1 00:00:06,179 --> 00:00:09,720 Speaker 1: Welcome to Kobe Time, a podcast series on Markets and 2 00:00:09,728 --> 00:00:13,239 Speaker 1: Economies from Devious Group Research. I'm ba chief economist, welcoming 3 00:00:13,250 --> 00:00:15,670 Speaker 1: you to our 108th episode. 4 00:00:16,388 --> 00:00:20,038 Speaker 1: If any book can make general audiences understand the Silicon 5 00:00:20,047 --> 00:00:23,358 Speaker 1: age and finally recognize how it rivals the atomic age 6 00:00:23,368 --> 00:00:27,159 Speaker 1: for drama and import chip war is it? That's what 7 00:00:27,169 --> 00:00:29,589 Speaker 1: Virginia Hefferman wrote at the end of her New York 8 00:00:29,599 --> 00:00:30,518 Speaker 1: Times book review. 9 00:00:30,829 --> 00:00:33,759 Speaker 1: I'm very glad to have the Chip Wars author on 10 00:00:33,770 --> 00:00:38,020 Speaker 1: Coy Time. Chris Miller is associated professor of International History 11 00:00:38,119 --> 00:00:41,759 Speaker 1: at the Fletcher School at Tufts University. He published Chip 12 00:00:41,770 --> 00:00:44,720 Speaker 1: War last year and it has been a major success 13 00:00:44,729 --> 00:00:48,319 Speaker 1: winning the Financial Times best business book award for 2022. 14 00:00:48,330 --> 00:00:50,659 Speaker 1: Chris Miller, a warm welcome to Cope time. 15 00:00:51,720 --> 00:00:54,330 Speaker 1: Thank you for having me. So great to have you. 16 00:00:54,340 --> 00:00:57,889 Speaker 1: Uh Chris, I'll begin by playing the devil's advocate. Uh 17 00:00:57,900 --> 00:01:01,689 Speaker 1: the nuclear era quantum mechanics, DNS double he legs man 18 00:01:01,700 --> 00:01:04,410 Speaker 1: on the moon. Even in my world, the futures and 19 00:01:04,419 --> 00:01:08,679 Speaker 1: options in finance. All of these civilization, transforming advances took 20 00:01:08,690 --> 00:01:11,449 Speaker 1: place in the 20th century with little or no shape, 21 00:01:11,459 --> 00:01:12,830 Speaker 1: processing power behind them, 22 00:01:13,290 --> 00:01:18,300 Speaker 1: modern cities, industrial agriculture, amazing bridges and tunnels, aviation. All 23 00:01:18,309 --> 00:01:21,330 Speaker 1: these predate the computer age. And yet in your book, 24 00:01:21,339 --> 00:01:25,699 Speaker 1: you state that today's military, economic and geopolitical power are 25 00:01:25,709 --> 00:01:29,589 Speaker 1: built on a foundation of computer chips. Are you overseeing 26 00:01:29,599 --> 00:01:30,080 Speaker 1: the case? 27 00:01:33,760 --> 00:01:36,379 Speaker 2: Well, I, I think you're right that humanity has had 28 00:01:36,389 --> 00:01:40,199 Speaker 2: lots of advances before chips were invented. Um, but it's 29 00:01:40,209 --> 00:01:43,519 Speaker 2: also the case today that it's hard to go about 30 00:01:43,529 --> 00:01:47,400 Speaker 2: your daily life without interacting with not just dozens but 31 00:01:47,410 --> 00:01:52,440 Speaker 2: hundreds or thousands of semiconductors in your phone. There are 32 00:01:52,449 --> 00:01:53,000 Speaker 2: dozens of semi 33 00:01:53,480 --> 00:01:57,319 Speaker 2: devices in your computer in the data centers that process 34 00:01:57,330 --> 00:01:59,540 Speaker 2: and store your data in your car. If it's a 35 00:01:59,550 --> 00:02:04,080 Speaker 2: new car, it will have on average 1000 semiconductors inside 36 00:02:04,089 --> 00:02:08,418 Speaker 2: in your refrigerator in your dishwasher in your microwave. Uh 37 00:02:08,429 --> 00:02:11,729 Speaker 2: And almost every device you touch, there are lots and 38 00:02:11,740 --> 00:02:12,960 Speaker 2: lots of semiconductors and 39 00:02:13,309 --> 00:02:17,169 Speaker 2: they are so ubiquitous around us that we don't even 40 00:02:17,179 --> 00:02:20,250 Speaker 2: think about them. Uh But in fact, as we learned 41 00:02:20,258 --> 00:02:24,220 Speaker 2: uh during the pandemic era, shortages, modern economy simply can't 42 00:02:24,229 --> 00:02:25,960 Speaker 2: function without them. 43 00:02:26,470 --> 00:02:29,168 Speaker 2: And so in that sense, I think they're the most 44 00:02:29,179 --> 00:02:33,619 Speaker 2: economically critical technology because during the pandemic, the largest dollar 45 00:02:33,630 --> 00:02:37,869 Speaker 2: value uh disruption to manufacturing uh in terms of component 46 00:02:37,880 --> 00:02:41,440 Speaker 2: shortages came from uh semiconductor shortages. And so I think 47 00:02:41,449 --> 00:02:46,199 Speaker 2: that was one very brief and relatively small illustration of 48 00:02:46,210 --> 00:02:49,978 Speaker 2: just how critically economically important semiconductors have become 49 00:02:51,580 --> 00:02:56,960 Speaker 1: economic criticality. Aside. Do you also see security and conflict 50 00:02:56,970 --> 00:03:01,070 Speaker 1: propensity coming in with the availability or lack there of chips. 51 00:03:04,309 --> 00:03:07,330 Speaker 2: Well, certainly militaries are just as reliant on ships as 52 00:03:07,339 --> 00:03:11,600 Speaker 2: the rest of us. Um militaries have reoriented the way 53 00:03:11,610 --> 00:03:15,720 Speaker 2: they devise systems to take advantage of the sensing, the communications, 54 00:03:15,729 --> 00:03:19,279 Speaker 2: the computing power that semiconductors enable. And just like today, 55 00:03:19,288 --> 00:03:22,119 Speaker 2: it's impossible to buy a car without many, many chips 56 00:03:22,130 --> 00:03:22,339 Speaker 2: and stuff. 57 00:03:22,505 --> 00:03:25,384 Speaker 2: So too, it's impossible to feel a modern tank or 58 00:03:25,395 --> 00:03:29,755 Speaker 2: a plane or a drone without lots of chips as well. 59 00:03:29,764 --> 00:03:33,535 Speaker 2: And just like in cars or consumer devices, the more 60 00:03:33,544 --> 00:03:36,544 Speaker 2: advanced the communication, the more advanced the sensing, the more 61 00:03:36,554 --> 00:03:39,085 Speaker 2: advanced the computing, the more chips are inside the same 62 00:03:39,095 --> 00:03:40,434 Speaker 2: is true of defense systems 63 00:03:40,750 --> 00:03:45,279 Speaker 2: as well. Which is why today every major military is 64 00:03:45,289 --> 00:03:48,360 Speaker 2: thinking harder than ever about its access to semiconductors, both 65 00:03:48,369 --> 00:03:51,350 Speaker 2: in terms of assuring that it can access the chips 66 00:03:51,360 --> 00:03:54,279 Speaker 2: that it needs. But also trying to find new ways 67 00:03:54,289 --> 00:03:57,600 Speaker 2: to apply the capabilities that chips enable the defense systems. 68 00:03:57,610 --> 00:03:59,520 Speaker 2: If you talk to defense ministries, you'll hear a lot 69 00:03:59,529 --> 00:04:03,779 Speaker 2: today about autonomy about applying A I to defense systems 70 00:04:03,789 --> 00:04:07,580 Speaker 2: and intelligence capabilities. And all of that means using more 71 00:04:07,589 --> 00:04:10,419 Speaker 2: and more semiconductors for military purposes. 72 00:04:11,259 --> 00:04:12,820 Speaker 1: Right? And in your book also, I think there are 73 00:04:12,830 --> 00:04:17,209 Speaker 1: these fascinating chapters about the Soviet Union's attempts to get 74 00:04:17,220 --> 00:04:20,750 Speaker 1: some degree of uh security on ship making and how 75 00:04:20,760 --> 00:04:24,950 Speaker 1: spectacularly fi you know, it all ended in largely failure. 76 00:04:24,959 --> 00:04:26,899 Speaker 1: And you point out that by the early nineties, the 77 00:04:26,910 --> 00:04:28,730 Speaker 1: gap was so insurmountable that even 78 00:04:28,988 --> 00:04:31,488 Speaker 1: Soviet defense specialists would say one of the major reasons 79 00:04:31,500 --> 00:04:34,719 Speaker 1: for their decline was the tech related uh gap with 80 00:04:34,730 --> 00:04:38,959 Speaker 1: the US and its allies. Um Chris, your book has 81 00:04:38,970 --> 00:04:42,518 Speaker 1: just fantastic first. So 67 chapters on the history of 82 00:04:42,529 --> 00:04:44,880 Speaker 1: chip making. And we go all the way back to 83 00:04:44,890 --> 00:04:48,720 Speaker 1: the forties and fifties, the uh the noise and Shockley 84 00:04:48,730 --> 00:04:49,859 Speaker 1: of the world. And 85 00:04:49,945 --> 00:04:52,975 Speaker 1: by the way, I did my phd at Urbana Champagne, Illinois. 86 00:04:52,984 --> 00:04:55,565 Speaker 1: I did not know about John Barin till I read 87 00:04:55,575 --> 00:04:57,674 Speaker 1: your book. And now I know why there's a Barine 88 00:04:57,684 --> 00:05:03,315 Speaker 1: quad at Illinois. Um So looking at at that uh and, 89 00:05:03,325 --> 00:05:06,815 Speaker 1: and the incredible success that Silicon Valley has had over 90 00:05:06,825 --> 00:05:10,844 Speaker 1: the last 50 60 years in terms of pushing uh innovation. 91 00:05:11,024 --> 00:05:14,974 Speaker 1: So help us sort of walk through the critical ingredients 92 00:05:14,984 --> 00:05:17,875 Speaker 1: behind this continuous series of innovation in the area of 93 00:05:17,885 --> 00:05:18,734 Speaker 1: computer chips. 94 00:05:21,339 --> 00:05:22,390 Speaker 1: Well, the the the 95 00:05:22,399 --> 00:05:24,649 Speaker 2: key to the rate of innovation in the chip industry 96 00:05:24,660 --> 00:05:28,019 Speaker 2: is to understand Moore's law. Um Gordon Moore was one 97 00:05:28,029 --> 00:05:31,160 Speaker 2: of the co-founders of Intel even before he founded Intel, 98 00:05:31,170 --> 00:05:33,029 Speaker 2: though he was involved in the chip industry. And he 99 00:05:33,040 --> 00:05:37,190 Speaker 2: realized very, very early on that the number of transistors 100 00:05:37,200 --> 00:05:39,790 Speaker 2: per chip was doubling every year or two. 101 00:05:40,079 --> 00:05:43,079 Speaker 2: Um And at the time a cutting edge ship would 102 00:05:43,089 --> 00:05:46,169 Speaker 2: have a very small number of transistors on it. Today, 103 00:05:46,178 --> 00:05:50,089 Speaker 2: your phone has billions of transistors, but just uh six 104 00:05:50,100 --> 00:05:53,410 Speaker 2: years after the first ship was invented, that he saw 105 00:05:53,420 --> 00:05:56,238 Speaker 2: a pattern of doubling every year or two and predicted 106 00:05:56,250 --> 00:06:00,269 Speaker 2: that that rate of doubling would continue for some time. 107 00:06:00,279 --> 00:06:01,049 Speaker 2: He thought 108 00:06:01,890 --> 00:06:03,920 Speaker 2: continue for at least a decade. It turns out it 109 00:06:03,928 --> 00:06:06,880 Speaker 2: continued all the way up basically to the present. And 110 00:06:06,890 --> 00:06:09,799 Speaker 2: what that's meant is that the chip industry has improved 111 00:06:09,809 --> 00:06:14,678 Speaker 2: its capabilities more rapidly over a longer time horizon than 112 00:06:14,690 --> 00:06:17,399 Speaker 2: any other type of technology in all of human history. 