1 00:00:02,520 --> 00:00:11,879 Speaker 1: Bloomberg Audio Studios, podcasts, radio news. This is Masters in 2 00:00:11,960 --> 00:00:16,119 Speaker 1: Business with Barry Ritholts on Bloomberg Radio on. 3 00:00:16,200 --> 00:00:20,200 Speaker 2: The latest Masters in Business podcast. My conversation with Song 4 00:00:20,280 --> 00:00:25,840 Speaker 2: Yi Yun. She is founder and managing partner at Principal Ventures, 5 00:00:26,120 --> 00:00:31,440 Speaker 2: an AI focused venture capital investment firm. She has a fascinating, 6 00:00:31,520 --> 00:00:38,199 Speaker 2: fascinating background all sorts of MIT Corporation Advisory Board, fifty 7 00:00:38,240 --> 00:00:41,560 Speaker 2: Women to Watch in Business from the Wall Street Journal, 8 00:00:42,400 --> 00:00:45,920 Speaker 2: named to the advisory board for the Center for Asia 9 00:00:45,960 --> 00:00:50,240 Speaker 2: Pacific Policy as well as the National Academy of Engineering 10 00:00:50,280 --> 00:00:54,840 Speaker 2: of Korea. She has a fascinating background in gaming, telecom 11 00:00:55,240 --> 00:00:58,840 Speaker 2: and AI. I found this conversation to be fascinating, and 12 00:00:58,920 --> 00:01:01,760 Speaker 2: I think you will also with no further ado, my 13 00:01:01,960 --> 00:01:07,280 Speaker 2: discussion with Principal Ventures. Sang Yi you that is quite 14 00:01:07,400 --> 00:01:11,560 Speaker 2: a CV. I went through. Let's roll back though to 15 00:01:11,600 --> 00:01:15,000 Speaker 2: where it all began. You get a bachelor's in science 16 00:01:15,080 --> 00:01:19,560 Speaker 2: from Korea's Advanced Institute of Science and Technology and then 17 00:01:19,600 --> 00:01:26,080 Speaker 2: a PhD in computational neuroscience from MIT. That's such a 18 00:01:26,120 --> 00:01:29,640 Speaker 2: fascinating area. What was the original career plan? 19 00:01:31,800 --> 00:01:36,680 Speaker 3: That's a very good question. I mean, I think growing 20 00:01:36,760 --> 00:01:38,600 Speaker 3: up in South Korea. I didn't know what are the 21 00:01:38,680 --> 00:01:42,480 Speaker 3: career options that I had. I just really enjoyed learning 22 00:01:42,600 --> 00:01:49,520 Speaker 3: science and engineering subject So when I was young, I 23 00:01:49,600 --> 00:01:54,520 Speaker 3: realized for some people, like singing is very natural, huh, 24 00:01:54,600 --> 00:01:57,960 Speaker 3: some people dancing as natural. I cannot sing, I cannot dance, 25 00:01:58,520 --> 00:02:02,240 Speaker 3: But like speaking to computers and programming was very natural 26 00:02:02,280 --> 00:02:04,560 Speaker 3: to me. So I started programming when I was nine, 27 00:02:05,200 --> 00:02:09,360 Speaker 3: and that led me to major in electrical engineering as 28 00:02:09,400 --> 00:02:14,839 Speaker 3: an undergrad at CHIST. And along the way, I realized 29 00:02:14,919 --> 00:02:16,960 Speaker 3: I just wanted to become a better engineer because I 30 00:02:17,040 --> 00:02:20,600 Speaker 3: was study as an engineering student. To be a better engineer, 31 00:02:20,680 --> 00:02:23,959 Speaker 3: you need to understand how human brain works. So for example, 32 00:02:24,480 --> 00:02:28,960 Speaker 3: I was studying signal processing algorithm, and those signal processing 33 00:02:29,000 --> 00:02:33,680 Speaker 3: algorithms are look best to your eyes when it's not 34 00:02:34,080 --> 00:02:39,399 Speaker 3: necessarily mathematically the best, but take into consideration what frequencies 35 00:02:39,520 --> 00:02:44,480 Speaker 3: most sensitive to human eyes. So understanding human brain and 36 00:02:44,520 --> 00:02:48,040 Speaker 3: human perception will enable you become a better engineer. So 37 00:02:48,040 --> 00:02:51,680 Speaker 3: I was kind of exploring what would be the subject 38 00:02:51,960 --> 00:02:54,480 Speaker 3: or kind of the major that I can pursue to 39 00:02:54,600 --> 00:02:59,280 Speaker 3: have a better good understanding both engineering and human brain. 