1 00:00:00,577 --> 00:00:03,767 Quantum computing doesn't suit every 2 00:00:03,863 --> 00:00:07,678 problem, right? You should have a supermassive to 3 00:00:09,890 --> 00:00:13,657 work with that. It's not like that. If you can classically 4 00:00:13,657 --> 00:00:17,296 solve a problem, you don't go for the quantum 5 00:00:17,424 --> 00:00:20,935 computer, quantum computing and quantum 6 00:00:20,935 --> 00:00:23,820 computers. Welcome to Impact 7 00:00:23,820 --> 00:00:30,425 Quantum. 8 00:00:31,900 --> 00:00:35,462 Quantum podcast, they're breaking the mold. Science has got 9 00:00:35,462 --> 00:00:38,945 beats to fold. Hello and welcome back to Impact Quantum, where we 10 00:00:38,977 --> 00:00:42,443 discuss the emerging field and entire industry that is quantum 11 00:00:42,443 --> 00:00:46,054 computing. We don't need to have a PhD necessarily, although 12 00:00:46,456 --> 00:00:49,136 probably helps. You just need to be curious. 13 00:00:49,890 --> 00:00:53,694 And with me is the most quantum curious person I know, 14 00:00:53,822 --> 00:00:57,609 Candace Cahootley. How's it going, Candace? It's great, Frank. How are 15 00:00:57,609 --> 00:01:01,060 you today? I'm doing fantastic. I'm doing fantastic. It's, 16 00:01:01,124 --> 00:01:04,719 um, we finally broke our ice, uh, below freezing, 17 00:01:05,152 --> 00:01:08,250 uh, record here, and, uh, snow's actually melting. 18 00:01:09,149 --> 00:01:12,952 Oh my goodness. Oh, that's just the beginning. How fantastic. It just snowed 19 00:01:12,952 --> 00:01:16,692 another foot last night because, because I'm in Montreal, 20 00:01:16,708 --> 00:01:20,463 so the temperature just hit freezing, and so I was like, oh, it warmed up, 21 00:01:20,479 --> 00:01:23,914 and then it snowed a foot. So that's how— that's what happens here when it 22 00:01:23,914 --> 00:01:27,512 warms up. But that's okay. That's okay. Today we have 23 00:01:27,689 --> 00:01:30,854 a great guest. We're going to be speaking with Dr. 24 00:01:30,918 --> 00:01:34,662 Helaina Bahrami, and she is a lady 25 00:01:34,694 --> 00:01:38,276 of many different titles. She has her hands in a lot of different pies. 26 00:01:38,775 --> 00:01:42,373 She's an AI and machine learning lead expert. She's an 27 00:01:42,406 --> 00:01:45,699 innovator. She is an entrepreneur. She 28 00:01:46,743 --> 00:01:50,005 works at the Oakland University of 29 00:01:50,037 --> 00:01:53,822 Technology. Where she's doing postdoctoral research. 30 00:01:54,592 --> 00:01:58,166 Helena, there's so much that you do, you know, I couldn't, I couldn't even sum 31 00:01:58,214 --> 00:02:01,981 it up. How are you today? Hello, and 32 00:02:02,943 --> 00:02:06,405 thank you very much for having me today. I'm doing very great, 33 00:02:06,533 --> 00:02:10,284 especially meeting with you, very lovely 34 00:02:10,316 --> 00:02:13,569 people, and I'm very excited to have our discussion. 35 00:02:14,547 --> 00:02:18,122 Awesome. And you are in Auckland because I thought I 36 00:02:18,410 --> 00:02:21,985 heard that as Oakland. And so this is Lord of the Rings, not 37 00:02:22,177 --> 00:02:25,945 MC Hammer. Sorry, I don't want to— I don't, I don't like calling out because 38 00:02:26,009 --> 00:02:29,632 I also have a New York accent. So yes, it's Auckland, New Zealand. 39 00:02:29,792 --> 00:02:33,239 And as we said in the virtual green room, that it's already tomorrow where she 40 00:02:33,239 --> 00:02:35,259 is, and we asked her how the future is. So 41 00:02:36,589 --> 00:02:40,052 yeah, yeah, uh, please forgive my 42 00:02:40,052 --> 00:02:43,386 accent, uh, so I'm not a native, but yes, you're right, I'm in 43 00:02:43,482 --> 00:02:47,138 Auckland. Okay. No problem. We have accents too, just, uh, 44 00:02:47,988 --> 00:02:49,864 we don't notice it. Um, so 45 00:02:51,821 --> 00:02:55,669 how did you— this— you do a lot of interesting things. So, so what 46 00:02:55,669 --> 00:02:59,454 exactly are you working on now that excites you at 47 00:02:59,454 --> 00:03:03,174 the moment? I actually, I can divide it into 48 00:03:03,382 --> 00:03:06,958 3 sectors, mainly focusing 49 00:03:06,990 --> 00:03:10,486 because my main power is artificial intelligence and machine 50 00:03:10,582 --> 00:03:13,453 learning. However, I'm very 51 00:03:14,207 --> 00:03:17,302 interested and very 52 00:03:17,736 --> 00:03:20,786 passionate about quantum mechanics, quantum physics, 53 00:03:21,573 --> 00:03:25,056 and because of some personal and also 54 00:03:25,056 --> 00:03:26,485 some kind of, you know, 55 00:03:29,760 --> 00:03:33,244 I can say it like a mission, I'm trying to 56 00:03:33,789 --> 00:03:37,450 innovate in the healthcare field to help people. So 57 00:03:38,846 --> 00:03:41,511 to merge these three fields, quantum 58 00:03:42,137 --> 00:03:45,985 computers, quantum computing, and also AI 59 00:03:45,985 --> 00:03:49,350 and machine learning, and neuroscience to help 60 00:03:49,350 --> 00:03:53,036 neurodegenerative— to help to provide some sort of solution for 61 00:03:53,036 --> 00:03:55,456 neurodegenerative disease. I tried to 62 00:03:56,962 --> 00:04:00,503 come up with some ideas to marry these 63 00:04:00,888 --> 00:04:04,654 three disciplines. You may 64 00:04:04,654 --> 00:04:08,499 heard about quantum machine learning, that it is right now a 65 00:04:08,499 --> 00:04:12,110 new emerging field. So mainly it's like 66 00:04:12,110 --> 00:04:15,594 that, using the power of quantum physics 67 00:04:16,028 --> 00:04:19,834 and quantum computing to empower 68 00:04:19,978 --> 00:04:22,869 the memory capacity and also 69 00:04:23,415 --> 00:04:27,237 algorithmic power to address 70 00:04:27,446 --> 00:04:30,979 or to solve an intractable problem. 71 00:04:31,220 --> 00:04:34,897 When I talk about intractable problem, it 72 00:04:34,897 --> 00:04:38,575 means that for classical computers, it requires 73 00:04:38,992 --> 00:04:42,376 many years to solve a specific problem. 74 00:04:43,819 --> 00:04:47,377 Drug discovery is amongst one of those problems that 75 00:04:47,490 --> 00:04:50,696 classical computers fail to 76 00:04:51,016 --> 00:04:54,607 properly address and find a solution 77 00:04:54,960 --> 00:04:57,493 because of the nature of this problem, which is 78 00:04:57,589 --> 00:05:01,019 combinatorial. Consider that you want to 79 00:05:01,051 --> 00:05:04,706 find a drug compound. It has 80 00:05:04,963 --> 00:05:08,553 a kind of different combination between molecules 81 00:05:08,810 --> 00:05:11,920 and having constraints due to the 82 00:05:12,755 --> 00:05:16,512 intercellular and cellular environment inside the 83 00:05:16,512 --> 00:05:20,269 body. So that requires a lot of computation, 84 00:05:20,366 --> 00:05:24,091 a lot of parameters to consider. Quantum 85 00:05:24,155 --> 00:05:27,013 computing can provide a very 86 00:05:28,073 --> 00:05:31,669 promising capacity, both in terms of 87 00:05:31,685 --> 00:05:35,459 memory and computation, to address this kind 88 00:05:35,459 --> 00:05:39,052 of problem. Interesting. 89 00:05:39,309 --> 00:05:42,866 So where did you start? Like, how did you get into this? You went 90 00:05:42,914 --> 00:05:46,360 from— because you do have a lot of things going on, and you 91 00:05:46,376 --> 00:05:49,709 mentioned quantum machine learning, you mentioned quantum, you touched on 92 00:05:50,061 --> 00:05:53,218 drug discovery, you also touched on 93 00:05:54,629 --> 00:05:55,398 quantum biology. 