113 00:06:17,410 --> 00:06:21,558 Speaker 2: Nothing else comes remotely close to that rate of technological improvement, 114 00:06:21,570 --> 00:06:24,920 Speaker 2: exponential growth for uh more than half a century. It's 115 00:06:24,928 --> 00:06:26,980 Speaker 2: something that the rest of the economy can only dream 116 00:06:26,988 --> 00:06:27,380 Speaker 2: of 117 00:06:27,709 --> 00:06:30,429 Speaker 2: in most companies if sales grow by 5% or if 118 00:06:30,440 --> 00:06:33,339 Speaker 2: technology improves by 10% it's a great success. And in 119 00:06:33,350 --> 00:06:35,829 Speaker 2: the chip industry, that's an extraordinary failure. And so III 120 00:06:35,859 --> 00:06:39,700 Speaker 2: I like to compare that to other types of veggies, aviation, 121 00:06:39,709 --> 00:06:43,579 Speaker 2: for example. Um you know what if planes flew twice 122 00:06:43,589 --> 00:06:46,549 Speaker 2: as fast every year while also being reduced in cost 123 00:06:46,559 --> 00:06:48,399 Speaker 2: on a regular basis? Will they be flying 124 00:06:48,678 --> 00:06:51,339 Speaker 2: uh many times faster than the speed of light at 125 00:06:51,350 --> 00:06:54,450 Speaker 2: this point? If that rate of growth were sustained over 126 00:06:54,459 --> 00:06:56,928 Speaker 2: half a century. And so it's it's physically impossible for 127 00:06:56,940 --> 00:07:00,019 Speaker 2: that to happen. But the chip industry has done something 128 00:07:00,209 --> 00:07:04,269 Speaker 2: similar and and when you look at the proliferation of 129 00:07:04,279 --> 00:07:08,230 Speaker 2: chips across society and across economy, that's been made possible 130 00:07:08,238 --> 00:07:11,369 Speaker 2: because Moore's law has driven down the cost of computing. 131 00:07:11,470 --> 00:07:13,549 Speaker 2: You want a data point that the first chips that 132 00:07:13,559 --> 00:07:16,630 Speaker 2: were available commercially had just four transistors on them 133 00:07:16,910 --> 00:07:18,959 Speaker 2: today. You can go to Amazon and buy a, a 134 00:07:18,970 --> 00:07:21,980 Speaker 2: thumb drive with a billion transistors or 10 billion transistors 135 00:07:21,989 --> 00:07:25,920 Speaker 2: for about $10. Which means that ballpark, the cost of 136 00:07:25,929 --> 00:07:31,290 Speaker 2: computing has fallen by roughly 1 billion fold uh since 137 00:07:31,299 --> 00:07:34,100 Speaker 2: uh since the 19 fifties. And that's all possible. Thanks 138 00:07:34,109 --> 00:07:34,760 Speaker 2: to Moore's Law, 139 00:07:35,559 --> 00:07:40,279 Speaker 1: right? I was looking up the Apple iphone 12 had 140 00:07:40,290 --> 00:07:43,970 Speaker 1: the A 14 processor, something like 10 billion I CS. 141 00:07:44,320 --> 00:07:47,920 Speaker 1: And the iphone 14 pro max has the A 16 142 00:07:47,929 --> 00:07:51,489 Speaker 1: processor with 16 billion. There's a 60% increase in the 143 00:07:51,500 --> 00:07:54,239 Speaker 1: I CS just in the latest ones. Um I wanna 144 00:07:54,250 --> 00:07:57,100 Speaker 1: talk to you later about the likelihood that this will 145 00:07:57,109 --> 00:07:59,510 Speaker 1: not go on forever and what that means. But uh 146 00:07:59,589 --> 00:08:01,510 Speaker 1: before we go in that direction, 147 00:08:01,750 --> 00:08:03,459 Speaker 1: uh Chris, you spent a lot of time in your 148 00:08:03,470 --> 00:08:06,489 Speaker 1: book talking about the immense complexity and the cost associated 149 00:08:06,500 --> 00:08:11,230 Speaker 1: with producing chips uh share with us some idea of how, 150 00:08:11,239 --> 00:08:13,019 Speaker 1: how complex and how coffee that is. 151 00:08:14,679 --> 00:08:17,970 Speaker 2: Well, to make an advanced chip, you need to make 152 00:08:17,980 --> 00:08:20,359 Speaker 2: a chip with billions of transistors, each of which is 153 00:08:20,369 --> 00:08:23,190 Speaker 2: roughly the size of a Coronavirus. So it's, it's the 154 00:08:23,200 --> 00:08:26,859 Speaker 2: most complex manufacturing humans have ever done. You think in 155 00:08:26,869 --> 00:08:31,429 Speaker 2: most manufactured goods, they have bolt tolerances measured in millimeters 156 00:08:31,440 --> 00:08:35,530 Speaker 2: or 10th of millimeters. But in the chip industry, misplacing 157 00:08:35,539 --> 00:08:40,689 Speaker 2: a single atom can at times cause catastrophic failure to 158 00:08:40,719 --> 00:08:42,309 Speaker 2: the way the semiconductor works. 159 00:08:42,405 --> 00:08:46,255 Speaker 2: So there's no other industry that combines the scale manufacturing 160 00:08:46,265 --> 00:08:50,695 Speaker 2: transistors by the billions with the level of precision required. 161 00:08:50,705 --> 00:08:53,405 Speaker 2: And and doing this is it's, it's simply the most 162 00:08:53,414 --> 00:08:55,875 Speaker 2: complex uh manufacturing that we do. And as a result, 163 00:08:55,885 --> 00:08:58,195 Speaker 2: it requires the most complex supply chain 164 00:08:58,590 --> 00:09:02,530 Speaker 2: that exists. And so today, it's fundamentally impossible, simply impossible 165 00:09:02,539 --> 00:09:06,340 Speaker 2: to create an advantage ship without drawing on machine tools, 166 00:09:06,349 --> 00:09:12,010 Speaker 2: software materials, um software from multiple different uh companies in 167 00:09:12,020 --> 00:09:15,260 Speaker 2: multiple different countries. And there are several that stand out. 168 00:09:15,270 --> 00:09:20,200 Speaker 2: Uh the US Japan, the Netherlands, uh Taiwan and Korea 169 00:09:20,210 --> 00:09:21,718 Speaker 2: are the five big players 170 00:09:22,099 --> 00:09:25,460 Speaker 2: uh in the chip industry. Uh They're the only five 171 00:09:25,469 --> 00:09:29,358 Speaker 2: countries that have um uh a substantial share of the 172 00:09:29,369 --> 00:09:32,349 Speaker 2: chip industry's revenue in double digit terms. Uh And you 173 00:09:32,359 --> 00:09:34,950 Speaker 2: can't produce a cutting edge uh chip without drawing on 174 00:09:34,960 --> 00:09:37,369 Speaker 2: expertise from uh each of those countries. 175 00:09:38,109 --> 00:09:41,049 Speaker 1: And if I am not just taken like a lithography, 176 00:09:41,059 --> 00:09:44,440 Speaker 1: the extreme ultraviolet lithography machines cost like $300 million to 177 00:09:44,450 --> 00:09:45,239 Speaker 1: make or something like that. 178 00:09:46,109 --> 00:09:48,869 Speaker 2: That, that's right. Yeah, the, the, the next generation lithography 179 00:09:48,880 --> 00:09:52,439 Speaker 2: machines will cost 300 billion or so dollars and like 180 00:09:52,450 --> 00:09:54,679 Speaker 2: a cutting edge ship making facility as a result of 181 00:09:54,690 --> 00:09:56,770 Speaker 2: all the machinery you need inside of it uh can 182 00:09:56,780 --> 00:10:01,478 Speaker 2: cost $20 billion or so the most expensive factories in 183 00:10:01,489 --> 00:10:04,349 Speaker 2: human history and 70% or so. The cost of a 184 00:10:04,359 --> 00:10:07,169 Speaker 2: new chip fab fab is what we call a manufacturing 185 00:10:07,179 --> 00:10:10,239 Speaker 2: facility is just to pay for the machine tools. So these, 186 00:10:10,369 --> 00:10:10,809 Speaker 2: these tool 187 00:10:10,903 --> 00:10:13,482 Speaker 2: tools that can lay down thin films and materials, just 188 00:10:13,492 --> 00:10:17,552 Speaker 2: four atoms thick with basically complete deformity or carve canyons 189 00:10:17,562 --> 00:10:20,002 Speaker 2: into silicon, just a couple of atoms wide. This is 190 00:10:20,013 --> 00:10:23,093 Speaker 2: the most complex machinery humans have ever made that requires 191 00:10:23,102 --> 00:10:25,513 Speaker 2: the supply chain that stretches from the Netherlands to the 192 00:10:25,523 --> 00:10:28,852 Speaker 2: US to Japan. And you can't make advanced ships without 193 00:10:28,862 --> 00:10:30,932 Speaker 2: these tools. And these tools themselves 194 00:10:31,015 --> 00:10:34,976 Speaker 2: are are not just made exclusively in these three countries. 195 00:10:34,995 --> 00:10:37,655 Speaker 2: They're also made by a tiny number of firms. Uh 196 00:10:37,666 --> 00:10:39,296 Speaker 2: And in the case of some of the tools, there's 197 00:10:39,306 --> 00:10:41,036 Speaker 2: just one firm that knows how to make them. And 198 00:10:41,046 --> 00:10:43,556 Speaker 2: so you mentioned the lithography tools when it comes to 199 00:10:43,565 --> 00:10:47,125 Speaker 2: advanced lithography, a SML and the Netherlands has 100% market 200 00:10:47,135 --> 00:10:50,296 Speaker 2: share uh of the tools that are necessary to make 201 00:10:50,306 --> 00:10:51,036 Speaker 2: cutting edge ships. 202 00:10:51,830 --> 00:10:53,659 Speaker 1: I think there's one part in your book, you have 203 00:10:53,669 --> 00:10:56,409 Speaker 1: this amazing analogy. As you say that the Carl Zeiss 204 00:10:56,609 --> 00:11:00,440 Speaker 1: lens or mirror that is required to reflect the laser 205 00:11:00,580 --> 00:11:03,770 Speaker 1: in the in the uh fat process for Eu V. 206 00:11:03,919 --> 00:11:06,289 Speaker 1: It's like uh the, the perfection in that is like, 207 00:11:06,299 --> 00:11:08,780 Speaker 1: you know, trying to hit a golf ball on the 208 00:11:08,789 --> 00:11:10,789 Speaker 1: moon from here like the laser has to be that 209 00:11:10,799 --> 00:11:12,320 Speaker 1: accurate or the pure 210 00:11:12,419 --> 00:11:15,010 Speaker 1: on the class has to be like if there's one 211 00:11:15,020 --> 00:11:18,429 Speaker 1: millimeter of impurity, it, it's like the size of Germany, 212 00:11:18,440 --> 00:11:20,949 Speaker 1: that's the size of the mirror and that's a tolerance level, 213 00:11:20,960 --> 00:11:23,210 Speaker 1: which is absolutely mind boggling. But I, I really like 214 00:11:23,219 --> 00:11:25,250 Speaker 1: the way you portray it because I think we sort 215 00:11:25,260 --> 00:11:29,489 Speaker 1: of understand the word complexity, we're really understanding it. And 216 00:11:29,500 --> 00:11:31,369 Speaker 1: I think that sort of analogy really, really helped. Uh 217 00:11:31,380 --> 00:11:32,929 Speaker 1: definitely it helped me. 218 00:11:33,200 --> 00:11:37,780 Speaker 1: Um OK, so it's a truly globalized process, all sorts 219 00:11:37,789 --> 00:11:41,549 Speaker 1: of companies have very specific and world dominating roles in it. 220 00:11:41,729 --> 00:11:43,799 Speaker 1: So I take it you don't believe the Chinese can 221 00:11:43,809 --> 00:11:44,789 Speaker 1: do it all by themselves. 222 00:11:47,049 --> 00:11:48,919 Speaker 2: Well, I think it depends on, on what we mean. If, 223 00:11:48,929 --> 00:11:53,460 Speaker 2: if we mean can China produce 2023 levels of technology 224 00:11:54,020 --> 00:11:56,679 Speaker 2: self sufficiently at some point in the future? I think 225 00:11:56,690 --> 00:11:58,239 Speaker 2: the answer is yes, they will be able to, the 226 00:11:58,250 --> 00:12:00,978 Speaker 2: question is how far in the future will that be possible? 