40 00:02:59,520 --> 00:03:03,000 Speaker 3: And that was that led me to study competition and 41 00:03:03,040 --> 00:03:04,000 Speaker 3: re science that I might say. 42 00:03:03,919 --> 00:03:07,920 Speaker 2: So computational neuroscience isn't so much about using computers to 43 00:03:08,080 --> 00:03:12,919 Speaker 2: understand people as opposed to understanding as opposed to understanding 44 00:03:12,960 --> 00:03:18,119 Speaker 2: neuroscience to create better software, better interfaces, better human interaction 45 00:03:18,240 --> 00:03:19,160 Speaker 2: with technology. 46 00:03:19,240 --> 00:03:21,239 Speaker 3: Is that rare? Exactly? Yeah, that's right. 47 00:03:21,680 --> 00:03:24,920 Speaker 2: Huh so pretty fascinating. Early in your career, you're at 48 00:03:25,000 --> 00:03:27,400 Speaker 2: McKenzie for a few years and then you eventually move 49 00:03:27,480 --> 00:03:32,160 Speaker 2: into SK telecom. Tell us your focus at both places. 50 00:03:32,680 --> 00:03:35,640 Speaker 3: Yeah, So, I mean, I think after after my PhD, 51 00:03:35,720 --> 00:03:39,880 Speaker 3: I wanted to go into the instead of staying in academia, 52 00:03:40,200 --> 00:03:43,400 Speaker 3: I wanted to go into the business world, and going 53 00:03:43,400 --> 00:03:45,600 Speaker 3: to McKinsey was the kind of the best way to 54 00:03:45,680 --> 00:03:52,440 Speaker 3: transition from being a being a PhD student to go 55 00:03:52,560 --> 00:03:56,839 Speaker 3: into the real world. So it was a really fascinating 56 00:03:57,400 --> 00:04:01,839 Speaker 3: experience for me, very fast paced, able to work with 57 00:04:02,200 --> 00:04:05,600 Speaker 3: big conglomerates and the leaders of the businesses and in 58 00:04:05,640 --> 00:04:10,200 Speaker 3: the areas of strategy and corporate finance, et cetera. And 59 00:04:11,200 --> 00:04:15,960 Speaker 3: s K was one of the forum's clients. And I 60 00:04:15,960 --> 00:04:18,119 Speaker 3: don't want to like date myself. It was a time 61 00:04:18,200 --> 00:04:22,800 Speaker 3: that everyone was rushing into three G rollout if you remember, 62 00:04:24,080 --> 00:04:26,800 Speaker 3: it was an interesting transition, just like we see today 63 00:04:26,880 --> 00:04:32,480 Speaker 3: because in two G della communication is all about voice communication, 64 00:04:32,800 --> 00:04:36,360 Speaker 3: and three G what was promised was the data transmission 65 00:04:36,880 --> 00:04:41,440 Speaker 3: including videos and images and high fidelity audios. 66 00:04:41,640 --> 00:04:45,440 Speaker 2: If I'm remembering correctly, it was voice and text, and 67 00:04:45,480 --> 00:04:49,440 Speaker 2: then it was image and some video and eventually what 68 00:04:49,520 --> 00:04:52,560 Speaker 2: was it for G five G was full Internet, right right. 69 00:04:52,480 --> 00:04:58,240 Speaker 3: Yeah, that's right. So as Telcos are one of the 70 00:04:58,360 --> 00:05:04,320 Speaker 3: big CAPEX investors in making that transition, and we're thinking 71 00:05:04,360 --> 00:05:07,240 Speaker 3: about how we can do the content delivery in the 72 00:05:07,320 --> 00:05:11,839 Speaker 3: most personal personalized content delivery was one of the challenges 73 00:05:12,520 --> 00:05:17,840 Speaker 3: that requires artificial intelligence and like a data data driven 74 00:05:18,279 --> 00:05:21,359 Speaker 3: delivery system. So I thought that was an interesting challenge 75 00:05:21,440 --> 00:05:25,039 Speaker 3: to take on. So I moved to SKA Telecom to 76 00:05:25,120 --> 00:05:26,520 Speaker 3: lead that effort, and. 77 00:05:26,480 --> 00:05:30,640 Speaker 2: Then you end up at NCSoft where you're president and 78 00:05:30,760 --> 00:05:35,479 Speaker 2: chief strategy officer. I'm curious what those experiences taught you, 79 00:05:36,040 --> 00:05:42,160 Speaker 2: not just about corporate governance and culture, but just these 80 00:05:42,160 --> 00:05:46,359 Speaker 2: big institutions that tend to have legacy technology. You know, 81 00:05:46,520 --> 00:05:49,200 Speaker 2: there tends to be some group that really wants to 82 00:05:49,240 --> 00:05:52,640 Speaker 2: move forward rapidly and adapt all the latest greatest tech 83 00:05:53,240 --> 00:05:56,159 Speaker 2: and then another group that hey, this is expensive, what's 84 00:05:56,240 --> 00:05:59,599 Speaker 2: the what's the ROI in this? How did you find 85 00:05:59,600 --> 00:06:05,240 Speaker 2: yourself navigating a big telecom like sk or a small 86 00:06:05,320 --> 00:06:08,279 Speaker 2: or more nimble gaming company like NCSoft. 