94 00:05:59,340 --> 00:06:03,122 Wow. That's all I got to say. I'm suitably impressed. But 95 00:06:03,122 --> 00:06:04,981 how did you start in this field? Like, did you, 96 00:06:07,033 --> 00:06:10,784 you know, how did you get started with quantum computing? Did 97 00:06:10,832 --> 00:06:14,391 it start in— did you start in AI, ML, or did you start in 98 00:06:14,407 --> 00:06:17,357 medical drug research or something else entirely? 99 00:06:18,767 --> 00:06:22,502 So basically, if I wanted to give a little bit 100 00:06:22,518 --> 00:06:26,173 background, I always, from early childhood, was 101 00:06:26,350 --> 00:06:29,299 interested in quantum physics and 102 00:06:30,133 --> 00:06:33,371 physics in general. But my path in 103 00:06:33,723 --> 00:06:37,314 education education and professional life ended 104 00:06:37,426 --> 00:06:40,360 into machine learning and AI, which I really love. 105 00:06:41,674 --> 00:06:45,329 But for my PhD, I started 106 00:06:45,586 --> 00:06:48,776 to work on a, you know, my PhD was 107 00:06:48,968 --> 00:06:51,677 about a brain-like 108 00:06:52,799 --> 00:06:56,518 computational model. It has a kind of 3D 109 00:06:56,598 --> 00:06:58,714 structure that has 110 00:07:00,253 --> 00:07:03,812 coordinates like 1,000 111 00:07:03,812 --> 00:07:07,531 scaled-down coordinates of a real neuron in brain. 112 00:07:08,108 --> 00:07:11,555 And this brain-like computational 113 00:07:11,571 --> 00:07:15,225 model was very interesting 114 00:07:15,322 --> 00:07:18,864 for me to explore, specifically with the application of 115 00:07:19,281 --> 00:07:22,984 neurodegenerative disease like dementia. So when 116 00:07:23,096 --> 00:07:25,709 I was doing my PhD at the early years, I 117 00:07:25,982 --> 00:07:27,905 realized that there is a 118 00:07:29,124 --> 00:07:31,256 computational kind of 119 00:07:32,955 --> 00:07:36,706 issue for this huge structure. And when I 120 00:07:36,738 --> 00:07:40,492 was doing some research, and because I was 121 00:07:40,492 --> 00:07:44,169 interested in the quantum physics and quantum computing, 122 00:07:44,827 --> 00:07:47,798 I thought, why not using the benefit 123 00:07:48,681 --> 00:07:52,518 of both memory and computation 124 00:07:52,775 --> 00:07:55,521 of the quantum physics and quantum mechanics? 125 00:07:56,308 --> 00:07:59,985 However, it was at the beginning, it was like trying 126 00:08:00,161 --> 00:08:03,357 to getting inspiration from quantum 127 00:08:04,352 --> 00:08:08,021 mechanics, and I tried to improve that 128 00:08:08,278 --> 00:08:11,370 framework by trying to 129 00:08:11,883 --> 00:08:15,361 use concept of quantum mechanics and quantum physics. 130 00:08:15,953 --> 00:08:19,335 At the moment, what I'm doing, I'm trying to 131 00:08:19,335 --> 00:08:22,363 convert that concepts to a physical 132 00:08:22,860 --> 00:08:26,433 circuit, adapting to the physical circuits of quantum 133 00:08:26,449 --> 00:08:30,135 computers. So it is not just because when we talk about 134 00:08:30,873 --> 00:08:34,373 getting inspiration, especially in the AI machine 135 00:08:34,405 --> 00:08:38,147 learning. It means that we simplified some 136 00:08:38,243 --> 00:08:41,840 of the concept and we tried to 137 00:08:41,936 --> 00:08:45,693 benefit from the general idea. But right now I'm trying 138 00:08:45,790 --> 00:08:48,792 with the help of more knowledge that I acquire 139 00:08:49,804 --> 00:08:53,240 around quantum machine learning, I'm trying to converting 140 00:08:53,352 --> 00:08:56,772 those models to quantum 141 00:08:57,206 --> 00:09:00,160 circuits to be adaptable to 142 00:09:01,622 --> 00:09:05,090 an algorithm that is compatible with quantum 143 00:09:05,154 --> 00:09:08,382 computers. And because of, 144 00:09:08,992 --> 00:09:12,380 you know, when you are starting to solving a problem, 145 00:09:13,183 --> 00:09:16,732 specifically in the area of health, you 146 00:09:17,438 --> 00:09:20,184 realize that it is not just one single 147 00:09:20,634 --> 00:09:23,427 angle to look at. When I was 148 00:09:24,503 --> 00:09:28,303 Starting my journey, I wanted to diagnose, I 149 00:09:28,335 --> 00:09:31,992 wanted to predict the risk of getting 150 00:09:32,360 --> 00:09:35,824 dementia, different type of dementia actually, like 151 00:09:35,953 --> 00:09:38,759 Alzheimer's disease, frontotemporal 152 00:09:39,176 --> 00:09:42,512 dementia, Lewy body dementia, all those categories 153 00:09:42,768 --> 00:09:46,312 that the brain, the neuron are getting 154 00:09:46,456 --> 00:09:50,289 affected by aging and environmental factors. 155 00:09:50,722 --> 00:09:53,623 Then I realized that, okay, we 156 00:09:53,912 --> 00:09:57,038 understand and we predict it. What's the next step? 157 00:09:57,711 --> 00:10:01,382 So for the next step, it requires a treatment 158 00:10:01,382 --> 00:10:05,133 plan. For the treatment plan, we need to look 159 00:10:05,133 --> 00:10:08,724 at the medicine and precision medicine, 160 00:10:08,916 --> 00:10:12,058 if I want to be exact, so that we can help a 161 00:10:12,122 --> 00:10:15,617 person according to their kind of 162 00:10:16,450 --> 00:10:20,009 biological signature instead of providing one 163 00:10:20,073 --> 00:10:23,794 solution that one size fits all. So it was 164 00:10:24,692 --> 00:10:27,934 the beginning of my journey and how it evolves to— 165 00:10:28,704 --> 00:10:32,507 and I can say that although it's very interesting, but 166 00:10:33,373 --> 00:10:36,952 the more I acquire knowledge, 167 00:10:37,546 --> 00:10:41,333 the more I understand that I'm still in the beginning of the 168 00:10:41,381 --> 00:10:44,430 way and there are a lot to learn and to work on. 169 00:10:46,596 --> 00:10:49,532 You know, I'm a little curious to the culture that's going on 170 00:10:50,527 --> 00:10:54,250 where you are. Do you find— are there a lot are there a lot of 171 00:10:54,299 --> 00:10:57,220 women in the room with you? Are you one of few? 172 00:10:58,954 --> 00:11:02,325 How is the gender, how does gender play 173 00:11:02,341 --> 00:11:04,814 out in quantum for you? 174 00:11:07,302 --> 00:11:10,529 It's, if you ask me like 175 00:11:10,609 --> 00:11:14,430 maybe 5 years ago or maybe 10 years 176 00:11:14,526 --> 00:11:18,315 ago, I would say that it is obviously 177 00:11:18,315 --> 00:11:21,766 very dominant by male, not 178 00:11:21,766 --> 00:11:25,523 female. But recently I can see a lot of 179 00:11:25,699 --> 00:11:29,039 female are entering in these 180 00:11:29,376 --> 00:11:33,213 fields, machine learning, computer science, data science, and 181 00:11:33,309 --> 00:11:36,745 also physics. I know myself a lot of 182 00:11:37,242 --> 00:11:40,838 very intelligent ladies that are working 183 00:11:41,095 --> 00:11:44,788 in the photonic physics 184 00:11:44,868 --> 00:11:48,544 and quantum mechanics. And also 185 00:11:48,657 --> 00:11:52,430 it's still kind of, it's 186 00:11:52,430 --> 00:11:55,962 proportional to, you know, it's more oriented 187 00:11:55,962 --> 00:11:58,177 towards or more kind of 188 00:11:59,253 --> 00:12:02,881 dominated by male. But I think that 189 00:12:02,929 --> 00:12:06,285 little by little the balance is happening. 