227 00:12:01,630 --> 00:12:06,159 Speaker 2: And because the industry races forward very rapidly. If any 228 00:12:06,169 --> 00:12:09,000 Speaker 2: country catches up to the cutting edge in 20 years time, 229 00:12:09,010 --> 00:12:11,659 Speaker 2: it's a meaningless accomplishment because the cutting edge will be 230 00:12:11,669 --> 00:12:15,099 Speaker 2: so far uh so far advanced. So I I think 231 00:12:15,109 --> 00:12:19,250 Speaker 2: answering your question requires specifying the time horizon. Do I 232 00:12:19,260 --> 00:12:24,419 Speaker 2: think China can indigenously develop cutting edge capabilities soon using 233 00:12:24,429 --> 00:12:27,640 Speaker 2: solely domestically produced technology? I doubt it. Uh just because 234 00:12:27,650 --> 00:12:29,419 Speaker 2: I don't think any country can do it uh 235 00:12:30,164 --> 00:12:32,864 Speaker 2: soon. Uh And you know, if I look at the 236 00:12:32,875 --> 00:12:37,094 Speaker 2: largest producers of uh in the semi hunter industry, the US, 237 00:12:37,104 --> 00:12:41,723 Speaker 2: uh uh Taiwan, Japan, Korea, none of them have any 238 00:12:41,734 --> 00:12:44,664 Speaker 2: sort of remotely plausible pathway over the next 10 years 239 00:12:44,674 --> 00:12:47,304 Speaker 2: to be completely self sufficient ship making. And so it 240 00:12:47,315 --> 00:12:49,674 Speaker 2: seems highly implausible to me that China which is today 241 00:12:49,684 --> 00:12:52,825 Speaker 2: still a small player in revenue terms, um, has a 242 00:12:52,835 --> 00:12:54,164 Speaker 2: plausible pathway of its own. 243 00:12:54,820 --> 00:12:56,869 Speaker 1: I mean, this may be an unfair question because this 244 00:12:56,880 --> 00:12:59,669 Speaker 1: happened just like last week. So the made 60 phone 245 00:12:59,679 --> 00:13:01,739 Speaker 1: that Huawei came up with and just a couple of 246 00:13:01,750 --> 00:13:03,809 Speaker 1: days ago, I saw that they're also increasing some foldable 247 00:13:03,820 --> 00:13:06,799 Speaker 1: phone and these seem to have nine nanometer chips and 248 00:13:06,809 --> 00:13:10,699 Speaker 1: then made by Ken uh or a K nine process 249 00:13:10,710 --> 00:13:14,109 Speaker 1: or something like that. Um Is it still like a 250 00:13:14,119 --> 00:13:16,349 Speaker 1: made in China phone or there is still a huge 251 00:13:16,359 --> 00:13:19,429 Speaker 1: amount of design and other elements that are coming from 252 00:13:19,440 --> 00:13:21,130 Speaker 1: a globalized supply chain process? 253 00:13:22,820 --> 00:13:25,289 Speaker 2: Well, we're still learning more about the the chip inside 254 00:13:25,299 --> 00:13:27,460 Speaker 2: of the Huawei phone. I think what we can pretty 255 00:13:27,469 --> 00:13:30,439 Speaker 2: safely assume thus far is that uh it's clear that 256 00:13:30,450 --> 00:13:33,469 Speaker 2: the pro the manufacturing process used to manufacture the chip 257 00:13:33,479 --> 00:13:35,460 Speaker 2: is what's called a seven nanometer process. So this is 258 00:13:35,469 --> 00:13:39,130 Speaker 2: a process that um TS MC first brought online 259 00:13:39,400 --> 00:13:43,960 Speaker 2: uh about five years ago uh in Taiwan. Um It's 260 00:13:43,969 --> 00:13:47,229 Speaker 2: a process that is now used in in mid and 261 00:13:47,239 --> 00:13:51,049 Speaker 2: low range smartphones, newer smartphones will have uh tsm CS 262 00:13:51,059 --> 00:13:53,950 Speaker 2: most advanced process, uh five nanometers 263 00:13:54,419 --> 00:13:58,640 Speaker 2: uh in them. Um We don't know about the efficiency 264 00:13:58,650 --> 00:14:02,699 Speaker 2: of the manufacturing yet. So Huawei's manufacturing partner, Smi um 265 00:14:02,710 --> 00:14:04,590 Speaker 2: hasn't released any data nor do I think they will 266 00:14:04,599 --> 00:14:07,478 Speaker 2: any time soon. So uh whether this chip is being 267 00:14:07,489 --> 00:14:11,939 Speaker 2: produced at huge inefficiency or relatively efficient, we're still going 268 00:14:11,950 --> 00:14:14,179 Speaker 2: to find out um what we do know is that it, 269 00:14:14,190 --> 00:14:18,020 Speaker 2: it was uh produced largely using imported tools. Um because 270 00:14:18,030 --> 00:14:20,960 Speaker 2: until recently, SMI was able to buy the most 271 00:14:21,039 --> 00:14:23,260 Speaker 2: advanced as well as the sort of second generation most 272 00:14:23,270 --> 00:14:26,690 Speaker 2: advanced tools from the US, from Japan uh and from 273 00:14:26,700 --> 00:14:29,570 Speaker 2: the Netherlands. Uh And so it had purchased a large 274 00:14:29,580 --> 00:14:31,330 Speaker 2: quantity of those tools and is now still able to 275 00:14:31,340 --> 00:14:34,320 Speaker 2: use uh at least some of those. Uh and this 276 00:14:34,330 --> 00:14:36,890 Speaker 2: ship was made largely using those tools. So, you know, 277 00:14:36,900 --> 00:14:40,010 Speaker 2: in some ways, it's, it's certainly uh evidence of incremental 278 00:14:40,020 --> 00:14:42,669 Speaker 2: progress in the ship industry. It's probably not evidence of 279 00:14:42,679 --> 00:14:45,750 Speaker 2: catching up because TS MC is also moving forward. And 280 00:14:45,760 --> 00:14:47,750 Speaker 2: so the gap between TS MC and 281 00:14:47,989 --> 00:14:52,630 Speaker 2: China's capabilities, uh it doesn't appear to be decreasing. Um 282 00:14:53,080 --> 00:14:55,780 Speaker 2: Is it e evidence of self sufficiency for China? Not 283 00:14:55,789 --> 00:14:59,599 Speaker 2: really because again, the machines that were producing these chips 284 00:14:59,609 --> 00:15:04,340 Speaker 2: were uh largely imported uh machines. So it's, it's, it's 285 00:15:04,349 --> 00:15:06,400 Speaker 2: progress of a sort, but I, I wouldn't call it 286 00:15:06,409 --> 00:15:06,900 Speaker 2: a breakthrough. 287 00:15:07,659 --> 00:15:09,440 Speaker 1: Tell me a little bit about the term that you 288 00:15:09,450 --> 00:15:12,669 Speaker 1: use inefficiency or efficiency. So I think there the yield 289 00:15:12,679 --> 00:15:14,909 Speaker 1: of semiconductor is like a very important part of the 290 00:15:15,169 --> 00:15:17,780 Speaker 1: bottom line of any manufacturing process. So give us a 291 00:15:17,789 --> 00:15:20,150 Speaker 1: sense of, you know what kind of yield is needed 292 00:15:20,159 --> 00:15:22,840 Speaker 1: and how complex it is the process of getting decently 293 00:15:22,849 --> 00:15:25,010 Speaker 1: efficient yield out of a semiconductor production. 294 00:15:26,070 --> 00:15:27,989 Speaker 2: Yeah, you yield is the key variable in it. That 295 00:15:28,000 --> 00:15:29,840 Speaker 2: that just means the share of chips you produce that 296 00:15:29,849 --> 00:15:34,090 Speaker 2: actually work. Um And because chips are so complex at 297 00:15:34,099 --> 00:15:37,539 Speaker 2: the cutting edge, uh it is very, very, very easy 298 00:15:37,549 --> 00:15:40,000 Speaker 2: to make errors in the manufacturing process that cause a 299 00:15:40,010 --> 00:15:42,760 Speaker 2: chip not to work. And there are ways you can 300 00:15:42,770 --> 00:15:45,109 Speaker 2: try to make your chip more or less resilient to 301 00:15:45,119 --> 00:15:48,619 Speaker 2: having errors. But whenever any company brings online, a new 302 00:15:48,695 --> 00:15:52,645 Speaker 2: manufacturing process more advanced with smaller transistors, there's always a 303 00:15:52,655 --> 00:15:56,044 Speaker 2: very uh significant learning curve that has to be overcome 304 00:15:56,174 --> 00:15:59,304 Speaker 2: and yields always start low and then they're always intended 305 00:15:59,315 --> 00:16:03,385 Speaker 2: to end uh very high in the 90% level um 306 00:16:03,395 --> 00:16:08,565 Speaker 2: uh as close to 100 as possible. Um And the, the, 307 00:16:08,575 --> 00:16:11,905 Speaker 2: the pro the process of improving yields is therefore critical 308 00:16:11,914 --> 00:16:15,895 Speaker 2: to financial viability for companies. And so a key question 309 00:16:15,905 --> 00:16:17,065 Speaker 2: in your ability to 310 00:16:17,320 --> 00:16:20,130 Speaker 2: price uh effectively, if you're a chip maker is what 311 00:16:20,140 --> 00:16:23,210 Speaker 2: yield you could produce at. Because if you're producing a 50% yield. 312 00:16:23,609 --> 00:16:26,729 Speaker 2: In other words, you gotta spend um amount of, you know, you, 313 00:16:26,739 --> 00:16:29,400 Speaker 2: you need to, you need to, to fund the production 314 00:16:29,409 --> 00:16:31,059 Speaker 2: of twice as many chips as you actually get. In 315 00:16:31,070 --> 00:16:33,080 Speaker 2: terms of working chips, you're gonna be a lot less 316 00:16:33,090 --> 00:16:36,710 Speaker 2: competitive than a company with 90% yield. Uh And so we, 317 00:16:36,719 --> 00:16:39,409 Speaker 2: we don't have good data uh about yields from companies. 318 00:16:39,419 --> 00:16:42,349 Speaker 2: They're very secretive about their yield data. Um But we 319 00:16:42,359 --> 00:16:44,190 Speaker 2: know that the the leading chip makers 320 00:16:44,284 --> 00:16:47,065 Speaker 2: uh end up with very high yields uh in the 90% 321 00:16:47,075 --> 00:16:51,054 Speaker 2: range um for their, their, their mature processes. And so 322 00:16:51,065 --> 00:16:53,534 Speaker 2: if you look at the new Huawei phone, you know, 323 00:16:53,544 --> 00:16:56,205 Speaker 2: a fundamental question is, what's the yield? And I don't 324 00:16:56,215 --> 00:16:57,484 Speaker 2: think we're gonna know the answer to that question any 325 00:16:57,494 --> 00:17:00,994 Speaker 2: time soon. But any assessment of how viable a processes 326 00:17:01,005 --> 00:17:03,494 Speaker 2: is or not has to uh has to try to 327 00:17:03,505 --> 00:17:04,954 Speaker 2: get us a sense of the yield. 328 00:17:05,680 --> 00:17:08,930 Speaker 1: So Chris, you mentioned that the Huawei phone uh could have, 329 00:17:08,939 --> 00:17:12,609 Speaker 1: you know, homegrown processors, but the machine tools, the critical 330 00:17:12,619 --> 00:17:17,400 Speaker 1: component of preparing that semiconductor are all imported. Um So 331 00:17:17,410 --> 00:17:21,310 Speaker 1: give us a sense of state of sanctions in play 332 00:17:21,319 --> 00:17:25,119 Speaker 1: right now. Are we saying that China is restricted from 333 00:17:25,130 --> 00:17:30,199 Speaker 1: importing any lithography machines going forward or is the idea that, 334 00:17:30,209 --> 00:17:32,150 Speaker 1: you know, they'll just get stuck in whatever they were 335 00:17:32,160 --> 00:17:33,889 Speaker 1: allowed to import and 336 00:17:34,400 --> 00:17:37,010 Speaker 1: they can keep on importing that but nothing else after that. 337 00:17:38,689 --> 00:17:41,489 Speaker 2: So over the past uh year, the US, Japan and 338 00:17:41,500 --> 00:17:44,449 Speaker 2: the Netherlands have each released a set of controls which 339 00:17:44,459 --> 00:17:49,000 Speaker 2: have uh basically tried to stop China from buying new 340 00:17:49,010 --> 00:17:52,699 Speaker 2: tools that could be used at sub 16 or 14 341 00:17:52,709 --> 00:17:55,959 Speaker 2: nanometer production. So China has been able in the pass 342 00:17:55,969 --> 00:17:59,119 Speaker 2: to buy lots of tools and has bought huge volumes 343 00:17:59,130 --> 00:18:02,479 Speaker 2: of tools that used for 14 nanometers, 10 nanometers and 344 00:18:02,489 --> 00:18:04,489 Speaker 2: seven is the the note after that which is the 345 00:18:04,500 --> 00:18:07,319 Speaker 2: note that the Huawei phone was uh produced on 346 