87 00:06:09,160 --> 00:06:11,120 Speaker 3: Yeah, I mean that's a really great question, and I 88 00:06:11,200 --> 00:06:16,320 Speaker 3: think it's learned to be persistent and resilient and patient 89 00:06:17,040 --> 00:06:22,680 Speaker 3: in both places. I was criticized for, like ask suggesting 90 00:06:22,760 --> 00:06:25,840 Speaker 3: something that was not the norm at the time. So, 91 00:06:25,880 --> 00:06:28,520 Speaker 3: for example, when I was an NCSoft, one of the 92 00:06:28,560 --> 00:06:31,000 Speaker 3: things that I it was very obvious to me was 93 00:06:31,080 --> 00:06:34,880 Speaker 3: it was full of data in gaming, the businesses and 94 00:06:34,920 --> 00:06:39,080 Speaker 3: like all offered in a distised form. You have transaction data, 95 00:06:39,120 --> 00:06:41,480 Speaker 3: you have a behavior data of the gamers and everything. 96 00:06:41,800 --> 00:06:43,760 Speaker 3: So it was possible to do a lot of things 97 00:06:44,160 --> 00:06:47,520 Speaker 3: on the data driven way, which is very kind of 98 00:06:48,000 --> 00:06:51,800 Speaker 3: not today it's it's it's a lot of the companies 99 00:06:51,800 --> 00:06:54,960 Speaker 3: are doing it, But back then it was it was 100 00:06:55,000 --> 00:06:58,800 Speaker 3: not very common to have understanding in both gaming business 101 00:06:59,000 --> 00:07:02,240 Speaker 3: and as well as kind of AI and kind of 102 00:07:02,320 --> 00:07:07,440 Speaker 3: data driven business business process modeling. So when I suggested 103 00:07:07,920 --> 00:07:11,120 Speaker 3: things like, oh, like it's let's do a small thing 104 00:07:11,240 --> 00:07:13,920 Speaker 3: like like charm prediction, because you can you can see 105 00:07:13,920 --> 00:07:17,840 Speaker 3: the customer player behavior within the game and see how 106 00:07:17,960 --> 00:07:20,840 Speaker 3: much they're engaged, and you can predict if that player 107 00:07:20,920 --> 00:07:23,120 Speaker 3: is about to turn out or continue with a with 108 00:07:23,200 --> 00:07:27,200 Speaker 3: a game, and some kind of interventions could help them 109 00:07:27,480 --> 00:07:29,600 Speaker 3: stay engaged with a game. So that was the one 110 00:07:30,400 --> 00:07:33,760 Speaker 3: application areas that I didn't identify it which could be 111 00:07:33,840 --> 00:07:36,520 Speaker 3: very straightforward, but then I was told, like there is 112 00:07:36,560 --> 00:07:40,440 Speaker 3: a strong pushback from the developers and even the business people. 113 00:07:40,800 --> 00:07:42,800 Speaker 3: They said, oh, you're saying it because you don't understand 114 00:07:42,800 --> 00:07:45,800 Speaker 3: a gaming business. You're you're you're saying you're a kind 115 00:07:45,840 --> 00:07:48,440 Speaker 3: of a you're not a heavy gamer in or whatever, but. 116 00:07:48,800 --> 00:07:52,600 Speaker 2: You understand, hey, because this as much to acquire a 117 00:07:52,680 --> 00:07:56,160 Speaker 2: client or a gamer, and if we see this behavior, 118 00:07:56,280 --> 00:07:59,240 Speaker 2: a five percentage of those folks are tapping out, what 119 00:07:59,240 --> 00:08:01,440 Speaker 2: can we do to keep in and paying monthly fees? 120 00:08:01,600 --> 00:08:04,320 Speaker 3: Right right? Exactly. Yeah, So even with a very clear 121 00:08:04,440 --> 00:08:09,120 Speaker 3: data and the case presented, it was not an easy 122 00:08:10,240 --> 00:08:13,920 Speaker 3: task to have get everyone's buy in. But I think 123 00:08:13,920 --> 00:08:17,080 Speaker 3: it gradually. Just the reason I kind of mentioned that 124 00:08:17,400 --> 00:08:20,440 Speaker 