190 00:12:07,665 --> 00:12:11,454 I don't generally think female don't 191 00:12:11,454 --> 00:12:15,114 have capacities, but culture and 192 00:12:15,291 --> 00:12:18,485 also the passion. So one 193 00:12:18,902 --> 00:12:22,401 angle is that the society that you are living in, 194 00:12:22,915 --> 00:12:26,558 what they are promoting for you as a lady. They promote you 195 00:12:26,670 --> 00:12:30,009 to be going in a kind of softer 196 00:12:30,298 --> 00:12:33,796 areas of science. But some 197 00:12:34,166 --> 00:12:37,986 research, I think that I recently heard 198 00:12:38,018 --> 00:12:41,613 that there was a recent research around the mathematical 199 00:12:41,629 --> 00:12:45,287 ability of female 200 00:12:45,383 --> 00:12:49,009 and male. And it was like that in our 201 00:12:49,009 --> 00:12:52,747 countries, like I think that it was one of the Scandinavian 202 00:12:53,726 --> 00:12:56,726 countries that female were 203 00:12:57,175 --> 00:13:00,977 outcompeting male in terms of mathematical kind of 204 00:13:01,667 --> 00:13:05,469 ability and power. And it says a lot. It means that 205 00:13:06,175 --> 00:13:09,705 not, not, it is not something genetic, something related to your 206 00:13:09,705 --> 00:13:13,410 gender. It's just how you are 207 00:13:13,635 --> 00:13:16,874 cultivated, what, what, uh, the opportunity that your, 208 00:13:17,323 --> 00:13:20,835 your country, your, uh, culture, your society provides for you. 209 00:13:21,814 --> 00:13:25,630 Makes a lot of sense. I always thought that that was more less biology and 210 00:13:25,646 --> 00:13:28,661 more sociology. Yeah, I agree. 211 00:13:30,393 --> 00:13:34,034 You know, you mentioned quantum biology and Frank knows I 212 00:13:34,226 --> 00:13:38,027 love everything quantum biology. Like it's a, it's a little bit, a 213 00:13:38,027 --> 00:13:40,978 little bit of an issue for me now, but like I wanted to talk a 214 00:13:40,978 --> 00:13:44,472 little bit of your, of your brain is a computational model. 215 00:13:45,129 --> 00:13:48,607 And I, I wondered if— do you 216 00:13:48,655 --> 00:13:51,924 see there to be meaningful parallels between 217 00:13:52,357 --> 00:13:56,091 quantum systems and how the brain handles 218 00:13:56,283 --> 00:13:59,905 things like uncertainty or ambiguity or 219 00:14:00,674 --> 00:14:04,071 probability? Part of actually my, 220 00:14:04,231 --> 00:14:07,773 my thesis, because, you know, my, uh, 221 00:14:08,542 --> 00:14:11,732 The work that I've done in my PhD, it was around 222 00:14:12,630 --> 00:14:14,618 building a computational model that 223 00:14:15,034 --> 00:14:18,817 resembles biological brain. And at that 224 00:14:18,817 --> 00:14:22,408 time, I read an article from Sir 225 00:14:22,472 --> 00:14:26,095 Penrose. He's a, I think, 226 00:14:26,480 --> 00:14:29,942 Nobel laureate for, I 227 00:14:29,942 --> 00:14:33,036 think, quantum physics. And 228 00:14:33,549 --> 00:14:37,284 he suggested that maybe consciousness 229 00:14:37,461 --> 00:14:40,683 and the way that brain thinks can be 230 00:14:40,699 --> 00:14:43,039 modeled by the quantum 231 00:14:46,134 --> 00:14:49,821 concepts, quantum mechanics concepts. So, and they 232 00:14:49,821 --> 00:14:53,605 tried with his colleague, tried to go to microtubule 233 00:14:53,653 --> 00:14:56,587 to say that, yeah, there might be some 234 00:14:57,276 --> 00:15:01,092 kind of quantum mechanical thing happening 235 00:15:01,188 --> 00:15:04,170 there. But I want to answer you in this way. 236 00:15:05,212 --> 00:15:08,883 Quantum mechanics is the fundamental rules of our 237 00:15:08,883 --> 00:15:11,352 physical world. So our 238 00:15:11,368 --> 00:15:15,119 brain is, 239 00:15:15,376 --> 00:15:18,726 includes molecules, cells, 240 00:15:18,822 --> 00:15:22,541 molecules have atoms, atoms have subatomic kind 241 00:15:22,589 --> 00:15:26,212 of domain. And when we go from the 242 00:15:26,212 --> 00:15:29,402 quantum realm to the physical or classical realm, 243 00:15:30,781 --> 00:15:34,083 because of confinement with 244 00:15:34,724 --> 00:15:38,539 many bodies, many other kind 245 00:15:38,539 --> 00:15:41,922 of environmental association with other other 246 00:15:41,922 --> 00:15:44,485 molecules, other atoms, the 247 00:15:44,950 --> 00:15:48,410 quantum nature faded. Doesn't 248 00:15:48,971 --> 00:15:52,784 go away, it faded. And it's like that you have a general rule in 249 00:15:52,880 --> 00:15:56,389 the quantum realm. When you go to the classical realm, 250 00:15:56,902 --> 00:16:00,395 it becomes more kind of a special case. So basically, 251 00:16:00,731 --> 00:16:04,368 it is true to say that it is governed by quantum mechanical 252 00:16:04,416 --> 00:16:08,069 rule. However, I, because I'm 253 00:16:08,213 --> 00:16:12,060 very interested in consciousness and I tried to with a 254 00:16:12,060 --> 00:16:15,703 very— at that time, I was a little bit ignorant. I thought that 255 00:16:16,137 --> 00:16:19,299 we can computationally model consciousness 256 00:16:20,311 --> 00:16:23,328 while we don't know what consciousness actually is. 257 00:16:24,019 --> 00:16:26,507 So I think 258 00:16:28,497 --> 00:16:31,933 we can't model consciousness at this 259 00:16:32,751 --> 00:16:35,962 scale because it 260 00:16:36,042 --> 00:16:39,669 requires not just considering the brain as an 261 00:16:39,669 --> 00:16:43,149 individual physical system. Our 262 00:16:43,182 --> 00:16:46,020 brain doesn't evolve just by 263 00:16:46,373 --> 00:16:50,046 isolation, right? So you learn by having 264 00:16:50,238 --> 00:16:53,397 social connection with others. Your 265 00:16:54,071 --> 00:16:57,166 mind, your consciousness, shaped by 266 00:16:57,695 --> 00:17:01,208 environments, by communication with 267 00:17:01,288 --> 00:17:04,977 other individuals or other entities in 268 00:17:05,073 --> 00:17:08,312 the world. So it is not an isolated model that you can 269 00:17:09,066 --> 00:17:12,787 provide a computational, you know, mathematical model to 270 00:17:12,787 --> 00:17:15,963 say that, yeah, consciousness arise from this physical 271 00:17:16,236 --> 00:17:20,070 neuronal activity. If we want to truly model such a 272 00:17:20,230 --> 00:17:24,080 thing, we need to model it 273 00:17:24,080 --> 00:17:27,801 within this complex psychosocial 274 00:17:27,993 --> 00:17:31,169 association with others, with the nature, with the 275 00:17:31,603 --> 00:17:34,827 environment. And that is the correct way to 276 00:17:35,292 --> 00:17:39,126 look at. And I think that But right now we don't have the capacity 277 00:17:39,319 --> 00:17:43,010 in terms of computation to test 278 00:17:43,010 --> 00:17:44,053 such an idea. 279 00:17:46,925 --> 00:17:50,424 Wow, that's a lot. I mean, that's a lot to take in. 280 00:17:51,017 --> 00:17:54,708 But you're right. Like, you know, if you want to use 281 00:17:54,949 --> 00:17:58,768 AI terms for this, right, humans are not 282 00:17:59,121 --> 00:18:02,844 neural networks that exist in isolation. They interact with 283 00:18:02,909 --> 00:18:06,648 other neural networks. Um, yeah, I 284 00:18:06,680 --> 00:18:09,697 mean, yeah, there's a lot, there's a lot to unpack there, and I just can't 285 00:18:09,713 --> 00:18:12,795 imagine what the computational power to simulate that would be. 286 00:18:13,806 --> 00:18:17,626 It's probably beyond, beyond our company, beyond my 287 00:18:17,674 --> 00:18:20,964 comprehension, that's for sure. Oh, go ahead. 