00:18:07,680 --> 00:18:10,540 Speaker 2: and so until recently and actually the the Dutch regulations 347 00:18:10,550 --> 00:18:12,410 Speaker 2: don't come into force until the end of this year. 348 00:18:12,420 --> 00:18:15,849 Speaker 2: So it's still the case that um that Chinese buyers 349 00:18:15,859 --> 00:18:18,800 Speaker 2: can buy uh certain tools from the Netherlands. But once 350 00:18:18,810 --> 00:18:21,979 Speaker 2: all these restrictions are enforced, the uh the the net 351 00:18:21,989 --> 00:18:26,079 Speaker 2: effect will be to limit China's purchase of Chinese firms 352 00:18:26,089 --> 00:18:28,849 Speaker 2: purchase of any new tools that can be used primarily 353 00:18:28,859 --> 00:18:30,670 Speaker 2: for sub 14 nanometer 354 00:18:31,099 --> 00:18:36,479 Speaker 2: uh production. Now most chips by volume are actually produced 355 00:18:36,489 --> 00:18:40,198 Speaker 2: at larger geometries. In other words, older manufacturing technologies at 356 00:18:40,209 --> 00:18:44,130 Speaker 2: 40 nanometers or 100 and 30 nanometers. But in dollar terms, 357 00:18:44,140 --> 00:18:46,339 Speaker 2: the most expensive chips are the ones that are produced 358 00:18:46,349 --> 00:18:49,729 Speaker 2: at more advanced technologies. And so inside your smart, 359 00:18:49,800 --> 00:18:53,119 Speaker 2: the main application processor is produced at something pretty close 360 00:18:53,130 --> 00:18:56,659 Speaker 2: to cutting edge inside your PC. Inside data centers, the 361 00:18:56,670 --> 00:18:59,609 Speaker 2: key chips are uh are are are also pretty close 362 00:18:59,619 --> 00:19:00,719 Speaker 2: to the cutting edge. And so for a lot of 363 00:19:00,729 --> 00:19:04,859 Speaker 2: key technology areas, there are huge performance gains to be 364 00:19:04,869 --> 00:19:07,589 Speaker 2: had by producing on a cutting edge technology or something 365 00:19:07,599 --> 00:19:08,430 Speaker 2: very close to it. 366 00:19:09,619 --> 00:19:13,310 Speaker 1: So the implications of this could be rather dire for 367 00:19:13,319 --> 00:19:16,449 Speaker 1: China's medium term outlook. So for the time being, they 368 00:19:16,459 --> 00:19:19,489 Speaker 1: can produce some chips and get on with it. But 369 00:19:19,500 --> 00:19:20,630 Speaker 1: in terms of pushing 370 00:19:21,439 --> 00:19:24,310 Speaker 1: gps that you need to train A I models or 371 00:19:24,319 --> 00:19:26,150 Speaker 1: if you want to run the next generation of data 372 00:19:26,160 --> 00:19:30,188 Speaker 1: centers or develop the next generation of weapons, um they're 373 00:19:30,199 --> 00:19:32,409 Speaker 1: basically doomed to failure. 374 00:19:34,319 --> 00:19:36,790 Speaker 2: Well, II I think I would, I would differentiate first 375 00:19:36,800 --> 00:19:40,510 Speaker 2: between the the restrictions on tools used to produce ships 376 00:19:40,520 --> 00:19:42,159 Speaker 2: below 14 nanometers 377 00:19:42,579 --> 00:19:45,750 Speaker 2: um and the restrictions on specific chips themselves. So the US, 378 00:19:45,760 --> 00:19:48,849 Speaker 2: Japan and the Netherlands have restricted tools used for sub 379 00:19:48,869 --> 00:19:52,739 Speaker 2: 14 nanometers, but they haven't restricted the transfer of chips 380 00:19:52,750 --> 00:19:56,020 Speaker 2: to China, sub 14 nanometers. And so today, China imports 381 00:19:56,030 --> 00:19:59,229 Speaker 2: almost all of its sub 14 nanometer chips from Taiwan, 382 00:19:59,239 --> 00:20:02,949 Speaker 2: from Korea, from the US uh from from elsewhere, which 383 00:20:02,959 --> 00:20:05,560 Speaker 2: is the reason that China spends as much money each 384 00:20:05,569 --> 00:20:07,369 Speaker 2: year importing chips, it spends importing oil, 385 00:20:07,449 --> 00:20:11,159 Speaker 2: it's largely smartphone processors, PC processors. And so it's actually 386 00:20:11,170 --> 00:20:15,060 Speaker 2: not much of a change to China's chip buying patterns 387 00:20:15,150 --> 00:20:17,800 Speaker 2: to have it keep buying chips from abroad. Um In 388 00:20:17,810 --> 00:20:21,050 Speaker 2: so far as the regulations uh by limiting China's domestic 389 00:20:21,060 --> 00:20:24,540 Speaker 2: production force China to buy chips uh abroad. There are 390 00:20:24,550 --> 00:20:28,169 Speaker 2: also as you mentioned, specific US rules that ban the 391 00:20:28,180 --> 00:20:31,089 Speaker 2: transfer of GP US or chips used to train A 392 00:20:31,099 --> 00:20:32,239 Speaker 2: I systems to China. 393 00:20:32,560 --> 00:20:35,750 Speaker 2: Uh but these are very, very specific rules governing just 394 00:20:35,760 --> 00:20:38,409 Speaker 2: one type of chip. And so the US has via 395 00:20:38,420 --> 00:20:40,989 Speaker 2: its rules kept wide open the transfer of any chip 396 00:20:41,000 --> 00:20:44,479 Speaker 2: for smartphones, any chip for autos, any chip for P CS. 397 00:20:44,930 --> 00:20:48,270 Speaker 2: Um almost all types of chips except GP US. So 398 00:20:48,280 --> 00:20:52,089 Speaker 2: China can still buy very advanced ships from abroad for smartphones. 399 00:20:52,270 --> 00:20:54,910 Speaker 2: The only restriction is when it comes to the chips 400 00:20:54,920 --> 00:20:57,140 Speaker 2: that are used to train A I systems. And those 401 00:20:57,150 --> 00:20:59,229 Speaker 2: are the chips called GP US that are largely produced 402 00:20:59,239 --> 00:20:59,949 Speaker 2: by NVIDIA. 403 00:21:00,680 --> 00:21:03,310 Speaker 1: Would the state of sanctions remain where they are or 404 00:21:03,319 --> 00:21:05,329 Speaker 1: the net is about to get tighter. 405 00:21:06,890 --> 00:21:10,079 Speaker 2: Well, I think a lot depends on um first off 406 00:21:10,089 --> 00:21:12,520 Speaker 2: uh the direction of of us politics US has an 407 00:21:12,530 --> 00:21:16,599 Speaker 2: election uh next year which could bring policy change. Um 408 00:21:16,609 --> 00:21:21,020 Speaker 2: but second Chinese policy um as well, I think, you know, 409 00:21:21,030 --> 00:21:23,760 Speaker 2: one of the key questions, the key uncertainties is 410 00:21:24,000 --> 00:21:27,560 Speaker 2: what cost China is willing to bear to pursue its 411 00:21:27,569 --> 00:21:31,510 Speaker 2: self-sufficiency drive. And we've seen the Chinese government spend tens 412 00:21:31,520 --> 00:21:34,959 Speaker 2: of billions of dollars a year from 2014 to present 413 00:21:34,989 --> 00:21:39,819 Speaker 2: towards self sufficiency. I think uh that rate will certainly 414 00:21:39,829 --> 00:21:42,229 Speaker 2: continue it. But I wonder if it will even intensify 415 00:21:42,239 --> 00:21:44,000 Speaker 2: and if China is willing to spend more money on 416 00:21:44,010 --> 00:21:47,810 Speaker 2: self sufficiency. In other words, produce more chips at a 417 00:21:47,819 --> 00:21:51,089 Speaker 2: very inefficient manner domestically, you could end up buying fewer 418 00:21:51,099 --> 00:21:52,109 Speaker 2: chips from abroad. 419 00:21:52,400 --> 00:21:54,399 Speaker 2: And if China starts buying fewer chips from abroad, it 420 00:21:54,410 --> 00:21:56,890 Speaker 2: also loses some of its leverage because right now, one 421 00:21:56,900 --> 00:22:00,750 Speaker 2: of the factors limiting foreign restrictions on China is that 422 00:22:00,760 --> 00:22:02,770 Speaker 2: foreign and chip firms still sell a lot of chips 423 00:22:02,780 --> 00:22:05,688 Speaker 2: to China. But if they begin to be restricted from 424 00:22:05,699 --> 00:22:09,510 Speaker 2: market access by China's self sufficiency drive, then you'll have 425 00:22:09,520 --> 00:22:14,089 Speaker 2: less lobbying in Washington, in Tokyo uh in European capital 426 00:22:14,099 --> 00:22:16,900 Speaker 2: saying be careful with your restrictions, you're gonna hurt our 427 00:22:16,910 --> 00:22:20,040 Speaker 2: market because if that market's already declining, that'll become much 428 00:22:20,050 --> 00:22:20,869 Speaker 2: less powerful 429 00:22:21,229 --> 00:22:23,800 Speaker 2: argument. And that's I think a very tricky balance for China. 430 00:22:23,810 --> 00:22:26,260 Speaker 2: It wants self sufficiency for sure, but it also values 431 00:22:26,270 --> 00:22:29,030 Speaker 2: the fact that because its market is valuable, it gives 432 00:22:29,040 --> 00:22:32,109 Speaker 2: it some bargaining power in its relations with the West. 433 00:22:32,459 --> 00:22:35,780 Speaker 1: It's, it's really fascinating, right? So if I don't have 434 00:22:35,790 --> 00:22:40,319 Speaker 1: an advanced domestic industry, interdependence remains uh but the moment 435 00:22:40,329 --> 00:22:43,109 Speaker 1: I progress interdependence, we can and maybe there is sanctions 436 00:22:43,119 --> 00:22:45,938 Speaker 1: at that point, but it's too late because I am independent. 437 00:22:46,140 --> 00:22:51,389 Speaker 1: Um Chris, your book showcases uh TSM CS singular role 438 00:22:51,400 --> 00:22:55,188 Speaker 1: in the manufacturing of the world's most advanced chips. Um 439 00:22:55,199 --> 00:22:57,530 Speaker 1: And you mentioned earlier as well that there are other 440 00:22:57,540 --> 00:23:00,709 Speaker 1: single nodes of vulnerability. A smo being another one. So 441 00:23:00,719 --> 00:23:04,140 Speaker 1: how risky it is. Is it for like one company 442 00:23:04,150 --> 00:23:07,079 Speaker 1: at this day and age being responsible for 100% of 443 00:23:07,089 --> 00:23:09,290 Speaker 1: something that the world needs critically? 444 00:23:11,400 --> 00:23:13,939 Speaker 2: Well, I don't think anyone designing a supply chain would 445 00:23:13,949 --> 00:23:18,750 Speaker 2: have opted in uh to a system of complete vulnerability, 446 00:23:18,760 --> 00:23:21,089 Speaker 2: complete reliance on a single firm. Like we're all reliant, 447 00:23:21,099 --> 00:23:24,489 Speaker 2: for example, on a S MLS lithography machines, I think 448 00:23:24,500 --> 00:23:27,219 Speaker 2: everyone would prefer a competitive market. Certainly the buyers of 449 00:23:27,229 --> 00:23:27,750 Speaker 2: lithography 450 00:23:27,829 --> 00:23:31,400 Speaker 2: machines would prefer a competitive market right now. They're price takers. 451 00:23:31,430 --> 00:23:36,069 Speaker 2: Um But I think there's particular concern around uh TS 452 00:23:36,079 --> 00:23:40,379 Speaker 2: MC because of the risks to Taiwan. Uh that I 453 00:23:40,390 --> 00:23:44,050 Speaker 2: think most people assess have grown somewhat in, in recent years. 454 00:23:44,500 --> 00:23:48,910 Speaker 2: The, the, the Taiwan question I think is, is important, 455 00:23:48,920 --> 00:23:52,380 Speaker 2: not just because of TSM CS dynamic as being the 456 00:23:52,390 --> 00:23:55,679 Speaker 2: central player of in producing advanced ships, but also because 457 00:23:55,729 --> 00:23:59,390 Speaker 2: it has become this critical producer of advanced ships precisely 458 00:23:59,400 --> 00:24:04,050 Speaker 2: as the us ability to credibly defend Taiwan has deteriorated. 