3: tangible example was a small, very tangible area that we 125 00:08:20,480 --> 00:08:23,240 Speaker 3: can apply to technology and once you show the success, 126 00:08:23,280 --> 00:08:25,600 Speaker 3: I think there is a gradually one by one we're 127 00:08:25,640 --> 00:08:30,200 Speaker 3: able to adopt integrate that into our business process to 128 00:08:30,320 --> 00:08:33,560 Speaker 3: end up having a large AI lab that does like 129 00:08:34,320 --> 00:08:37,079 Speaker 3: all of those things in a more centralized way. 130 00:08:37,160 --> 00:08:40,720 Speaker 2: So what I'm hearing from you is very systems oriented framework, 131 00:08:40,800 --> 00:08:45,040 Speaker 2: both for gaming and telecom. I know the big mobile 132 00:08:45,080 --> 00:08:49,000 Speaker 2: companies in the US are constantly fighting their own churn rate. 133 00:08:50,280 --> 00:08:53,760 Speaker 2: So having a top down systems approach sounds like you 134 00:08:53,800 --> 00:09:00,120 Speaker 2: could be really proactive in terms of maintaining clients. You 135 00:09:00,160 --> 00:09:02,920 Speaker 2: would think there's buy in from everybody, but it sounds 136 00:09:02,960 --> 00:09:08,319 Speaker 2: like there's a little salesmanship involved to get everybody behind 137 00:09:08,400 --> 00:09:10,080 Speaker 2: that that approach. 138 00:09:09,840 --> 00:09:11,840 Speaker 3: Right, yeah, right, yeah. 139 00:09:11,559 --> 00:09:16,120 Speaker 2: So let's talk a little bit about what's going on 140 00:09:16,200 --> 00:09:20,640 Speaker 2: in the world of AI. I've heard you discuss various 141 00:09:20,679 --> 00:09:23,400 Speaker 2: things that are just short term hype. How do you 142 00:09:23,559 --> 00:09:29,080 Speaker 2: figure out when you're evaluating an AI system, either for 143 00:09:29,120 --> 00:09:33,160 Speaker 2: an investment or just to use the technology in a company, 144 00:09:33,520 --> 00:09:37,000 Speaker 2: how do you figure out what's valuable and what's just hype? 145 00:09:37,080 --> 00:09:38,280 Speaker 2: What are you looking at? 146 00:09:39,080 --> 00:09:41,840 Speaker 3: I mean, I think we talk a lot about hype 147 00:09:41,840 --> 00:09:45,040 Speaker 3: cycle and kind of bubble being built up in this 148 00:09:45,840 --> 00:09:47,840 Speaker 3: in this kind of AI era, but I think it's 149 00:09:47,880 --> 00:09:54,040 Speaker 3: not unheard of them in every platform shift, there was 150 00:09:54,600 --> 00:09:59,600 Speaker 3: overcapacity built, not just in AI infrastructure, but it happened 151 00:09:59,600 --> 00:10:01,840 Speaker 3: with inter like fiber optics. 152 00:10:02,360 --> 00:10:04,080 Speaker 2: Railroad, pelgram, wherever you go. 153 00:10:04,320 --> 00:10:08,040 Speaker 3: So there's always a kind of excess capacity that's built, 154 00:10:08,080 --> 00:10:11,079 Speaker 3: and I think that's when people what people are pointing 155 00:10:11,120 --> 00:10:13,880 Speaker 3: to is that is that area that there is some 156 00:10:13,960 --> 00:10:17,440 Speaker 3: kind of bubble built up. But on the other hand, 157 00:10:17,480 --> 00:10:20,040 Speaker 3: if you talk about application of the technology, if you 158 00:10:20,200 --> 00:10:24,400 Speaker 3: find the application and real business problems that you can 159 00:10:25,240 --> 00:10:29,480 Speaker 3: apply this technology to solve, to be more efficient or 160 00:10:30,320 --> 00:10:33,640 Speaker 3: bring out some insights that humans were not able to 161 00:10:33,679 --> 00:10:35,520 Speaker 3: do it, And I think there is a great area 162 00:10:35,600 --> 00:10:38,360 Speaker 3: to apply the technology, and I think there are so 163 00:10:38,600 --> 00:10:41,480 Speaker 3: many of them out there. So that's why we are 164 00:10:41,600 --> 00:10:46,679 Speaker 3: so excited about the development technology and the prospect of 165 00:10:46,760 --> 00:10:47,560 Speaker 3: it going forward. 