288 00:18:21,269 --> 00:18:25,040 Yeah, yeah, if we want, because, you know, there was an attempt, I think 289 00:18:25,040 --> 00:18:28,362 that's 4 years ago, to build a 290 00:18:28,362 --> 00:18:31,685 spiking neural network like computers 291 00:18:32,198 --> 00:18:35,293 that can handle, like we have 292 00:18:35,389 --> 00:18:38,708 83 billion neurons in our brain and 293 00:18:38,852 --> 00:18:42,314 consider that the connectivity goes to the roof. 294 00:18:42,635 --> 00:18:46,178 So they tried to build such a machine. Even 295 00:18:46,322 --> 00:18:48,839 such a machine requires a lot of 296 00:18:50,394 --> 00:18:54,177 fine-tuning, a lot of energy to make 297 00:18:54,177 --> 00:18:57,816 it run again. Still, there is, I was 298 00:18:57,912 --> 00:19:00,525 looking at the success and/or 299 00:19:01,454 --> 00:19:05,302 news about how they go on with that idea. Idea. 300 00:19:05,606 --> 00:19:09,261 But you're right, the problem is the idea might be there, 301 00:19:09,389 --> 00:19:12,067 but right now we are limited with the 302 00:19:12,996 --> 00:19:16,731 maybe power or at some extent 303 00:19:16,844 --> 00:19:19,601 knowledge of how to build or 304 00:19:20,611 --> 00:19:24,282 manipulate those information. Also too, 305 00:19:24,298 --> 00:19:27,889 maybe this will be resolved at some point in the future. I mean, I'm still 306 00:19:27,889 --> 00:19:31,030 amazed at the fact that you're on the other side of the planet and we're 307 00:19:31,030 --> 00:19:34,691 having a real real-time video conversation. Like, that wasn't that long ago that was 308 00:19:35,847 --> 00:19:39,427 impossible to, impractical to, now it's an everyday occurrence. 309 00:19:41,707 --> 00:19:44,903 I think it's fantastic having 3 separate countries on the call right now. It's not 310 00:19:44,903 --> 00:19:48,612 even, it's not even an issue, right? That's fantastic. 311 00:19:50,538 --> 00:19:54,376 So when you speak to organizations about quantum, what's the biggest 312 00:19:54,376 --> 00:19:55,869 misconception you still hear? 313 00:19:59,046 --> 00:20:01,371 One thing that is sad because 314 00:20:03,295 --> 00:20:07,096 last year I attended a conference and I 315 00:20:07,368 --> 00:20:11,040 was telling that because my work mainly 316 00:20:11,056 --> 00:20:14,680 has a research nature in it. It's not like that 317 00:20:14,921 --> 00:20:18,753 there is a solution ready, I'm just trying to build an 318 00:20:18,753 --> 00:20:22,249 app around that. I was asking 319 00:20:22,618 --> 00:20:26,018 that question, How I can build some 320 00:20:26,146 --> 00:20:29,979 sort of research group that people 321 00:20:29,995 --> 00:20:33,732 can help me to build this idea, 322 00:20:33,748 --> 00:20:36,699 especially in drug discovery. You know, in drug discovery, 323 00:20:37,774 --> 00:20:41,318 we have the problem of a huge 324 00:20:41,367 --> 00:20:44,478 database of molecules and information 325 00:20:44,654 --> 00:20:48,135 related to drug compounds. We just 326 00:20:48,135 --> 00:20:51,743 scratched the surface, like 10%, and from that 327 00:20:51,968 --> 00:20:55,095 10%, it takes like, uh, 10 to 328 00:20:55,159 --> 00:20:58,720 15 years, or sometimes 20 years, to 329 00:20:58,944 --> 00:21:02,312 do some sort of research on what compound 330 00:21:03,274 --> 00:21:07,091 suits for this specific disease. And after that, going 331 00:21:07,187 --> 00:21:10,940 to virtual screening, then going to test, uh, phases, 332 00:21:11,116 --> 00:21:14,901 and going to regulation phases, then to, uh, release 333 00:21:14,901 --> 00:21:18,557 to the market. So it is, first of all, a long, uh, kind of, 334 00:21:18,814 --> 00:21:22,647 uh, period of time. And 335 00:21:22,823 --> 00:21:26,656 also it costs a lot of money, billions and billions 336 00:21:26,752 --> 00:21:30,376 of dollars. It's not like that if you fail the 337 00:21:30,376 --> 00:21:33,375 first phase of research, it's just, just maybe a 338 00:21:34,305 --> 00:21:37,801 couple of thousands of dollars. It's billions of billions of dollars. 339 00:21:38,378 --> 00:21:42,066 And 90% of that's considered that I 340 00:21:42,066 --> 00:21:45,851 mentioned, 10% of those information we are looking 341 00:21:45,995 --> 00:21:49,352 at. 90% of those trials 342 00:21:49,689 --> 00:21:53,383 fail at the earliest stage. Some of them fail 343 00:21:53,463 --> 00:21:56,579 at the final stage. It means that you already 344 00:21:57,478 --> 00:22:00,915 invest a lot of money during this research stage. 345 00:22:02,055 --> 00:22:05,251 Then there is a potential. Quantum 346 00:22:05,267 --> 00:22:08,784 computing can provide a potential to 347 00:22:09,395 --> 00:22:12,960 solve a problem that a classical machine 348 00:22:13,137 --> 00:22:16,341 can solve like in 300 years. Within 349 00:22:17,133 --> 00:22:20,849 a couple of hours. Even though we are 350 00:22:20,978 --> 00:22:24,009 in the noisy intermediate-scale quantum 351 00:22:24,606 --> 00:22:28,248 era, it can be handled. Right now, I 352 00:22:28,248 --> 00:22:31,361 think that quantum— 353 00:22:32,991 --> 00:22:36,590 I think that it was Atom Quantum. Yes, they 354 00:22:36,800 --> 00:22:40,269 propose quantum computing that can handle more than 355 00:22:40,269 --> 00:22:43,881 1,000 qubits. When we 356 00:22:44,122 --> 00:22:47,571 say 1,000 qubits, maybe it's like not 357 00:22:47,876 --> 00:22:51,662 big enough for handling information. But if 358 00:22:51,807 --> 00:22:55,593 we add the quantum physical, 359 00:22:56,219 --> 00:22:59,796 quantum mechanical concept to that, it provides 360 00:22:59,909 --> 00:23:03,294 us a huge space to store 361 00:23:03,984 --> 00:23:07,369 and compute information, process 362 00:23:07,449 --> 00:23:11,171 information, so that can help to reduce 363 00:23:11,204 --> 00:23:13,768 the time and also 364 00:23:15,162 --> 00:23:16,764 save some budget. If you 365 00:23:18,976 --> 00:23:22,196 fail earlier, you can save a lot of budget. But the 366 00:23:22,276 --> 00:23:24,535 problem is, first of all, 367 00:23:28,028 --> 00:23:31,217 quantum computing doesn't suit every 368 00:23:31,249 --> 00:23:34,710 problem. You should have a supermassive 369 00:23:34,854 --> 00:23:38,635 to work with that. It's not 370 00:23:38,683 --> 00:23:41,747 like that. If you can classically solve a 371 00:23:42,389 --> 00:23:45,761 problem, you don't go for the quantum computer, 372 00:23:45,906 --> 00:23:49,663 quantum computing and quantum computers, because 373 00:23:49,695 --> 00:23:52,922 there are two angles. One of them requires a 374 00:23:52,970 --> 00:23:56,519 specialist. It's a cost to build 375 00:23:56,535 --> 00:24:00,244 such an algorithm. Those algorithms are a little 376 00:24:00,453 --> 00:24:04,065 bit complicated. You need to both know 377 00:24:04,627 --> 00:24:08,398 quantum mechanics rules. You need to know computer 378 00:24:08,398 --> 00:24:11,430 science, and also if it is like 379 00:24:11,767 --> 00:24:14,783 information technology, if you are working with some piece of 380 00:24:15,537 --> 00:24:18,761 information. And the other fact is that 381 00:24:20,028 --> 00:24:23,525 how many— because 382 00:24:24,841 --> 00:24:28,290 right now there are a lot of companies that are using, especially 383 00:24:28,290 --> 00:24:31,787 pharmaceutical companies, or this 384 00:24:33,086 --> 00:24:36,649 cryptographies that they try to use quantum 385 00:24:36,954 --> 00:24:40,726 computers, but it is not widely accepted, uh, 386 00:24:41,432 --> 00:24:45,252 again due to the lack of, uh, experts in 387 00:24:45,300 --> 00:24:48,141 this field. And also, uh, 388 00:24:48,831 --> 00:24:52,667 quantum computers are expensive, not, uh, easily accessible. 389 00:24:53,229 --> 00:24:56,503 Um, uh, although there are, uh, Google and 390 00:24:56,680 --> 00:25:00,290 IBM and also, uh, another major 391 00:25:00,290 --> 00:25:03,305 player that provides cloud quantum 392 00:25:03,433 --> 00:25:06,946 computers, it still, it is not widely used. So 393 00:25:07,443 --> 00:25:10,603 when I talk about, let's move to quantum 394 00:25:11,710 --> 00:25:15,559 solution, there are 3 major problem. 