459 00:24:04,780 --> 00:24:07,290 Speaker 2: I think that's the second half dynamic. It's not just 460 00:24:07,300 --> 00:24:09,380 Speaker 2: a chip story, it's also a story of the military 461 00:24:09,390 --> 00:24:12,400 Speaker 2: balance shifting in Taiwan's favor. And so for most of 462 00:24:12,410 --> 00:24:17,760 Speaker 2: the last 50 years, uh tech executives, investors, um electronics 463 00:24:17,770 --> 00:24:19,849 Speaker 2: firms didn't have to think that hard about the military 464 00:24:19,859 --> 00:24:21,930 Speaker 2: balance because it was obvious if there were a war, 465 00:24:21,939 --> 00:24:24,780 Speaker 2: everyone knew who would win, everyone knew which would win. 466 00:24:24,790 --> 00:24:26,859 Speaker 2: Beijing and Washington and Taipei everywhere. 467 00:24:27,150 --> 00:24:28,780 Speaker 2: Today. If there were a war, there's a lot of 468 00:24:28,790 --> 00:24:31,219 Speaker 2: uncertainty about what the outcome would be. And I think 469 00:24:31,229 --> 00:24:36,520 Speaker 2: that has inspired uh an increased sense of uncertainty about 470 00:24:36,530 --> 00:24:39,790 Speaker 2: whether China might be willing to escalate. And if so 471 00:24:39,800 --> 00:24:41,739 Speaker 2: what the ramifications would be for the supply chain. And 472 00:24:41,750 --> 00:24:44,510 Speaker 2: so that's really the reason why we're talking about uh 473 00:24:44,520 --> 00:24:49,119 Speaker 2: Taiwan risk is that the military balance has shifted dramatically 474 00:24:49,130 --> 00:24:49,938 Speaker 2: in China's favor. 475 00:24:51,040 --> 00:24:53,319 Speaker 1: Right. Yeah, I don't see anybody forcing a SML to 476 00:24:53,329 --> 00:24:57,369 Speaker 1: start making lithography missions elsewhere because is not seen as 477 00:24:57,410 --> 00:24:59,790 Speaker 1: a note of vulnerability out. There are these, you know, 478 00:24:59,800 --> 00:25:02,099 Speaker 1: terrific chapters in your book where you sort of talk 479 00:25:02,109 --> 00:25:05,790 Speaker 1: about the war game scenarios around which China Taiwan, uh 480 00:25:05,800 --> 00:25:08,209 Speaker 1: the situation can unravel. And I think that was also 481 00:25:08,219 --> 00:25:11,339 Speaker 1: very instructive. Ok. I want to come back to something 482 00:25:11,349 --> 00:25:14,469 Speaker 1: we talked about earlier, Gordon Moore and his incredible Christian 483 00:25:14,479 --> 00:25:17,290 Speaker 1: call 55 years ago that, you know, 484 00:25:17,550 --> 00:25:21,500 Speaker 1: chip speed processing speed can double 18 to 24 months. 485 00:25:21,599 --> 00:25:23,319 Speaker 1: How long can that last Chris? 486 00:25:26,020 --> 00:25:28,310 Speaker 2: Well, I, nobody knows the answer to that question. Even 487 00:25:28,319 --> 00:25:30,949 Speaker 2: the smartest people in the industry are just speculating. I 488 00:25:30,959 --> 00:25:32,489 Speaker 2: think we can say a couple of things. 489 00:25:33,290 --> 00:25:38,160 Speaker 2: One is that since the 19 eighties, Moore's law's death 490 00:25:38,170 --> 00:25:40,800 Speaker 2: has been regularly predicted and it's always been wrong. 491 00:25:41,609 --> 00:25:44,479 Speaker 2: In the 19 eighties. There were very distinguished computer uh 492 00:25:44,489 --> 00:25:48,209 Speaker 2: engineers who were highly confident that Morris Law would come 493 00:25:48,219 --> 00:25:50,319 Speaker 2: to an end that they were completely wrong in the 494 00:25:50,329 --> 00:25:53,650 Speaker 2: early two thousands. Gordon Moore himself said that he couldn't 495 00:25:53,660 --> 00:25:57,219 Speaker 2: fathom ways that the industry could keep Mori's law alive 496 00:25:57,229 --> 00:25:58,218 Speaker 2: for more than 497 00:25:58,560 --> 00:26:01,060 Speaker 2: and here we are two decades later. Uh and it's 498 00:26:01,069 --> 00:26:04,139 Speaker 2: still alive. So I think we should be very cognizant 499 00:26:04,150 --> 00:26:06,979 Speaker 2: of the fact that past predictions of Moore's Law's failure 500 00:26:06,989 --> 00:26:10,760 Speaker 2: have always been wrong. Uh And that should I think 501 00:26:10,770 --> 00:26:12,698 Speaker 2: sensitize us to the fact that there are lots of 502 00:26:12,709 --> 00:26:14,859 Speaker 2: smart people trying to find ways to keep it alive. 503 00:26:15,280 --> 00:26:17,400 Speaker 2: I think also it's the case that we, we have 504 00:26:17,410 --> 00:26:19,099 Speaker 2: a pretty clear line of sight. 505 00:26:19,319 --> 00:26:21,369 Speaker 2: If you look at the road maps uh that have 506 00:26:21,380 --> 00:26:24,290 Speaker 2: been put out by TS MC or Intel uh over 507 00:26:24,300 --> 00:26:26,770 Speaker 2: the next decade, how we're gonna get at least another 508 00:26:26,780 --> 00:26:31,150 Speaker 2: decade of transistor shrinkage beyond that point, nobody knows. But 509 00:26:31,160 --> 00:26:33,770 Speaker 2: it's always been the case that more than a decade out, 510 00:26:33,780 --> 00:26:36,760 Speaker 2: nobody knows. It's just too far out to, to, to 511 00:26:36,770 --> 00:26:37,869 Speaker 2: realistically plan, 512 00:26:38,510 --> 00:26:41,650 Speaker 1: right? Certainly will be the place. No, no, please go ahead. 513 00:26:42,560 --> 00:26:44,849 Speaker 2: Well, II I it, it certainly will be the case 514 00:26:44,859 --> 00:26:47,469 Speaker 2: that at some point you can't string transistors any further. 515 00:26:47,479 --> 00:26:50,468 Speaker 2: We're already measuring them in terms of nanometers. Uh Soon 516 00:26:50,479 --> 00:26:53,729 Speaker 2: we'll be in the Angstrom era uh defining them in 517 00:26:53,739 --> 00:26:56,130 Speaker 2: terms of angst drums. We'll have to invent I think 518 00:26:56,140 --> 00:26:59,859 Speaker 2: new terminology to measure the uh minute distances by the 519 00:26:59,869 --> 00:27:01,819 Speaker 2: end of the decade. Uh But at some point, we'll 520 00:27:01,829 --> 00:27:04,209 Speaker 2: hit real physical limits. But even at that point, I 521 00:27:04,219 --> 00:27:06,890 Speaker 2: think the the past decade has actually given us a 522 00:27:06,900 --> 00:27:09,310 Speaker 2: lot of new technique 523 00:27:09,400 --> 00:27:14,359 Speaker 2: to provide better semiconductors um with the same size transistors. So, 524 00:27:14,369 --> 00:27:17,189 Speaker 2: for example, there's new ways to design semiconductors and GP 525 00:27:17,199 --> 00:27:19,290 Speaker 2: U is the best example. What makes GP U a 526 00:27:19,300 --> 00:27:21,520 Speaker 2: great chip for A I is not that the transistors 527 00:27:21,530 --> 00:27:23,750 Speaker 2: are any smaller than the chip inside of your iphone. 528 00:27:23,760 --> 00:27:25,569 Speaker 2: It's that they, they're designed in a different way. And 529 00:27:25,579 --> 00:27:27,910 Speaker 2: so I think we're still in the very early stages 530 00:27:27,920 --> 00:27:32,280 Speaker 2: of exploring ways chip design can produce uh performance enhancements. 531 00:27:32,459 --> 00:27:34,449 Speaker 2: And then second, there's a whole new trend in the 532 00:27:34,459 --> 00:27:36,349 Speaker 2: industry called advanced packaging, 533 00:27:36,790 --> 00:27:39,219 Speaker 2: which sounds very boring, but actually, it is going to 534 00:27:39,229 --> 00:27:41,969 Speaker 2: be revolutionary in terms of letting us put different types 535 00:27:41,979 --> 00:27:45,000 Speaker 2: of chips together in different ways, which will provide huge 536 00:27:45,010 --> 00:27:48,550 Speaker 2: performance improvements in terms of speed of data transfer, in 537 00:27:48,560 --> 00:27:52,489 Speaker 2: terms of managing heat dissipation, huge issue when you've got 538 00:27:52,500 --> 00:27:53,979 Speaker 2: lots of electrons flying around. 539 00:27:54,329 --> 00:27:56,900 Speaker 2: Um and also in terms of cost, letting you combine, 540 00:27:56,910 --> 00:27:58,780 Speaker 2: for example, a low cost ship and a high cost 541 00:27:58,790 --> 00:28:02,280 Speaker 2: ship in a way that brings down your overall device cost. 542 00:28:02,290 --> 00:28:04,459 Speaker 2: And so we're in the early stages of of this 543 00:28:04,469 --> 00:28:06,329 Speaker 2: trend as well. And so when I look at the 544 00:28:06,339 --> 00:28:10,099 Speaker 2: trajectory we have of at least another decade of shrinking 545 00:28:10,109 --> 00:28:13,859 Speaker 2: transistors further lost the design renaissance is underway. Plus the 546 00:28:13,869 --> 00:28:17,140 Speaker 2: advanced packaging shifts. I'm pretty optimistic that we've got a 547 00:28:17,150 --> 00:28:19,599 Speaker 2: long pathway for further advances. 548 00:28:20,569 --> 00:28:23,530 Speaker 1: So even if the exponential increase in the number of 549 00:28:23,540 --> 00:28:27,399 Speaker 1: transistors per chip stops, we can probably have computing power 550 00:28:27,410 --> 00:28:29,380 Speaker 1: per chip continue to increase. 551 00:28:29,739 --> 00:28:31,219 Speaker 2: And that's right. That's exactly right. 552 00:28:31,709 --> 00:28:33,880 Speaker 1: Right. You know, you thanks to your book. I mean, 553 00:28:33,890 --> 00:28:35,650 Speaker 1: I'm doing stuff that I've never done before. I now 554 00:28:35,660 --> 00:28:38,219 Speaker 1: go to the IBM website and they have this newsletter 555 00:28:38,229 --> 00:28:42,020 Speaker 1: from their scientists about to your point about chip design 556 00:28:42,180 --> 00:28:45,229 Speaker 1: and issues related heat dissipation. So, yeah. Um I have 557 00:28:45,239 --> 00:28:48,650 Speaker 1: no idea exactly what they're talking about, but it's nonetheless 558 00:28:48,689 --> 00:28:50,910 Speaker 1: sort of energizing to see that there's all sorts of 559 00:28:50,920 --> 00:28:53,920 Speaker 1: thinking going on for the next and the next next 560 00:28:53,930 --> 00:28:55,359 Speaker 1: uh generation of frog 561 00:28:56,170 --> 00:28:59,810 Speaker 1: um Chris, we talked about military uh issues earlier. I 562 00:28:59,819 --> 00:29:02,530 Speaker 1: want to bring that back. Uh Even in your book, 563 00:29:02,540 --> 00:29:06,109 Speaker 1: uh from the very beginning when rockets were being developed, 564 00:29:06,119 --> 00:29:09,949 Speaker 1: uh military uh experts saw the application of chips in 565 00:29:09,959 --> 00:29:12,920 Speaker 1: the guidance system and you have this uh really cool 566 00:29:12,930 --> 00:29:14,420 Speaker 1: chapter in your book where you talk about the early 567 00:29:14,430 --> 00:29:17,050 Speaker 1: nineties during the Gulf War. First time we saw the 568 00:29:17,060 --> 00:29:19,189 Speaker 1: application of, you know, advanced guidance systems. 569 00:29:19,459 --> 00:29:23,599 Speaker 1: Um So military applications are always, always at the forefront 570 00:29:23,609 --> 00:29:25,510 Speaker 1: of ship demand. Is that a good thing? 571 00:29:27,380 --> 00:29:28,949 Speaker 2: Well, I, I don't know if it's a good thing. 