166 00:10:47,960 --> 00:10:51,880 Speaker 2: So I've heard you discuss various priorities, whether it's a 167 00:10:51,920 --> 00:10:59,360 Speaker 2: corporate entity pursuing upgrading their AI infrastructure or you know, 168 00:10:59,640 --> 00:11:05,079 Speaker 2: just tot AI native startup. You talk about durability, defensibility, 169 00:11:05,280 --> 00:11:09,160 Speaker 2: real world impact. Explain what those three things mean. 170 00:11:10,559 --> 00:11:17,120 Speaker 3: So in making in kind of making that adoption of 171 00:11:17,160 --> 00:11:19,760 Speaker 3: the technology, there are there are two ways to think 172 00:11:19,800 --> 00:11:25,320 Speaker 3: about it. One is adopting the technology without really changing 173 00:11:25,360 --> 00:11:29,120 Speaker 3: the current work process, right, So you can, for example, 174 00:11:29,160 --> 00:11:31,880 Speaker 3: there are a lot of talk about co pilot or 175 00:11:31,880 --> 00:11:35,120 Speaker 3: like augmenting what we do make it faster like a search, 176 00:11:35,600 --> 00:11:40,280 Speaker 3: and if those things are necessarily changing our current established workflow, 177 00:11:40,640 --> 00:11:43,760 Speaker 3: but you are fully embracing what the technology can do 178 00:11:43,920 --> 00:11:46,719 Speaker 3: to make you more productive and efficient. So that's the 179 00:11:46,840 --> 00:11:49,040 Speaker 3: kind of one way of applying it, and I think 180 00:11:49,120 --> 00:11:52,800 Speaker 3: there will be some amount of the ry that will 181 00:11:52,840 --> 00:11:56,840 Speaker 3: be realized from such approaches. And the other is complete 182 00:11:56,880 --> 00:11:59,680 Speaker 3: redesign of the workflow. And I think that's a that's 183 00:11:59,679 --> 00:12:03,400 Speaker 3: a that's kind of we're a very early stage of 184 00:12:03,480 --> 00:12:06,400 Speaker 3: like kind of witnessing that, but I think that will 185 00:12:06,440 --> 00:12:10,400 Speaker 3: be a more interesting area to look out for and 186 00:12:10,720 --> 00:12:14,679 Speaker 3: could be more tremendous transformation and the value created from 187 00:12:14,720 --> 00:12:15,280 Speaker 3: such effort. 188 00:12:15,520 --> 00:12:18,720 Speaker 2: So tell us what you did at NCSoft, because a 189 00:12:18,760 --> 00:12:21,600 Speaker 2: lot of the work you put in there was about 190 00:12:21,640 --> 00:12:25,400 Speaker 2: transforming them to use AI. Was it, Hey, we're just 191 00:12:25,480 --> 00:12:29,160 Speaker 2: going to make all our developers and gamers and everybody 192 00:12:29,160 --> 00:12:31,880 Speaker 2: else a little more efficient, a little more productive or 193 00:12:31,920 --> 00:12:35,880 Speaker 2: did this require a clean sheet rethink of everything that 194 00:12:36,280 --> 00:12:37,200 Speaker 2: the company was doing. 195 00:12:37,400 --> 00:12:40,920 Speaker 3: Yeah, I mean I think it was like fifteen years ago, 196 00:12:41,040 --> 00:12:43,560 Speaker 3: and I think it's back then, I think technology was 197 00:12:43,640 --> 00:12:48,480 Speaker 3: not ready to fully redesign the game development work workflow. 198 00:12:48,520 --> 00:12:50,640 Speaker 3: But I think back then it was more of augmenting 199 00:12:50,640 --> 00:12:54,800 Speaker 3: the existing process. So for example, like we did things 200 00:12:54,840 --> 00:12:58,640 Speaker 3: like the marketing and Trump prediction and more efficiently we 201 00:12:58,760 --> 00:13:03,480 Speaker 3: had NLP that's design signed for Gamer, specialized for Gamer language. 202 00:13:03,720 --> 00:13:09,559 Speaker 3: Like we had augmenting, the animation tool that helps animators 203 00:13:09,600 --> 00:13:13,240 Speaker 3: to animate not just by pedal creatures, but like four 204 00:13:13,320 --> 00:13:17,480 Speaker 3: legged monsters in a way that's as efficient as you're 205 00:13:17,520 --> 00:13:20,240 Speaker 3: doing bipedal creatures, et cetera. So I think it's a 206 00:13:20,320 --> 00:13:23,760 Speaker 3: more focused on augmenting existing processes back then. But I 207 00:13:23,760 --> 00:13:28,800 Speaker 3: think the technology has advanced and matured today that I 208 00:13:28,800 --> 00:13:32,600 Speaker 