395 00:25:16,089 --> 00:25:19,729 One of them, can you find 396 00:25:19,729 --> 00:25:23,194 the expert for me that can do that? Expert are 397 00:25:23,675 --> 00:25:25,696 rare right now, but it is growing field. 398 00:25:27,477 --> 00:25:30,672 How much should I pay for this kind of 399 00:25:32,406 --> 00:25:36,051 supercomputers or computers, quantum computers? 400 00:25:36,710 --> 00:25:40,467 And besides of the business value that you need 401 00:25:40,563 --> 00:25:43,389 to convince. And also, 402 00:25:44,513 --> 00:25:48,270 if I solve it with the classical computers, why should I 403 00:25:48,270 --> 00:25:51,466 bother to go? And there is no kind of 404 00:25:53,022 --> 00:25:56,710 justification to go there. But yes, it's 405 00:25:56,710 --> 00:26:00,414 like it was difficult, and I, I thought that 406 00:26:00,719 --> 00:26:04,214 possibly, uh, it requires 407 00:26:04,567 --> 00:26:07,790 more kind of, uh, like 408 00:26:08,496 --> 00:26:12,103 podcast that you are providing, more kind of, um, 409 00:26:13,226 --> 00:26:16,465 public awareness in terms of technology and the 410 00:26:16,513 --> 00:26:20,249 capability so that business owners are ready to 411 00:26:20,281 --> 00:26:24,051 move. But again, it's, it's, 412 00:26:24,436 --> 00:26:28,094 it's, it's, whenever we try to 413 00:26:28,302 --> 00:26:31,158 provide a solution, we always say that 414 00:26:31,575 --> 00:26:35,088 quantum supremacy over classical computers. 415 00:26:35,601 --> 00:26:38,986 That's the most important part. Okay, 416 00:26:39,676 --> 00:26:43,366 so what part of drug discovery, what part of the drug 417 00:26:43,366 --> 00:26:46,767 discovery pipeline do you think quantum will impact first? 418 00:26:47,232 --> 00:26:51,045 Target identification, molecular modeling, 419 00:26:51,301 --> 00:26:54,746 or optimization? I think that all of the 420 00:26:54,746 --> 00:26:58,382 areas. So the nature of this— then if 421 00:26:58,767 --> 00:27:02,532 you look at the problem, the nature of the problem is quantum mechanical. 422 00:27:02,852 --> 00:27:04,647 You are looking at, uh, 423 00:27:05,047 --> 00:27:08,251 molecules, how they bind together, how their, 424 00:27:09,261 --> 00:27:12,833 uh, subatomic level, um, tries 425 00:27:12,945 --> 00:27:16,582 to provide the capacity of binding with, uh, between the 426 00:27:16,951 --> 00:27:20,273 drug itself and also the 427 00:27:20,562 --> 00:27:23,935 drug compound and target protein in terms of my, my 428 00:27:24,882 --> 00:27:28,560 own research area, which mostly is 429 00:27:28,833 --> 00:27:32,655 the protein and amino acid inside the brain. 430 00:27:33,233 --> 00:27:36,638 So the nature is quantum mechanical. It's all 431 00:27:36,702 --> 00:27:40,139 based quantum chemistry, quantum thermodynamics, 432 00:27:40,219 --> 00:27:43,591 quantum mechanics. And from the 433 00:27:43,672 --> 00:27:47,253 optimization, again, this drug discovery problem 434 00:27:48,088 --> 00:27:51,877 is not, this is the place that 435 00:27:52,182 --> 00:27:55,795 quantum supremacy plays an important role. You cannot 436 00:27:55,972 --> 00:27:58,991 solve this problem with 437 00:27:59,023 --> 00:28:02,796 classical computers. As I mentioned, 438 00:28:02,812 --> 00:28:05,815 it is a combinatorial problem. So 439 00:28:06,634 --> 00:28:09,958 you may need maybe 100 years to 440 00:28:10,408 --> 00:28:13,972 look at not even on a specific 441 00:28:14,663 --> 00:28:18,094 combination, of the drug molecules. 442 00:28:19,072 --> 00:28:22,421 And also for the optimization, 443 00:28:22,854 --> 00:28:26,188 we have quantum optimization that they can 444 00:28:26,845 --> 00:28:30,403 find the equilibrium or the 445 00:28:30,451 --> 00:28:34,233 lowest energy point for the 446 00:28:34,490 --> 00:28:38,160 structure of the compound. Because it's not just, you know, 447 00:28:39,746 --> 00:28:43,529 connecting atoms or molecules to build a new compound. You need 448 00:28:43,529 --> 00:28:47,315 to look at the structure, 449 00:28:48,086 --> 00:28:50,831 and it can define a lot about the 450 00:28:51,779 --> 00:28:54,364 hydrophilic, hydrophobic, and 451 00:28:55,809 --> 00:28:58,474 the kind of characteristic that can 452 00:28:58,924 --> 00:29:02,264 contribute that molecules. Again, 453 00:29:02,376 --> 00:29:05,652 it's because I'm more kind of 454 00:29:05,764 --> 00:29:08,943 expert in my own area. That molecules can 455 00:29:09,457 --> 00:29:13,035 pass the broad brain barrier 456 00:29:13,853 --> 00:29:17,221 and reach to the specific 457 00:29:17,237 --> 00:29:21,070 protein and try to, you know, solve 458 00:29:21,070 --> 00:29:24,518 the issue. And besides of that, this 459 00:29:24,550 --> 00:29:28,046 optimization is not just about the 460 00:29:28,046 --> 00:29:30,708 constraints that we see in the 461 00:29:31,911 --> 00:29:35,519 molecule compound physical shape. So 462 00:29:35,680 --> 00:29:38,949 consider that Specifically for the drug that 463 00:29:39,446 --> 00:29:42,475 requires to go through brain. Brain has a 464 00:29:43,052 --> 00:29:46,737 kind of protective lattice. It's called a 465 00:29:46,978 --> 00:29:50,519 blood-brain barrier. It doesn't allow everything passes through. 466 00:29:51,288 --> 00:29:54,285 It has a filter. So first of all, you need to be 467 00:29:54,285 --> 00:29:57,842 successfully passed through this barrier. And 468 00:29:57,939 --> 00:30:00,807 also drugs are not just a kind of, 469 00:30:02,121 --> 00:30:05,814 you know, a final solution solves the problem, right? They 470 00:30:05,958 --> 00:30:09,700 have toxicity. They can alter and change the 471 00:30:10,085 --> 00:30:13,859 cellular environment. They can help, but 472 00:30:13,923 --> 00:30:17,231 there are always side effects as well. How to 473 00:30:17,857 --> 00:30:21,149 increase the benefit and decrease 474 00:30:21,149 --> 00:30:24,923 the kind of harm 475 00:30:24,987 --> 00:30:28,665 that the drug can cause to the body is important. All of 476 00:30:28,777 --> 00:30:32,005 those parameters It can create a 477 00:30:32,134 --> 00:30:35,539 huge problem space that cannot 478 00:30:35,667 --> 00:30:38,944 easily solved by classical computers. 479 00:30:39,683 --> 00:30:43,088 Quantum computers can provide a 480 00:30:43,136 --> 00:30:46,734 very big space. You know, theoretically, if you 481 00:30:46,830 --> 00:30:50,380 heard about Hilbert space, if you heard about a 482 00:30:50,461 --> 00:30:54,219 kind of multidimensional space that you can look at 483 00:30:54,476 --> 00:30:58,026 possibilities and try to find 484 00:30:58,186 --> 00:31:01,671 the best solution 485 00:31:02,892 --> 00:31:06,425 quantum computers has. I think that ultimate 486 00:31:06,425 --> 00:31:08,770 solution for this specific problem. 487 00:31:11,179 --> 00:31:15,017 Interesting. I would say 488 00:31:15,017 --> 00:31:18,695 anybody listening for at least 5 minutes understands how incredibly 489 00:31:18,727 --> 00:31:22,566 brilliant you are. Yeah. Yeah. I mean, it won't take 5 minutes. Kind 490 00:31:22,582 --> 00:31:26,259 of. I know, right? It'll take like exactly 30 seconds. So 491 00:31:26,741 --> 00:31:29,567 you also are an entrepreneur. So 492 00:31:30,771 --> 00:31:34,544 what kind of leadership mindset does quantum 493 00:31:34,544 --> 00:31:38,173 demand? One of them is that 494 00:31:38,783 --> 00:31:42,428 to be open to failure. It's not like that, you know, 495 00:31:42,637 --> 00:31:45,993 every attempt is successful. The other thing 496 00:31:46,250 --> 00:31:49,301 is that you need to— if 497 00:31:49,397 --> 00:31:53,058 first of all, quantum mechanics is peculiar in 498 00:31:53,090 --> 00:31:55,943 nature. Right? It's like different than 499 00:31:56,937 --> 00:31:59,805 the classical tangible 500 00:31:59,885 --> 00:32:02,994 environment that you're facing. 