572 00:29:28,959 --> 00:29:31,650 Speaker 2: I think it's an inevitable thing. I think uh it 573 00:29:31,660 --> 00:29:35,890 Speaker 2: is inevitable today that militaries are going to try to 574 00:29:36,239 --> 00:29:40,050 Speaker 2: acquire whatever edge they can. Um And today that means 575 00:29:40,060 --> 00:29:42,069 Speaker 2: developing systems that are more autonomous 576 00:29:42,589 --> 00:29:45,209 Speaker 2: uh autonomous systems are a key focus of all of 577 00:29:45,219 --> 00:29:49,540 Speaker 2: the world's major militaries and deploying A I to defense 578 00:29:49,550 --> 00:29:53,300 Speaker 2: and intelligence systems we already see in the Russia Ukraine War. 579 00:29:53,310 --> 00:29:56,130 Speaker 2: Plenty of uh of examples of ways that A I 580 00:29:56,140 --> 00:29:59,280 Speaker 2: is is being used. For example, militaries have computer vision 581 00:29:59,290 --> 00:30:01,520 Speaker 2: algorithms in their intelligence systems. So 582 00:30:01,910 --> 00:30:04,280 Speaker 2: take a satellite photo and differentiate what's a tank and 583 00:30:04,290 --> 00:30:06,609 Speaker 2: what's a truck, not like be done automatically. That's, that's 584 00:30:06,619 --> 00:30:11,439 Speaker 2: pretty simple. Um But uh more decisions about um communications. 585 00:30:11,449 --> 00:30:15,280 Speaker 2: For example, if your adversary is jamming your frequencies of 586 00:30:15,290 --> 00:30:17,939 Speaker 2: your communication system, can you use an A I to 587 00:30:17,949 --> 00:30:21,439 Speaker 2: rapidly jump between frequency in a way that's faster than 588 00:30:21,449 --> 00:30:22,579 Speaker 2: your adversary system? 589 00:30:23,140 --> 00:30:25,709 Speaker 2: Uh These are some of the use cases of, of a, of, 590 00:30:25,719 --> 00:30:28,130 Speaker 2: of A I and military contacts. And then I think 591 00:30:28,140 --> 00:30:30,770 Speaker 2: if you look at the role that drones have played 592 00:30:30,780 --> 00:30:33,829 Speaker 2: in the Russia Ukraine war again, that provides a very clear, 593 00:30:33,839 --> 00:30:36,930 Speaker 2: I think worrisome but clear pathway for how militaries will 594 00:30:36,939 --> 00:30:42,430 Speaker 2: begin to deploy increasing degrees of autonomy in drones as well. 595 00:30:42,439 --> 00:30:45,290 Speaker 2: And so we're gonna, I think have a very complex 596 00:30:45,300 --> 00:30:46,449 Speaker 2: ethical debates over the next 597 00:30:46,525 --> 00:30:50,854 Speaker 2: several decades about in what ways are we comfortable deploying 598 00:30:50,864 --> 00:30:53,734 Speaker 2: autonomy in military systems? But just as we're running at 599 00:30:53,744 --> 00:30:56,954 Speaker 2: autonomous cars and just as autonomy is coming to all other, 600 00:30:56,964 --> 00:30:59,515 Speaker 2: the range P applications. So too, it's already coming to 601 00:30:59,525 --> 00:31:02,635 Speaker 2: military systems. Uh And there is a, I think very 602 00:31:02,645 --> 00:31:05,925 Speaker 2: real race between the world's leading militaries and above all 603 00:31:05,935 --> 00:31:08,915 Speaker 2: between China and the US to harness this technology for 604 00:31:08,925 --> 00:31:09,834 Speaker 2: defense purposes. 605 00:31:10,780 --> 00:31:13,599 Speaker 1: So robot soldiers here we come. Um Yeah, when I 606 00:31:13,609 --> 00:31:16,689 Speaker 1: see those Boston dynamics videos of those, you know, somersaulting 607 00:31:16,699 --> 00:31:19,579 Speaker 1: uh robots. The first thing I don't think initially everybody 608 00:31:19,589 --> 00:31:21,949 Speaker 1: said that that's like the next, you know, guy who's 609 00:31:21,959 --> 00:31:24,689 Speaker 1: shelving stuff at the Amazon warehouse. But to me, it 610 00:31:24,699 --> 00:31:27,790 Speaker 1: was more like somebody holding an AK-47 and, you know, 611 00:31:27,800 --> 00:31:30,349 Speaker 1: joining the front. Um Chris, you know, I mean, this 612 00:31:30,359 --> 00:31:33,810 Speaker 1: is just a stream of consciousness observation while the Ukraine 613 00:31:33,819 --> 00:31:36,589 Speaker 1: conflict shows all sorts of cutting edge application, 614 00:31:37,050 --> 00:31:39,689 Speaker 1: it still looks like a very first world war type 615 00:31:39,699 --> 00:31:42,969 Speaker 1: of uh terrain. When we look at the videos trenches 616 00:31:42,979 --> 00:31:45,660 Speaker 1: and something called dragon's teeth, which is basically, you know, 617 00:31:45,670 --> 00:31:48,670 Speaker 1: concrete boulders to stop tanks from going over. Um So 618 00:31:48,680 --> 00:31:50,719 Speaker 1: it seems like, you know, we have a lot of 619 00:31:50,729 --> 00:31:54,750 Speaker 1: shiny new toys. Uh but the adaption of them, the 620 00:31:54,760 --> 00:31:57,119 Speaker 1: full usage of them, it takes a while, right? I mean, 621 00:31:57,130 --> 00:31:59,250 Speaker 1: would you agree if I say that the sort of 622 00:31:59,260 --> 00:32:02,219 Speaker 1: compute power that we have right now, even the existing 623 00:32:02,229 --> 00:32:04,750 Speaker 1: computer power, forget about the stuff that's coming. We're barely 624 00:32:04,760 --> 00:32:05,880 Speaker 1: using a fraction of that. 625 00:32:07,619 --> 00:32:09,750 Speaker 2: I, I think that's right. II I think when you 626 00:32:09,760 --> 00:32:13,510 Speaker 2: look at the Russian Ukraine war, it's easy to miss 627 00:32:13,520 --> 00:32:16,280 Speaker 2: a lot of the ways in which computing has actually 628 00:32:16,290 --> 00:32:20,310 Speaker 2: been pretty decisive. Um So for example, there's, there's the 629 00:32:20,319 --> 00:32:22,329 Speaker 2: whole cyber war that didn't happen. 630 00:32:22,939 --> 00:32:28,849 Speaker 2: Um We collectively expected that Russia would launch devastating cyberattacks 631 00:32:28,859 --> 00:32:31,579 Speaker 2: and knock out Ukraine's cyber infrastructure ali accounts, but that 632 00:32:31,589 --> 00:32:32,349 Speaker 2: didn't happen 633 00:32:32,699 --> 00:32:36,290 Speaker 2: in no small part because Ukraine was defended by Google 634 00:32:36,300 --> 00:32:39,849 Speaker 2: and Microsoft and some of the key Western tech giants and, 635 00:32:39,859 --> 00:32:42,849 Speaker 2: and their computing capabilities are key to cyber defenses. And it, 636 00:32:42,969 --> 00:32:45,630 Speaker 2: you know, we've talked about autonomy and uh in, in 637 00:32:45,640 --> 00:32:48,209 Speaker 2: many parts of the economy, well, autonomy in both cyber 638 00:32:48,219 --> 00:32:51,689 Speaker 2: offense and cyber defense is coming soon, it's actually already 639 00:32:51,699 --> 00:32:54,180 Speaker 2: already arrived. And so uh that will be a AAA 640 00:32:54,189 --> 00:32:57,839 Speaker 2: new frontier of the application of, of A I. When 641 00:32:57,849 --> 00:33:00,729 Speaker 2: it comes to the intelligence that Ukraine has received from 642 00:33:00,739 --> 00:33:01,310 Speaker 2: the West, 643 00:33:01,869 --> 00:33:05,849 Speaker 2: we've already uh we've got many examples of um Ukraine 644 00:33:05,859 --> 00:33:09,260 Speaker 2: receiving very accurate targeting data from the West and Signals 645 00:33:09,270 --> 00:33:12,530 Speaker 2: Intelligence is one of the early use cases uh of 646 00:33:12,540 --> 00:33:14,920 Speaker 2: the application of A I to military systems. And so 647 00:33:14,930 --> 00:33:18,000 Speaker 2: we don't have uh great visibility into all of the 648 00:33:18,010 --> 00:33:22,089 Speaker 2: specific platforms that the US is using to uh give 649 00:33:22,099 --> 00:33:25,569 Speaker 2: Ukraine uh targeting data, for example, for its high on rockets. 650 00:33:25,579 --> 00:33:26,738 Speaker 2: But I don't think you can 651 00:33:27,130 --> 00:33:30,839 Speaker 2: um safely. Uh I think you have to assume that 652 00:33:30,969 --> 00:33:33,810 Speaker 2: uh there's a whole lot of computing uh involved in 653 00:33:33,819 --> 00:33:38,050 Speaker 2: the very rapid identification of targets uh transmission of those targets, 654 00:33:38,060 --> 00:33:41,199 Speaker 2: Ukrainian forces and the Ukrainians uh our system, if you 655 00:33:41,209 --> 00:33:44,130 Speaker 2: just press a button and the rocket flies automatically it 656 00:33:44,140 --> 00:33:45,760 Speaker 2: hit its targets. So those are just two ways in 657 00:33:45,770 --> 00:33:46,630 Speaker 2: which computing is made. 658 00:33:47,030 --> 00:33:49,420 Speaker 2: I think a real difference on the battlefield. But you know, 659 00:33:49,430 --> 00:33:51,689 Speaker 2: I think you're right that, uh, in some ways this 660 00:33:51,699 --> 00:33:53,750 Speaker 2: is a 21st century war, in other ways, it's 1/19 661 00:33:53,760 --> 00:33:58,650 Speaker 2: century war. Uh, and I, I think that speaks to the, uh, the, 662 00:33:59,050 --> 00:34:01,709 Speaker 2: the fact that regardless of the technologies or in some 663 00:34:01,719 --> 00:34:04,819 Speaker 2: ways because of the technologies, uh, warfare remains a very 664 00:34:04,829 --> 00:34:05,699 Speaker 2: brutal business. 665 00:34:07,030 --> 00:34:10,629 Speaker 1: Yeah, absolutely. So, we've sort of almost naturally progressed to 666 00:34:10,639 --> 00:34:13,290 Speaker 1: talking about Ukraine. That was supposed to be my final question. 667 00:34:13,419 --> 00:34:15,719 Speaker 1: And I think many fans of Chip War may not 668 00:34:15,729 --> 00:34:19,590 Speaker 1: be aware that your core expertise include Russia and Cold 669 00:34:19,600 --> 00:34:22,719 Speaker 1: War history. And you've written several books on Putin and 670 00:34:22,729 --> 00:34:25,719 Speaker 1: Imperial Russia. I'd love to do a separate podcast one day, 671 00:34:25,729 --> 00:34:28,229 Speaker 1: Chris with just talking about Russia because there's like so 672 00:34:28,239 --> 00:34:31,050 Speaker 1: much to talk about, particularly the historical context that you 673 00:34:31,060 --> 00:34:34,399 Speaker 1: have written so articulately about. But, uh, just give me 674 00:34:34,409 --> 00:34:36,360 Speaker 1: a sense, uh, we've talked just a little bit about 675 00:34:36,370 --> 00:34:36,719 Speaker 1: the 676 00:34:36,979 --> 00:34:40,379 Speaker 1: technology around the Ukraine conflict. But as a historian, you know, 677 00:34:40,389 --> 00:34:43,010 Speaker 1: what's your take on the course the conflict has taken 678 00:34:43,229 --> 00:34:46,429 Speaker 1: and what would be a path to resolution? 679 00:34:48,790 --> 00:34:51,419 Speaker 2: Well, I think the the key driver of, of the 680 00:34:51,429 --> 00:34:54,459 Speaker 2: conflict is, is a question about where does Russia's sphere 681 00:34:54,469 --> 00:34:57,569 Speaker 2: of control end? I think the, the Russian view which 682 00:34:57,739 --> 00:35:01,050 Speaker 2: President Putin has articulated is that Ukraine is an, ought 683 00:35:01,060 --> 00:35:03,959 Speaker 2: to be in Russia's sphere of control. And the Ukrainian 684 00:35:03,969 --> 00:35:06,399 Speaker 2: view and the Western view is that Ukraine should be 685 00:35:06,409 --> 00:35:08,620 Speaker 2: able to make its own choices about whether it wants 686 00:35:08,629 --> 00:35:10,870 Speaker 2: to be part of the Russian sphere. I mean, that's 687 00:35:10,879 --> 00:35:15,419 Speaker 2: the core of disagreement that has led to uh the conflict. 