3: think that there are more opportunities to kind of completely 209 00:13:33,160 --> 00:13:35,600 Speaker 3: redesign and come up with a new type of AI 210 00:13:35,720 --> 00:13:39,520 Speaker 3: native companies. So I see like an AI native entertainment 211 00:13:39,559 --> 00:13:41,920 Speaker 3: firms that we're thinking about what do you mean by 212 00:13:42,000 --> 00:13:45,800 Speaker 3: new type of entertainment? New type of engagement that's different 213 00:13:45,880 --> 00:13:51,079 Speaker 3: from polishing and like making higher fidelity graphics and games, 214 00:13:51,400 --> 00:13:54,760 Speaker 3: but like a completely new form of entertainment using fully 215 00:13:54,800 --> 00:13:56,960 Speaker 3: embracing the technology that we have today. 216 00:13:56,800 --> 00:14:00,200 Speaker 2: So I keep reading that Claude is writing its own 217 00:14:00,280 --> 00:14:04,160 Speaker 2: code and updating its own code. If you were at 218 00:14:04,960 --> 00:14:10,960 Speaker 2: a gaming shop today, I'm assuming, do you replace coders? 219 00:14:11,040 --> 00:14:14,199 Speaker 2: Do you have copilot work with coders? It was a 220 00:14:14,240 --> 00:14:16,920 Speaker 2: Wall Street Journal article last week. I think it was 221 00:14:16,920 --> 00:14:20,480 Speaker 2: Wall Street Journal. It could have been wired about coders 222 00:14:20,720 --> 00:14:23,720 Speaker 2: in Silicon Valley are just kind of sitting around watching 223 00:14:24,240 --> 00:14:28,000 Speaker 2: Claude rewrite their code. What is going on in the 224 00:14:28,000 --> 00:14:32,920 Speaker 2: world of software development now that Claude is capable of 225 00:14:33,000 --> 00:14:33,920 Speaker 2: updating itself. 226 00:14:34,240 --> 00:14:36,520 Speaker 3: Yeah, I think it's a really fascinating I think a 227 00:14:36,560 --> 00:14:40,680 Speaker 3: lot of the coding is done using tools like a Claude, 228 00:14:41,360 --> 00:14:46,480 Speaker 3: and it certainly makes it more efficient and like productive, 229 00:14:46,640 --> 00:14:49,440 Speaker 3: which means that we need a lot less people in 230 00:14:49,480 --> 00:14:53,000 Speaker 3: the loop to do the job in certain areas such 231 00:14:53,040 --> 00:14:57,560 Speaker 3: as like reviewing a kind of like detecting errors and 232 00:14:57,680 --> 00:15:00,000 Speaker 3: just kind of I think there are a certain area 233 00:15:00,000 --> 00:15:03,080 Speaker 3: area is that AI coders can do better, But there 234 00:15:03,080 --> 00:15:06,960 Speaker 3: are other areas that these more kind of heavy hand, 235 00:15:07,480 --> 00:15:12,400 Speaker 3: like more involvement of a kind of redesigning that the 236 00:15:12,520 --> 00:15:15,200 Speaker 3: schema and the structure and how things are going to 237 00:15:15,280 --> 00:15:18,320 Speaker 3: work and how it kind of is going to be 238 00:15:19,840 --> 00:15:26,120 Speaker 3: implemented in providing them kind of engaging experience as for gamers. 239 00:15:26,320 --> 00:15:30,480 Speaker 2: So my bias is that the humans are very creative 240 00:15:30,600 --> 00:15:35,720 Speaker 2: and very innovative, and you know, I'm thinking in the 241 00:15:35,760 --> 00:15:40,440 Speaker 2: sort of storylines we see on all the streaming shows 242 00:15:40,480 --> 00:15:45,760 Speaker 2: and some of the interesting novel gaming. I don't know 243 00:15:45,800 --> 00:15:47,880 Speaker 2: what to call it. I guess it's a narrative storyline. 244 00:15:48,840 --> 00:15:50,920 Speaker 2: Is that what people going to focus on, and just 245 00:15:51,000 --> 00:15:54,720 Speaker 2: the blocking and tackling, just the daily grind of putting 246 00:15:54,720 --> 00:15:58,200 Speaker 2: code into place we're going to let ai use. Is 247 00:15:58,240 --> 00:16:00,600 Speaker 2: that a today thing or is that a next you know, 248 00:16:00,680 --> 00:16:02,680 Speaker 2: is that going to change over the next couple of 249 00:16:02,720 --> 00:16:03,960 Speaker 2: decades or years. 