501 00:32:03,747 --> 00:32:07,080 The other thing is that you need to be patient. The field is 502 00:32:07,257 --> 00:32:10,926 growing, but slowly growing. It's not like that. So 503 00:32:11,503 --> 00:32:15,333 when I want to write an algorithm for classical machine, 504 00:32:16,070 --> 00:32:19,339 everything is already tested. 505 00:32:21,359 --> 00:32:23,988 There are a little bit innovation, 506 00:32:24,645 --> 00:32:26,441 although we saw a big 507 00:32:30,176 --> 00:32:33,846 leap in 2017 with language models, 508 00:32:34,039 --> 00:32:37,229 large language models. But anyway, it's like that 509 00:32:37,630 --> 00:32:40,916 gradually adding to the already built-in 510 00:32:40,916 --> 00:32:44,378 foundation. For the quantum mechanics and quantum 511 00:32:44,523 --> 00:32:47,600 computers, especially quantum machine learning, it's still 512 00:32:47,664 --> 00:32:51,448 new. 2019 quantum 513 00:32:51,816 --> 00:32:55,360 computers being used, like, 514 00:32:57,108 --> 00:33:00,940 widely accepted and used for research field. And 515 00:33:00,956 --> 00:33:03,217 right now we are entering the era that 516 00:33:04,019 --> 00:33:07,482 pharmaceutical companies and also finance 517 00:33:07,610 --> 00:33:11,042 and also 518 00:33:13,238 --> 00:33:17,055 companies that working on creating new 519 00:33:17,071 --> 00:33:20,311 materials, new Finding new 520 00:33:20,456 --> 00:33:24,261 compounds for physical materials. Consider a 521 00:33:24,309 --> 00:33:28,002 scenario that you are able to 522 00:33:28,950 --> 00:33:32,209 use this quantum computing power 523 00:33:32,883 --> 00:33:36,239 to build a material that responds 524 00:33:36,432 --> 00:33:40,157 to the environment, right? If you 525 00:33:40,687 --> 00:33:43,368 learn the property of the 526 00:33:44,155 --> 00:33:47,914 material, the molecule, you can build at 527 00:33:47,914 --> 00:33:51,367 nanoscale, build a kind of material that can 528 00:33:51,367 --> 00:33:55,110 respond to the environment, a physical material. It can change 529 00:33:55,608 --> 00:33:59,190 the architecture and construction 530 00:34:00,218 --> 00:34:03,687 science. And also what I'm working 531 00:34:03,800 --> 00:34:07,365 right now, as I mentioned, it was part of the drug 532 00:34:07,430 --> 00:34:10,929 discovery. I'm working on 533 00:34:12,004 --> 00:34:15,643 building a kind of physical lattice 534 00:34:16,172 --> 00:34:19,731 to play the role of blood-brain 535 00:34:19,859 --> 00:34:23,081 barrier so that when we are doing some 536 00:34:24,332 --> 00:34:27,955 drug virtual screening, before we go to the animal 537 00:34:27,955 --> 00:34:31,354 test and, you know, human test or using 538 00:34:31,706 --> 00:34:34,255 other models, we have a physical kind of 539 00:34:35,474 --> 00:34:38,985 model that can show if this drug can have 540 00:34:39,081 --> 00:34:42,769 ability to have 541 00:34:42,769 --> 00:34:46,202 ability to pass this barrier. So generally, 542 00:34:47,037 --> 00:34:50,662 it's like that. First of all, I 543 00:34:50,710 --> 00:34:54,545 think being open to failure and know that the field is 544 00:34:54,577 --> 00:34:58,170 growing slowly right now, but you 545 00:34:58,170 --> 00:35:01,251 never know. There might be, once the 546 00:35:02,502 --> 00:35:05,678 full error-tolerant quantum machines are 547 00:35:05,807 --> 00:35:08,502 available, I think that it will be a huge 548 00:35:10,042 --> 00:35:13,443 thing for humanity because it gives us a 549 00:35:13,491 --> 00:35:17,148 full kind of power to use this 550 00:35:17,420 --> 00:35:20,516 massive memory and also computation. 551 00:35:21,911 --> 00:35:25,150 Generally, I think another 552 00:35:26,401 --> 00:35:29,849 important aspect is that 553 00:35:30,827 --> 00:35:33,922 we shouldn't think of science as silos. 554 00:35:34,516 --> 00:35:37,804 One thing is that quantum computers quantum 555 00:35:37,804 --> 00:35:40,181 computing, and also 556 00:35:40,374 --> 00:35:43,858 machine learning. If you look at it as 557 00:35:43,906 --> 00:35:47,438 silos, they can do a little, 558 00:35:47,920 --> 00:35:51,067 but when you try to look at this as part of a big 559 00:35:51,212 --> 00:35:54,744 picture, you can solve many problems 560 00:35:55,708 --> 00:35:58,806 by just combining different angles of 561 00:35:59,625 --> 00:36:01,167 this physical world together. 562 00:36:03,493 --> 00:36:07,004 You know, one of my favorite TV shows of all time was this UK science 563 00:36:07,004 --> 00:36:09,313 show called, um, Connections. 564 00:36:11,493 --> 00:36:15,132 And it's really old, like it was old when I 565 00:36:15,132 --> 00:36:18,963 saw it in the '90s, I think. But basically it shows kind of historically 566 00:36:18,963 --> 00:36:22,650 how, um, things we take for granted today have a 567 00:36:22,650 --> 00:36:25,969 connection, right? Hence the name, right? So one of them was the, 568 00:36:26,402 --> 00:36:30,105 um, perfume from the gasoline spray, 569 00:36:30,810 --> 00:36:34,401 this is like the 1700s, 1800s, to the carburetor, 570 00:36:36,100 --> 00:36:39,947 right? Because basically you had to atomize the gasoline, right? So kind of like, and 571 00:36:40,027 --> 00:36:43,778 how historically these people interacted 572 00:36:43,810 --> 00:36:47,626 with one another from different disciplines, right? And I think that was, I 573 00:36:47,706 --> 00:36:51,216 haven't seen this in a very long time, but that was 574 00:36:51,264 --> 00:36:54,935 basically kind of the gist of the atomizer connection to the carburetor. 575 00:36:55,674 --> 00:36:58,788 Was, you know, people were tinkering around with this, and 576 00:36:58,788 --> 00:37:02,624 then they were— somebody was at a party, and I guess they 577 00:37:02,640 --> 00:37:06,428 were selling perfumes, and he saw them pump the thing and spray 578 00:37:06,428 --> 00:37:10,184 it, and he's like, that's it, you know, kind of like little 579 00:37:10,200 --> 00:37:13,957 moments like that. That's cross-pollination is always fun like that, you know. 580 00:37:17,809 --> 00:37:21,340 No, absolutely. So, hmm, so what kind of 581 00:37:21,340 --> 00:37:25,145 advice would you give someone who's mid-career but 582 00:37:25,193 --> 00:37:27,264 wants to pivot into quantum? 583 00:37:31,471 --> 00:37:35,244 I think last year I had a talk at university. 