688 00:35:15,429 --> 00:35:15,860 Speaker 2: And I think 689 00:35:16,239 --> 00:35:20,209 Speaker 2: there's no pathway in sight for that issue to be resolved. 690 00:35:20,219 --> 00:35:24,469 Speaker 2: And people speculate, welcome, you know, this territorial settlement, that 691 00:35:24,479 --> 00:35:25,409 Speaker 2: territorial settlement 692 00:35:25,770 --> 00:35:28,709 Speaker 2: resolve the conflict. But I think the core question is 693 00:35:28,790 --> 00:35:35,020 Speaker 2: um is Russia going to continue uh fighting to keep 694 00:35:35,030 --> 00:35:37,830 Speaker 2: Ukraine or try to keep Ukraine in its sphere or 695 00:35:37,840 --> 00:35:39,550 Speaker 2: is it willing to let Ukraine not be uh in 696 00:35:39,560 --> 00:35:42,989 Speaker 2: its sphere? And that's, I think the key uncertainty that 697 00:35:43,120 --> 00:35:46,158 Speaker 2: uh we just don't know uh Russia's willingness to fight 698 00:35:46,169 --> 00:35:49,189 Speaker 2: on to pursue that aim. And the Ukrainians will say 699 00:35:49,199 --> 00:35:50,840 Speaker 2: we're not going to do a cease fire now because 700 00:35:50,850 --> 00:35:53,459 Speaker 2: if we do, the Russians will start fighting again after 701 00:35:53,469 --> 00:35:55,040 Speaker 2: rebuilding for a couple of years. And 702 00:35:55,600 --> 00:35:57,330 Speaker 2: you know, I understand why they think that because that's 703 00:35:57,340 --> 00:36:00,020 Speaker 2: been the track record of, of recent history, there was 704 00:36:00,030 --> 00:36:02,639 Speaker 2: a cease fire in 2014, the Russians rebuilt and uh 705 00:36:02,649 --> 00:36:06,429 Speaker 2: attacked again. And so I, I think that that that 706 00:36:06,439 --> 00:36:09,770 Speaker 2: key question of, of ultimately, is Ukraine an independent country 707 00:36:09,780 --> 00:36:12,929 Speaker 2: or not? And does Russia recognize it is independent or not? 708 00:36:13,169 --> 00:36:17,090 Speaker 2: That's what's at stake? Uh And I don't see right 709 00:36:17,100 --> 00:36:22,050 Speaker 2: now um any clear pathway to 710 00:36:22,399 --> 00:36:25,689 Speaker 2: getting a resolution of, of that question that both Ukraine 711 00:36:25,699 --> 00:36:26,610 Speaker 2: and Russia could agree on, 712 00:36:28,129 --> 00:36:33,149 Speaker 1: right. Um gosh, uh not the most hopeful uh you know, prognosis. 713 00:36:33,159 --> 00:36:35,330 Speaker 1: And Chris, we, uh I, I was talking to a 714 00:36:35,340 --> 00:36:36,939 Speaker 1: friend of mine, she's Russian and she just came back 715 00:36:36,949 --> 00:36:39,820 Speaker 1: from Moscow and she said that you don't really feel 716 00:36:39,830 --> 00:36:41,860 Speaker 1: that there's a war going on. Uh when you walk 717 00:36:41,870 --> 00:36:44,280 Speaker 1: the streets of Moscow and when you go to electronic 718 00:36:44,290 --> 00:36:46,620 Speaker 1: stores that used to be, say, 719 00:36:46,729 --> 00:36:49,359 Speaker 1: you know, filled with Apple and Microsoft products, now they're 720 00:36:49,370 --> 00:36:51,979 Speaker 1: filled with uh made in China electronics products and they 721 00:36:51,989 --> 00:36:54,639 Speaker 1: look really good and they're shiny and people are buying 722 00:36:54,649 --> 00:36:58,040 Speaker 1: those things. So this sort of, you know, pivot toward 723 00:36:58,050 --> 00:37:01,280 Speaker 1: Chinese tech. I mean, can Russia sort of, you know, 724 00:37:01,290 --> 00:37:05,159 Speaker 1: withstand Western sanctions substantially better because of that. 725 00:37:07,360 --> 00:37:09,929 Speaker 2: Well, I don't really think the, the pivot to Chinese 726 00:37:09,939 --> 00:37:12,429 Speaker 2: tech is the key to Russia's withstanding of sanctions. I 727 00:37:12,439 --> 00:37:15,840 Speaker 2: I think the key is Russia's oil exports. Um That's 728 00:37:15,850 --> 00:37:17,770 Speaker 2: been the key and I think we've learned over the 729 00:37:17,780 --> 00:37:21,370 Speaker 2: course of the last 18 months of war. Um several 730 00:37:21,379 --> 00:37:27,899 Speaker 2: key things. Um One is that the Russian war effort 731 00:37:27,909 --> 00:37:30,129 Speaker 2: has been very stimulative for the economy. 732 00:37:30,689 --> 00:37:32,750 Speaker 2: Um So the Russian government is, is running a pretty 733 00:37:32,760 --> 00:37:36,360 Speaker 2: large budget deficit now. Um but that is produced a 734 00:37:36,370 --> 00:37:39,560 Speaker 2: sort of Keynesian impact on the Russian economy, 735 00:37:39,739 --> 00:37:42,280 Speaker 2: uh which I think ought to have been expected. But 736 00:37:42,290 --> 00:37:45,509 Speaker 2: most people myself included didn't fully uh expect in the 737 00:37:45,520 --> 00:37:48,729 Speaker 2: early days of the war. And so rather than unemployment 738 00:37:48,739 --> 00:37:52,209 Speaker 2: or declining incomes. We actually have very rapid income growth 739 00:37:52,219 --> 00:37:55,629 Speaker 2: driven by very low unemployment in Russia because of labor 740 00:37:55,639 --> 00:37:58,229 Speaker 2: shortages now. So that's, that's one surprise to the war. 741 00:37:58,239 --> 00:38:02,209 Speaker 2: I think the second surprise that is about the West's 742 00:38:02,219 --> 00:38:05,469 Speaker 2: willingness to impose sanctions because one of the really striking 743 00:38:05,479 --> 00:38:06,689 Speaker 2: acts of the war is that 744 00:38:06,939 --> 00:38:10,510 Speaker 2: uh the West at first started to impose very tough 745 00:38:10,520 --> 00:38:12,790 Speaker 2: oil sanctions on Russia and then immediately walked them all 746 00:38:12,800 --> 00:38:15,839 Speaker 2: back through the price cap scheme. Um And so now 747 00:38:15,850 --> 00:38:19,029 Speaker 2: Russian oil exports are basically unchanged since before the war. 748 00:38:19,040 --> 00:38:22,929 Speaker 2: Um precisely because the West has collectively decided that it 749 00:38:22,939 --> 00:38:25,810 Speaker 2: doesn't actually want to sanction Russian oil exports. It wants 750 00:38:25,820 --> 00:38:28,989 Speaker 2: more Russian oil exports to keep prices as low. 751 00:38:29,260 --> 00:38:31,100 Speaker 2: That is possible. And so if you look at the 752 00:38:31,110 --> 00:38:34,510 Speaker 2: Russian economic situation, that's the key dynamic is that a, 753 00:38:34,520 --> 00:38:39,060 Speaker 2: the domestic economy is being supported by deficit spending domestically 754 00:38:39,070 --> 00:38:42,600 Speaker 2: and b the external accounts are um they look relatively 755 00:38:42,610 --> 00:38:46,070 Speaker 2: stable because oil exports and prices fall on oil. But 756 00:38:46,080 --> 00:38:48,649 Speaker 2: uh volumes are uh almost as high as ever, 757 00:38:49,949 --> 00:38:52,439 Speaker 1: right? But uh I wanna stick with that tech issue. 758 00:38:52,449 --> 00:38:54,520 Speaker 1: I agree with you. I agree that the oil export 759 00:38:54,530 --> 00:38:58,659 Speaker 1: revenues are basically what's driving, you know, Russia's resilience. And 760 00:38:58,669 --> 00:38:59,739 Speaker 1: of course, you know, if they want to buy stuff 761 00:38:59,750 --> 00:39:02,000 Speaker 1: from China, they still have to pay and uh and 762 00:39:02,010 --> 00:39:04,239 Speaker 1: then that's where the revenue is coming from. But going 763 00:39:04,250 --> 00:39:06,259 Speaker 1: back to the tech issue, um 764 00:39:06,899 --> 00:39:09,000 Speaker 1: the, the view is that, you know, if you don't 765 00:39:09,010 --> 00:39:12,100 Speaker 1: have access to Western technology, if the Apple I OS 766 00:39:12,110 --> 00:39:16,080 Speaker 1: is not, you know, upgrading and you're not getting the 767 00:39:16,090 --> 00:39:20,790 Speaker 1: Microsoft 3 65 support that day to day commerce and 768 00:39:20,800 --> 00:39:23,570 Speaker 1: inventory management. Everything sort of comes to a standstill, 769 00:39:23,840 --> 00:39:25,810 Speaker 1: but clearly, that's not the case for Russia. So I'm 770 00:39:25,820 --> 00:39:30,589 Speaker 1: assuming that they're not getting those regular updates from Western tech. 771 00:39:30,659 --> 00:39:33,449 Speaker 1: And still, I don't hear any logistics failure or any 772 00:39:33,489 --> 00:39:36,409 Speaker 1: difficulty in running power plants or for that matter, running 773 00:39:36,419 --> 00:39:40,459 Speaker 1: the military in the case of Russia. So it's basically 774 00:39:40,469 --> 00:39:42,639 Speaker 1: what smuggling is that is that the way they're sort 775 00:39:42,649 --> 00:39:43,669 Speaker 1: of keeping themselves going. 776 00:39:44,820 --> 00:39:45,929 Speaker 1: Well, I think for, for 777 00:39:45,939 --> 00:39:50,259 Speaker 2: software tools, uh there, it's always been relatively easy to 778 00:39:50,270 --> 00:39:52,830 Speaker 2: pirate software tools and, and rush actually has been a 779 00:39:52,840 --> 00:39:56,350 Speaker 2: big market for Pirated software. And so if you can't 780 00:39:56,360 --> 00:39:58,709 Speaker 2: get the latest IO I OS update, you just use 781 00:39:58,719 --> 00:40:02,610 Speaker 2: the old version, it's slightly more vulnerable to being hacked 782 00:40:02,620 --> 00:40:05,049 Speaker 2: your performance declines over time. 783 00:40:05,310 --> 00:40:09,699 Speaker 2: Um But that's, that's pretty manageable. Um I think for, 784 00:40:09,709 --> 00:40:13,320 Speaker 2: for enterprise software, one of the surprising things has been 785 00:40:13,360 --> 00:40:18,479 Speaker 2: how long many enterprise software firms kept servicing Russian customers. 786 00:40:18,570 --> 00:40:19,320 Speaker 2: Um 787 00:40:19,870 --> 00:40:22,699 Speaker 2: And uh there's, it's been a, a matter of some debate, 788 00:40:22,709 --> 00:40:24,080 Speaker 2: I think in the West as well. That was the 789 00:40:24,090 --> 00:40:28,770 Speaker 2: right approach, but many enterprise software companies, um, essentially kept 790 00:40:28,780 --> 00:40:32,620 Speaker 2: Russian clients um, up to date with functional software for 791 00:40:32,629 --> 00:40:34,729 Speaker 2: a very long time, some even up to the present 792 00:40:34,899 --> 00:40:37,129 Speaker 2: uh day. So I, I think on the software side, 793 00:40:37,139 --> 00:40:41,199 Speaker 2: Russia has actually had very few difficulties um in accessing 794 00:40:41,209 --> 00:40:44,639 Speaker 2: maybe for certain niche types of industrial software. But generally speaking, 795 00:40:44,649 --> 00:40:47,889 Speaker 2: it hasn't, doesn't appear to have been a problem um for, 796 00:40:47,899 --> 00:40:49,459 Speaker 2: for tech components, 797 00:40:50,169 --> 00:40:53,799 Speaker 2: you know, there have been open discussions in the Russian 798 00:40:53,810 --> 00:40:57,439 Speaker 2: press about certain types of component shortages. Um but the 799 00:40:57,449 --> 00:41:02,469 Speaker 2: Russian manufacturing base is actually relatively small um outside of 800 00:41:02,479 --> 00:41:06,669 Speaker 2: the auto sector which has gotten uh clobbered by um 801 00:41:06,679 --> 00:41:09,629 Speaker 2: by sanctions of export controls. The other big sector for 802 00:41:09,639 --> 00:41:11,000 Speaker 2: manufacturing is the military. 