250 00:16:05,520 --> 00:16:07,560 Speaker 3: I think that's a really good question. I think it's 251 00:16:07,960 --> 00:16:10,720 Speaker 3: if you look at today, I mean, I think a 252 00:16:10,800 --> 00:16:14,320 Speaker 3: lot of the jobs, like like YouTubers, podcast is are 253 00:16:14,440 --> 00:16:16,920 Speaker 3: the type of job that didn't exist ten years ago. 254 00:16:17,320 --> 00:16:19,640 Speaker 3: I don't know what other jobs are going to be 255 00:16:19,680 --> 00:16:23,040 Speaker 3: created in the world where like the things that needed 256 00:16:23,200 --> 00:16:26,080 Speaker 3: needed like one hundred people attention can be done with 257 00:16:26,240 --> 00:16:29,080 Speaker 3: like a fraction of those people. There could be other 258 00:16:29,160 --> 00:16:31,000 Speaker 3: type of jobs there are the type of roles, But 259 00:16:31,080 --> 00:16:33,680 Speaker 3: I think, uh, that's a that's an evolution that will 260 00:16:33,720 --> 00:16:36,680 Speaker 3: have to see how it how it roll out. Then 261 00:16:37,000 --> 00:16:40,440 Speaker 3: then I can and kind of predicting exactly what type 262 00:16:40,440 --> 00:16:42,560 Speaker 3: of jobs we're going to exist in ten years from now. 263 00:16:42,720 --> 00:16:46,239 Speaker 2: Huh, really really interesting. Coming up, we continue our conversation 264 00:16:46,400 --> 00:16:51,000 Speaker 2: with Song Ye Yun uh, managing partner at Principal Ventures, 265 00:16:51,400 --> 00:16:56,480 Speaker 2: discussing AI and the modern economy. I'm Barry Ridults. You're 266 00:16:56,520 --> 00:17:11,000 Speaker 2: listening to Masters in Business on Bloomberg Radio. I'm Barry Ridalts. 267 00:17:11,040 --> 00:17:13,960 Speaker 2: You're listening to Masters in Business on Bloomberg Radio. My 268 00:17:14,119 --> 00:17:17,359 Speaker 2: extra special guest today is Sami Yun. She is the 269 00:17:17,560 --> 00:17:22,040 Speaker 2: founder and managing partner at Principal Venture Partners, an AI 270 00:17:22,240 --> 00:17:26,080 Speaker 2: focused venture capital firm. Previously, she was president and Chief 271 00:17:26,080 --> 00:17:31,840 Speaker 2: strategy officer at gaming company NCSoft. So before we start 272 00:17:31,880 --> 00:17:36,040 Speaker 2: talking about AI in more depth, I just have to 273 00:17:36,119 --> 00:17:40,639 Speaker 2: mention your book Push Play, Gaming for a Better World. 274 00:17:41,160 --> 00:17:45,200 Speaker 2: I love the concept that, hey, let's not forget about play. 275 00:17:45,320 --> 00:17:50,600 Speaker 2: It's really significant in terms of innovation and just being 276 00:17:50,640 --> 00:17:53,600 Speaker 2: an engine of change. Tell us a little bit about 277 00:17:53,600 --> 00:17:55,400 Speaker 2: what motivated push. 278 00:17:55,160 --> 00:17:59,200 Speaker 3: Play, right, I mean, I think it's a as you 279 00:17:59,359 --> 00:18:06,480 Speaker 3: just I think we have a tendency of not appreciate 280 00:18:06,720 --> 00:18:10,159 Speaker 3: appreciating the role of the play in our everyday life. 281 00:18:12,240 --> 00:18:13,879 Speaker 3: My motive is that we don't live to work, we 282 00:18:14,200 --> 00:18:16,439 Speaker 3: live to play, like we live to explore. When you 283 00:18:16,520 --> 00:18:19,520 Speaker 3: have like extra time, are you gonna do like one 284 00:18:19,520 --> 00:18:21,520 Speaker 3: more line of work, We're going to play. I think 285 00:18:21,560 --> 00:18:24,320 Speaker 3: it's a play is our kind of natural tendency or 286 00:18:24,560 --> 00:18:28,160 Speaker 3: homolludence as opposed to Homo sapiens. I mean, I think 287 00:18:28,200 --> 00:18:31,600 Speaker 3: it being an addition in addition to being a Homo sapiens. 