --> 00:14:26,370 S15: episodes as soon as they drop, follow the Morning Edition 258 00:14:26,370 --> 00:14:30,970 S15: on Apple, Spotify, or wherever you listen to podcasts. And 259 00:14:30,970 --> 00:14:32,930 S15: to stay up to date, sign up to our Morning 260 00:14:32,930 --> 00:14:35,490 S15: Edition newsletter to receive a summary of the day's most 261 00:14:35,490 --> 00:14:39,610 S15: important news in your inbox every morning. Links are in 262 00:14:39,610 --> 00:14:43,890 S15: the show. Notes. I'm Samantha Selinger. Morris, thanks for listening.