803 00:41:11,399 --> 00:41:14,259 Speaker 2: Um And the military has got the budget, it needs 804 00:41:14,270 --> 00:41:18,600 Speaker 2: to uh import components um from abroad. It probably does 805 00:41:18,610 --> 00:41:20,638 Speaker 2: sort of an inefficient manner. It probably faced supply chain 806 00:41:20,649 --> 00:41:23,229 Speaker 2: difficulties due to sanctions, but it's been able to import 807 00:41:23,239 --> 00:41:26,800 Speaker 2: at least some volume because we know that multi production continues. 808 00:41:26,810 --> 00:41:29,409 Speaker 2: Um And so I think those are kind of the 809 00:41:29,419 --> 00:41:31,859 Speaker 2: autos in the in the the defense industry, the two 810 00:41:32,000 --> 00:41:35,070 Speaker 2: sectors of Russian manufacturing that are most reliant on uh 811 00:41:35,080 --> 00:41:36,259 Speaker 2: foreign components. 812 00:41:37,649 --> 00:41:40,639 Speaker 1: But there's no area of sort of symbiosis that there 813 00:41:40,649 --> 00:41:44,110 Speaker 1: are certain areas of tech excellence in Russia that China 814 00:41:44,120 --> 00:41:45,658 Speaker 1: can exploit or vice versa. 815 00:41:47,659 --> 00:41:51,429 Speaker 2: I there are a, a small number of discrete examples. 816 00:41:51,439 --> 00:41:55,739 Speaker 2: One could point to um specific computer vision firms that 817 00:41:55,750 --> 00:41:59,850 Speaker 2: have been invested in by Chinese firms, but they're extraordinarily 818 00:41:59,860 --> 00:42:03,179 Speaker 2: small in number. I think that the tech dependence is 819 00:42:03,189 --> 00:42:07,029 Speaker 2: almost exclusively unidirectional Russia is becoming dependent on China. 820 00:42:07,550 --> 00:42:11,479 Speaker 1: OK. Totally different question. So I'm talking to you from Singapore. Actually, 821 00:42:11,489 --> 00:42:13,120 Speaker 1: I don't know where in the world you are right now, Chris, 822 00:42:13,129 --> 00:42:13,709 Speaker 1: where are you? 823 00:42:14,540 --> 00:42:16,590 Speaker 2: I'm, I'm today in Wyoming. 824 00:42:16,770 --> 00:42:19,419 Speaker 1: Oh My. OK. Uh I thought you were like in 825 00:42:19,429 --> 00:42:22,840 Speaker 1: China or something. Uh OK. So um I'm talking to 826 00:42:22,850 --> 00:42:26,290 Speaker 1: you from Singapore, Singapore over the last 50 years have 827 00:42:26,300 --> 00:42:30,439 Speaker 1: had a manufacturing footprint with respect to chips and semiconductors 828 00:42:30,449 --> 00:42:34,669 Speaker 1: and so on. Uh any sense of the excellence of 829 00:42:34,679 --> 00:42:37,969 Speaker 1: production in Singapore vis a vis a Korea, Japan and 830 00:42:37,979 --> 00:42:38,409 Speaker 1: so on. 831 00:42:40,820 --> 00:42:43,949 Speaker 2: Well, I think in terms of cutting edge production, the 832 00:42:43,959 --> 00:42:46,370 Speaker 2: the cutting edge has been in recent years in Taiwan 833 00:42:46,379 --> 00:42:49,929 Speaker 2: and Korea uh in the US, not in Singapore, but 834 00:42:49,939 --> 00:42:54,149 Speaker 2: there's a very uh substantial manufacturing base in Singapore. And 835 00:42:54,159 --> 00:42:55,860 Speaker 2: just because something is not cutting edge doesn't mean it's 836 00:42:55,870 --> 00:42:59,629 Speaker 2: not excellent. Um indeed, there are lots of very profitable 837 00:42:59,639 --> 00:43:03,090 Speaker 2: uh companies that do non cutting edge chip making often 838 00:43:03,100 --> 00:43:06,169 Speaker 2: with unique mix signal or analog or radio frequency 839 00:43:06,245 --> 00:43:09,915 Speaker 2: requirements as well. Uh And so III I would say 840 00:43:09,925 --> 00:43:12,495 Speaker 2: that Singapore's chip making has not been cutting edge, but 841 00:43:12,504 --> 00:43:14,674 Speaker 2: I wouldn't say that that means it has been excellent. 842 00:43:14,774 --> 00:43:16,995 Speaker 2: Um because there are many different ways you can actually 843 00:43:17,004 --> 00:43:21,074 Speaker 2: define what's a profitable business. What's a um what's a, 844 00:43:21,084 --> 00:43:24,404 Speaker 2: a technologically viable uh uh product? And I think Singapore 845 00:43:24,415 --> 00:43:26,915 Speaker 2: is likely to remain an important hub of chip making. 846 00:43:27,175 --> 00:43:30,014 Speaker 2: Uh and also uh a hub of assembling some of 847 00:43:30,024 --> 00:43:31,594 Speaker 2: the tools that are used to make chips, 848 00:43:32,179 --> 00:43:35,929 Speaker 1: right? OK. So final question since you mentioned Korea, uh 849 00:43:35,939 --> 00:43:38,669 Speaker 1: when people sort of, you know, dissect it through the 850 00:43:38,679 --> 00:43:40,589 Speaker 1: uh Huawei phone. So one of the things that both 851 00:43:40,600 --> 00:43:43,179 Speaker 1: mentioned that there are some Korean parts in it, I 852 00:43:43,189 --> 00:43:46,109 Speaker 1: think he, right, that's the company that produces those things. 853 00:43:46,120 --> 00:43:50,449 Speaker 1: Um So yes, Taiwan is at the absolute, you know, 854 00:43:50,459 --> 00:43:53,750 Speaker 1: singular you know, security with respect to cutting it chips 855 00:43:53,760 --> 00:43:56,550 Speaker 1: just gives a sense of how things are in Korea. 856 00:43:56,560 --> 00:43:57,489 Speaker 1: Perhaps also in Japan. 857 00:43:58,879 --> 00:44:02,830 Speaker 2: Korea is really critical as a producer of memory chips. Now, 858 00:44:02,840 --> 00:44:04,510 Speaker 2: there's two types of memory chips, both of which are 859 00:44:04,520 --> 00:44:07,189 Speaker 2: produced at the cutting edge in Korea. There's a number 860 00:44:07,199 --> 00:44:10,379 Speaker 2: of firms that produce cutting edge memory chips. Um uh 861 00:44:10,409 --> 00:44:13,750 Speaker 2: one in the US, uh one in Japan and, and 862 00:44:13,760 --> 00:44:18,330 Speaker 2: two in Korea, primarily um memory Chis are basically a commodity, 863 00:44:18,340 --> 00:44:21,010 Speaker 2: you can more or less swap out one company's memory 864 00:44:21,020 --> 00:44:22,070 Speaker 2: Chis for another. 865 00:44:22,360 --> 00:44:25,050 Speaker 2: And as a result, there's not nearly as much differentiation 866 00:44:25,060 --> 00:44:28,439 Speaker 2: between uh the firms and the pricing dynamics are also 867 00:44:28,449 --> 00:44:32,689 Speaker 2: quite different profitability is uh generally lower. Um But Korea 868 00:44:32,699 --> 00:44:35,570 Speaker 2: is by far the world's largest producer of uh memory 869 00:44:35,580 --> 00:44:38,830 Speaker 2: chips and Korean firms, Samsung and Hex are are absolutely 870 00:44:38,915 --> 00:44:42,533 Speaker 2: critical suppliers of of of memory chips when it comes 871 00:44:42,544 --> 00:44:45,435 Speaker 2: to Japan, Japan is actually not primarily important as a 872 00:44:45,445 --> 00:44:48,014 Speaker 2: chip maker. There, there are chips made in Japan, especially 873 00:44:48,024 --> 00:44:51,404 Speaker 2: for auto and industrial uses. Um But Japan is is 874 00:44:51,415 --> 00:44:54,665 Speaker 2: really unique as a producer of highly specialized chemicals used 875 00:44:54,675 --> 00:44:57,495 Speaker 2: in ship making and also some of the machine tools 876 00:44:57,504 --> 00:45:01,125 Speaker 2: needed to manufacture advanced ships. And that's really where Japanese 877 00:45:01,135 --> 00:45:04,104 Speaker 2: firms have uh pretty unique uh market positions. 878 00:45:04,820 --> 00:45:10,449 Speaker 1: I remember during the Fukushima earthquake and tsunami, uh this 879 00:45:10,459 --> 00:45:13,169 Speaker 1: is a power plant issued that there were that one 880 00:45:13,179 --> 00:45:15,350 Speaker 1: note of vulnerability that there were some chips that were 881 00:45:15,360 --> 00:45:17,949 Speaker 1: made for navigation systems which were not coming in. And 882 00:45:17,959 --> 00:45:21,000 Speaker 1: as a result, Thailand's auto production came to a standstill. 883 00:45:21,080 --> 00:45:22,429 Speaker 1: Uh And that was the first time I sort of 884 00:45:22,439 --> 00:45:25,049 Speaker 1: started thinking about these, you know, nodes of vulnerability. 885 00:45:25,260 --> 00:45:28,350 Speaker 1: Um Chris uh I believe in your book, you had 886 00:45:28,360 --> 00:45:30,139 Speaker 1: mentioned as far as the memory chips, you know, the, 887 00:45:30,149 --> 00:45:32,409 Speaker 1: the DRS and the NANS are concerned that is one 888 00:45:32,419 --> 00:45:35,149 Speaker 1: area where the Chinese can probably eat Korea's lunch. 889 00:45:36,860 --> 00:45:37,780 Speaker 1: Well, it's, it's, 890 00:45:37,790 --> 00:45:41,229 Speaker 2: it's probably true if China can get access to the 891 00:45:41,239 --> 00:45:44,610 Speaker 2: tools needed to make those ships. And until recently, China 892 00:45:44,620 --> 00:45:49,229 Speaker 2: could uh now, however, uh those tools are restricted, uh 893 00:45:49,239 --> 00:45:52,330 Speaker 2: Chinese firms can't purchase them. And actually the leading Chinese 894 00:45:52,340 --> 00:45:55,429 Speaker 2: memory producer, a company called YMTC is uh is uh 895 00:45:55,439 --> 00:45:59,250 Speaker 2: under additional us sanctions in addition to the the broad ones. 896 00:45:59,639 --> 00:46:00,989 Speaker 2: And so actually, 897 00:46:01,239 --> 00:46:04,270 Speaker 2: um I think one could argue that Korean firms have 898 00:46:04,280 --> 00:46:08,089 Speaker 2: seen a potential major competitor really uh impacted by these 899 00:46:08,100 --> 00:46:11,850 Speaker 2: controls in a way that might protect their market share 900 00:46:11,860 --> 00:46:13,388 Speaker 2: uh for some time to come. So there are lots 901 00:46:13,399 --> 00:46:17,489 Speaker 2: of unexpected um knock on effects of the controls that 902 00:46:17,500 --> 00:46:21,250 Speaker 2: have interesting implications uh for for a number of different countries. 903 00:46:21,260 --> 00:46:22,469 Speaker 2: But Korea being one of them, 904 00:46:23,379 --> 00:46:26,169 Speaker 1: lots of unexpected knock on effects. I think that's a 905 00:46:26,179 --> 00:46:29,169 Speaker 1: very good way to sort of summarize the the world, 906 00:46:29,179 --> 00:46:32,979 Speaker 1: the chip war dynamic. Uh Chris, I, I really appreciate 907 00:46:32,989 --> 00:46:35,409 Speaker 1: your time and thank you so much for your insights. 908 00:46:36,330 --> 00:46:37,209 Speaker 2: Well, thank you for having 909 00:46:37,219 --> 00:46:40,799 Speaker 1: me. Fantastic time with the uh talking about this and 910 00:46:40,810 --> 00:46:43,509 Speaker 1: great job with the book. Uh Kobe Time is for 911 00:46:43,520 --> 00:46:47,629 Speaker 1: information only and does not represent any trade recommendations. All 912 00:46:47,639 --> 00:46:50,939 Speaker 1: 108 episodes of the podcast are available on youtube and 913 00:46:50,949 --> 00:46:54,959 Speaker 1: all major podcast platforms including Apple Google and Spotify. As 914 00:46:54,969 --> 00:46:57,629 Speaker 1: for our research publications, webinars and live streams, you can 915 00:46:57,639 --> 00:47:01,219 Speaker 1: find them all by Googling devious research library. Have a 916 00:47:01,229 --> 00:47:02,060 Speaker 1: great day.