584 00:37:35,404 --> 00:37:38,696 One of the students said that because they were 585 00:37:39,803 --> 00:37:43,641 doing quantum mechanics and quantum computing courses, 586 00:37:44,138 --> 00:37:47,943 said that what is the future? What is the current market for 587 00:37:47,991 --> 00:37:51,583 us if we want to enter? I think that one 588 00:37:51,679 --> 00:37:55,269 important aspect is you need to 589 00:37:55,381 --> 00:37:59,147 be, first of all, passionate about the field that you 590 00:37:59,211 --> 00:38:02,512 are entering. If it is your passion and there is no 591 00:38:03,874 --> 00:38:07,688 kind of capacity at the market at the moment, 592 00:38:08,425 --> 00:38:11,053 you will make that passion 593 00:38:11,967 --> 00:38:15,220 by thinking of using it in 594 00:38:15,798 --> 00:38:19,236 other forms, not just directly, you know, going 595 00:38:19,268 --> 00:38:23,044 to a kind of building infrastructure for 596 00:38:23,044 --> 00:38:26,868 the whole pipeline of the quantum computers 597 00:38:27,125 --> 00:38:30,644 and quantum mechanics. But the thing is that 598 00:38:32,106 --> 00:38:35,367 if I want to make an analogy, when I 599 00:38:35,416 --> 00:38:39,079 started my career at artificial intelligence, I 600 00:38:39,175 --> 00:38:42,773 was in my own country and it was like 20 years 601 00:38:42,869 --> 00:38:46,480 ago, it was at the beginning of 602 00:38:46,496 --> 00:38:49,625 artificial intelligence. There wasn't 603 00:38:50,412 --> 00:38:54,119 many places that use, actually no places, no company 604 00:38:54,520 --> 00:38:57,938 or ordinary company were using 605 00:38:58,115 --> 00:39:01,742 artificial intelligence. And I remember that my mother said that, 606 00:39:02,624 --> 00:39:05,577 change your path to something that can 607 00:39:05,577 --> 00:39:09,364 be used. You cannot find a job. And I said that, 608 00:39:09,750 --> 00:39:13,471 no, I like this one. So Little by little, 609 00:39:13,599 --> 00:39:16,452 by finding research institute that are working there, 610 00:39:17,526 --> 00:39:21,101 I start to expand my knowledge work. And then 611 00:39:21,470 --> 00:39:24,484 there was a time that right now every 612 00:39:24,981 --> 00:39:28,411 place demands for AI machine learning 613 00:39:28,539 --> 00:39:31,745 skills. So I think that first is that 614 00:39:32,916 --> 00:39:36,539 make sure that this is your path. If you like it, 615 00:39:36,859 --> 00:39:40,481 you can find Although at the moment it's not 616 00:39:40,481 --> 00:39:44,279 too many, you can find places that you can 617 00:39:45,337 --> 00:39:49,038 use your knowledge, but be aware of the 618 00:39:49,343 --> 00:39:52,612 growth of this field. There is a promise of 619 00:39:53,124 --> 00:39:55,816 within 10 to 15 years 620 00:39:57,211 --> 00:40:00,399 fully tolerant quantum computers will be 621 00:40:00,480 --> 00:40:03,845 available. And we are right now, we 622 00:40:03,925 --> 00:40:06,316 achieved 1,000 qubits. 623 00:40:07,617 --> 00:40:10,780 And when I'm saying about qubits 624 00:40:10,925 --> 00:40:13,895 and these strange things, 625 00:40:14,393 --> 00:40:17,540 mainly I'm emphasizing on the 626 00:40:18,054 --> 00:40:21,120 capacity of storing knowledge, storing 627 00:40:21,217 --> 00:40:24,653 information, processing them, and 628 00:40:24,653 --> 00:40:28,201 extracting knowledge. So I think the 629 00:40:28,201 --> 00:40:31,960 main you know, advice 630 00:40:31,960 --> 00:40:34,756 that I can give to people is that 631 00:40:36,154 --> 00:40:39,320 if it is your passion, stick to that. And you can 632 00:40:39,818 --> 00:40:43,128 either by going to research 633 00:40:43,530 --> 00:40:47,033 fields, research companies, you can, 634 00:40:48,013 --> 00:40:51,837 you know, add more to your knowledge, help to build more 635 00:40:52,287 --> 00:40:55,501 kind of steps to this ladder. 636 00:40:55,966 --> 00:40:59,237 And don't be kind of— 637 00:41:01,017 --> 00:41:04,672 don't feel like that if this is my 638 00:41:04,945 --> 00:41:08,648 area, my field, there is not enough 639 00:41:08,808 --> 00:41:12,111 places to work in. Just wait. It's like that 640 00:41:12,720 --> 00:41:16,456 maybe within— even though maybe earlier, there 641 00:41:16,568 --> 00:41:20,047 will be huge demand of quantum 642 00:41:20,256 --> 00:41:24,040 computing skills and also quantum 643 00:41:24,040 --> 00:41:26,828 machine learning. Alongside with 644 00:41:28,030 --> 00:41:31,587 the quantum mechanical skills for building 645 00:41:31,635 --> 00:41:34,920 such a machine. So for 646 00:41:34,920 --> 00:41:38,301 quantum computers, we are not just, 647 00:41:39,006 --> 00:41:42,596 you know, we need to combine 3 different areas, 648 00:41:43,974 --> 00:41:47,547 quantum physics, nanotechnology, and also 649 00:41:49,101 --> 00:41:52,658 computer science to build such a 650 00:41:52,883 --> 00:41:56,696 beautiful, powerful machine, upon that we need to 651 00:41:56,712 --> 00:42:00,173 have skills about quantum 652 00:42:00,237 --> 00:42:03,458 information technology that by itself 653 00:42:03,858 --> 00:42:07,351 categorizes into different kind of 654 00:42:08,328 --> 00:42:11,757 classes, quantum computing, 655 00:42:12,013 --> 00:42:15,635 quantum communication, and also 656 00:42:16,580 --> 00:42:20,249 quantum sensing. So it's not just one field, it has a broad 657 00:42:20,826 --> 00:42:24,663 use it. I think that mainly is, 658 00:42:24,920 --> 00:42:28,757 uh, being patient and finding the correct 659 00:42:28,870 --> 00:42:32,466 institution or a place to, to start with. 660 00:42:32,707 --> 00:42:36,175 That's a good way 661 00:42:36,560 --> 00:42:40,269 to put it. That's, um, definitely would love to have you 662 00:42:40,269 --> 00:42:42,613 back on the show because there's a lot to unpack. We're going to need more 663 00:42:42,613 --> 00:42:46,450 than an hour, uh, to, to go through it 664 00:42:46,450 --> 00:42:50,222 because you have, you have your— you're one of the, I think, few people that 665 00:42:50,222 --> 00:42:53,957 we've spoken to that has a very unique perspective on multiple aspects of 666 00:42:54,695 --> 00:42:58,446 how quantum computing can assist in both medicine, 667 00:42:58,478 --> 00:43:02,277 AI, and biology. I think it's 668 00:43:02,325 --> 00:43:05,836 an interesting— it's an interesting take. And 669 00:43:06,236 --> 00:43:08,224 I think you have a very unique perspective on this. 670 00:43:10,164 --> 00:43:13,786 Thank you. Yeah, I'm 671 00:43:13,786 --> 00:43:17,041 just— I'm just unpacking this. There's just a lot. So, 672 00:43:17,796 --> 00:43:20,957 and I also like your idea that, you know, and you stuck with AI 673 00:43:21,792 --> 00:43:25,339 way before AI was cool. You said 20 years ago, and I, 674 00:43:25,660 --> 00:43:28,838 10 years ago, I made the decision to switch into AI. People thought I was 675 00:43:28,870 --> 00:43:32,112 crazy, right? Yes, yes, I 676 00:43:32,465 --> 00:43:36,060 completely, you know, I was, when, when 677 00:43:36,493 --> 00:43:39,944 this AI machine learning hype 678 00:43:39,944 --> 00:43:43,105 happened, a lot of people, 679 00:43:43,955 --> 00:43:47,210 they jump to AI machine learning. 680 00:43:48,028 --> 00:43:51,540 And I remember the time, and one of my friends said 681 00:43:51,572 --> 00:43:55,276 that, you know, from 682 00:43:55,341 --> 00:43:58,997 my own country, you know that everything has been changed. You 683 00:43:59,013 --> 00:44:02,845 can find a lot of jobs here because when we started 684 00:44:02,878 --> 00:44:06,534 together, it was very little places. And I said that, yes, I can 685 00:44:06,582 --> 00:44:08,891 see everywhere in the world right now. 686 00:44:10,334 --> 00:44:14,021 Market's demand for that. But at that time, it was like that. 687 00:44:14,742 --> 00:44:15,992 It's like a kind of 688 00:44:18,525 --> 00:44:22,132 something that you cannot use. It is just for research, 689 00:44:22,276 --> 00:44:26,043 for libraries, but very, very 690 00:44:26,043 --> 00:44:29,875 niche, very niche type of situations. Airline seat 691 00:44:29,875 --> 00:44:33,097 optimization or logistic optimization wasn't 692 00:44:33,097 --> 00:44:36,848 really mainstream. It was just starting to become— well, 693 00:44:36,864 --> 00:44:40,631 20 years ago, it was a different world. But I mean, 10 years 694 00:44:40,631 --> 00:44:44,228 ago, I think people started realizing like, hey, we have all this data and we 695 00:44:44,228 --> 00:44:48,050 can start doing something. Now, the hype is very real 696 00:44:48,163 --> 00:44:51,712 and everybody's an AI expert now. But I think 697 00:44:51,712 --> 00:44:55,535 quantum is very much the same thing. You can validate this by looking for jobs. 698 00:44:55,984 --> 00:44:59,807 There's not a lot of jobs, but there is a lot 699 00:44:59,807 --> 00:45:03,163 more this year than there was last year. I suspect that there's going to be 700 00:45:03,195 --> 00:45:07,034 more next year than this year. I think people forget, 701 00:45:07,114 --> 00:45:10,936 people always look at the hockey stick graphic or an exponential curve, Exponential 702 00:45:10,968 --> 00:45:13,760 curves during the first iteration or two 703 00:45:14,675 --> 00:45:18,333 don't really grow much, and then suddenly they explode. Yes, 704 00:45:18,510 --> 00:45:22,281 I totally agree. And the other aspect is that 705 00:45:22,939 --> 00:45:25,714 the success of AI is 706 00:45:27,752 --> 00:45:31,507 because of the growth of GPUs. So it was like 707 00:45:31,523 --> 00:45:35,053 both Internet of Things, GPUs, the 708 00:45:35,053 --> 00:45:38,872 computational power, and the data. When you have the Internet of Things, 709 00:45:38,985 --> 00:45:42,564 we could sense the world and can 710 00:45:43,061 --> 00:45:46,786 capture a lot of information. Social media, all of those things 711 00:45:47,252 --> 00:45:50,543 create the opportunity for AI and machine learning. Right now 712 00:45:51,089 --> 00:45:54,910 we have the same scenario. We 713 00:45:55,103 --> 00:45:58,860 are advancing in nanotechnology. We are advancing 714 00:45:59,117 --> 00:46:02,906 in error correction for the 715 00:46:03,291 --> 00:46:05,892 quantum computing. So I think that you're 716 00:46:07,000 --> 00:46:10,307 very Correct. And I think it's not linear 717 00:46:10,997 --> 00:46:14,785 like AI. There will be soon a 718 00:46:14,785 --> 00:46:17,402 huge jump to having very 719 00:46:17,996 --> 00:46:21,575 capable quantum machines and quantum 720 00:46:21,591 --> 00:46:25,203 computers that can help us to change the worldview 721 00:46:25,363 --> 00:46:28,349 even much more dramatically. 722 00:46:30,275 --> 00:46:34,112 Absolutely. So where can— I'm sorry, go ahead, Gannis. No, I 723 00:46:34,112 --> 00:46:37,544 was going to ask, you know, If you could give a piece of advice to 724 00:46:37,560 --> 00:46:40,575 leaders who feel that they're already late to quantum, 725 00:46:41,665 --> 00:46:45,418 what would you tell them right now? Again, 726 00:46:45,434 --> 00:46:49,266 it goes back to if the area 727 00:46:49,266 --> 00:46:52,249 that they work requires quantum, even to consider that 728 00:46:53,612 --> 00:46:55,824 we entered the area of 729 00:46:57,284 --> 00:47:00,972 quantum, full error tolerance 730 00:47:01,148 --> 00:47:04,580 quantum computers. If it makes sense, if 731 00:47:05,141 --> 00:47:08,781 quantum provides a supermassive, if it 732 00:47:09,150 --> 00:47:12,774 provides a kind of solution that benefits the 733 00:47:13,095 --> 00:47:15,837 company, they need to 734 00:47:16,542 --> 00:47:20,279 start looking for the option. So first of all, they need 735 00:47:20,311 --> 00:47:23,839 to understand the problem. If the problem has the nature to 736 00:47:23,839 --> 00:47:26,292 move on to this kind of 737 00:47:26,885 --> 00:47:29,627 computational sector, I think 738 00:47:31,538 --> 00:47:34,957 First of all, investing even in small amounts. Every 739 00:47:35,070 --> 00:47:38,858 company, if they want to grow, they have a 740 00:47:38,923 --> 00:47:42,150 budget for R&D. So they can 741 00:47:42,294 --> 00:47:45,858 invest in the future, for the future in this 742 00:47:45,970 --> 00:47:49,454 field. If their problem can be 743 00:47:50,177 --> 00:47:53,853 solved by quantum computers, again, it's 744 00:47:53,853 --> 00:47:56,951 like understanding of the nature of the quantum and 745 00:47:57,128 --> 00:48:00,915 understanding the capacity and 746 00:48:00,915 --> 00:48:04,251 capability of these powerful machines. 747 00:48:04,684 --> 00:48:08,147 It's not like that everybody should jump to the, like, 748 00:48:08,356 --> 00:48:11,964 AI. Maybe sometime in future comes, but at the moment, 749 00:48:12,221 --> 00:48:15,316 consider that quantum computers, 750 00:48:15,733 --> 00:48:18,363 although they promise to save energy, 751 00:48:19,678 --> 00:48:22,676 but it's, if for a small 752 00:48:22,949 --> 00:48:26,702 problem in compared to the classical machine, 753 00:48:26,990 --> 00:48:29,973 it uses much more energy. So the problem should be 754 00:48:30,662 --> 00:48:33,195 huge enough so that you can 755 00:48:34,429 --> 00:48:37,859 benefit from the computational power of the quantum 756 00:48:37,859 --> 00:48:41,513 mechanics and quantum computers, sorry. But generally, I 757 00:48:41,642 --> 00:48:45,392 think that investment doesn't 758 00:48:45,392 --> 00:48:49,095 harm in terms of R&D, research and development, 759 00:48:50,201 --> 00:48:53,118 and also learning the 760 00:48:55,186 --> 00:48:57,798 quantum mechanics and quantum computers 761 00:48:59,258 --> 00:49:02,806 can be a little bit challenging at the 762 00:49:02,871 --> 00:49:06,579 beginning, but it is like, as you say, it's 763 00:49:06,579 --> 00:49:10,224 like the curve is like at the beginning is very slow, but when 764 00:49:10,417 --> 00:49:13,772 you reach a certain point, it's very easy. It becomes 765 00:49:13,820 --> 00:49:17,513 natural to you. That's 766 00:49:17,561 --> 00:49:21,270 fantastic. That's good. So I think I cut off Frank, who was going to ask 767 00:49:21,270 --> 00:49:24,481 you, so if people want to find out more, 768 00:49:24,947 --> 00:49:28,546 about what you do, to follow and ask questions, 769 00:49:28,659 --> 00:49:30,330 where would we send them to? 770 00:49:32,659 --> 00:49:35,166 Both my company's 771 00:49:36,114 --> 00:49:39,456 website and also my LinkedIn. I can— 772 00:49:40,773 --> 00:49:44,517 daily I receive a lot of messages and I answer some. 773 00:49:44,629 --> 00:49:48,035 Sometimes you get some interesting ideas, you get some 774 00:49:48,903 --> 00:49:51,827 very passionate collaborators. Both my 775 00:49:52,340 --> 00:49:55,857 heliumai.co.nz and 776 00:49:55,953 --> 00:49:59,550 also my LinkedIn is 777 00:49:59,646 --> 00:50:02,809 a good channel to connect and communicate. 778 00:50:04,142 --> 00:50:06,406 Awesome. Fantastic. Fantastic. 779 00:50:07,835 --> 00:50:10,324 And with that, we'll play the outro music. 780 00:50:16,490 --> 00:50:19,877 They're breaking the mold. Science and sky pizza, 781 00:50:19,974 --> 00:50:20,889 bold and it's 782 00:50:20,969 --> 00:50:29,752 gold. 783 00:50:33,220 --> 00:50:36,994 The multiverse is skanking, skanking in time. Black holes are 784 00:50:36,994 --> 00:50:40,366 wailing in a horn line so fine. From plank scales to planets, they're 785 00:50:40,366 --> 00:50:44,016 connecting the dots. Kenneth and Frank, they're the cosmic 786 00:50:44,080 --> 00:50:48,161 hotshot. 787 00:50:49,623 --> 00:50:53,456 Quantum Podcast, turn it up fast. Kenneth and Frank blowing 788 00:50:53,585 --> 00:50:57,231 my mind at last. Quantum Podcast, they're breaking the 789 00:50:57,344 --> 00:51:01,071 mold. Science has got beats, it's bold 790 00:51:01,136 --> 00:51:01,875 and it's gold.