288 00:18:31,880 --> 00:18:34,199 Speaker 3: So I think a play is a very important and 289 00:18:34,280 --> 00:18:38,080 Speaker 3: I think it's not not only about like we're not 290 00:18:38,119 --> 00:18:42,280 Speaker 3: only talking about computer games, but in general, I think 291 00:18:42,320 --> 00:18:45,840 Speaker 3: play has played a play has played a very significant 292 00:18:45,920 --> 00:18:50,080 Speaker 3: role in human evolution. Whenever there is a new artifact 293 00:18:50,160 --> 00:18:53,920 Speaker 3: that was introduced in our on our culture, we start 294 00:18:53,920 --> 00:18:56,280 Speaker 3: with the playing with it. We create, We kind of 295 00:18:56,280 --> 00:18:58,639 Speaker 3: figure out how to how to kind of how to 296 00:18:58,680 --> 00:19:02,520 Speaker 3: make how to make a play out of it, how 297 00:19:02,560 --> 00:19:04,600 Speaker 3: to kind of create something out of it. And when 298 00:19:04,600 --> 00:19:06,920 Speaker 3: we have a good understanding of the material and kind 299 00:19:06,960 --> 00:19:09,080 Speaker 3: of utility in it, then you kind of turned that 300 00:19:09,160 --> 00:19:13,720 Speaker 3: into utility. I think gaming has been playing that role 301 00:19:14,000 --> 00:19:19,920 Speaker 3: very diligently in over the last couple of decades. For example, 302 00:19:19,960 --> 00:19:22,480 Speaker 3: if you look at some type of technology that has 303 00:19:22,560 --> 00:19:29,400 Speaker 3: been when it was fully tested and baked out, gaming 304 00:19:29,800 --> 00:19:33,119 Speaker 3: has been always the platform that was brave enough to 305 00:19:33,320 --> 00:19:38,360 Speaker 3: incorporate in our in our offering and have our players 306 00:19:38,880 --> 00:19:42,399 Speaker 3: test kind of try out. For example, we had a 307 00:19:42,480 --> 00:19:47,160 Speaker 3: VP of AI since early two thousands. AI technology has 308 00:19:47,240 --> 00:19:52,800 Speaker 3: been not mature enough to kind of apply to applications 309 00:19:52,800 --> 00:19:56,800 Speaker 3: like a driver lest cars twenty years ago, but it 310 00:19:56,880 --> 00:19:59,240 Speaker 3: was okay in gaming because the gaming is a very 311 00:19:59,240 --> 00:20:02,720 Speaker 3: low risk and and at the same time gamers are 312 00:20:02,960 --> 00:20:06,920 Speaker 3: inherently early adopters they want to try out new technology. 313 00:20:07,000 --> 00:20:09,640 Speaker 3: So I think it's a combination of those things make 314 00:20:09,720 --> 00:20:15,080 Speaker 3: it possible and almost inevitable for gaming companies to try 315 00:20:15,160 --> 00:20:19,320 Speaker 3: new technologies. So not just the AI, but like kuber 316 00:20:19,400 --> 00:20:24,679 Speaker 3: netics discussions, or like a cloud, or even like a 317 00:20:24,720 --> 00:20:28,760 Speaker 3: free to play freemium business models all tried out in 318 00:20:28,840 --> 00:20:32,760 Speaker 3: gaming first before they were adopted in other businesses. 319 00:20:33,600 --> 00:20:35,280 Speaker 2: Let me throw you a little bit of a curveball 320 00:20:35,520 --> 00:20:39,280 Speaker 2: about gaming. So when I was growing up, play was 321 00:20:39,400 --> 00:20:43,480 Speaker 2: totally unstructured. You go down to the schoolyard. The computer 322 00:20:43,600 --> 00:20:49,119 Speaker 2: games like Pong and Space Invaders very rudimentary now it 323 00:20:49,160 --> 00:20:53,840 Speaker 2: seems kids their lives are much more scheduled, their play 324 00:20:53,920 --> 00:20:57,919 Speaker 2: is more structured. How does that affect the sort of 325 00:20:58,000 --> 00:21:01,600 Speaker 2: experience you want to provide from a gaming company. 326 00:21:02,600 --> 00:21:05,040 Speaker 3: That's a very good question, and I think it's a 327 00:21:06,840 --> 00:21:10,560 Speaker 3: There are many aspects to your question. One is about 328 00:21:10,640 --> 00:21:14,439 Speaker 3: what is gaming for today? And I think the reason 329 00:21:14,480 --> 00:21:19,359 Speaker 3: why there's so much opportunity to play game and kind 330 00:21:19,359 --> 00:21:23,399 Speaker 3: of as a novelty is because computer happens to be 331 00:21:23,640 --> 00:21:26,760 Speaker 3: the most sophisticated in advanced device that we have today. 332 00:21:27,200 --> 00:21:29,840 Speaker 3: I think we're trying to figure out what is kind 333 00:21:29,880 --> 00:21:32,840 Speaker 3: of its limitations and what it can do and it 334 00:21:32,960 --> 00:21:35,800 Speaker 3: and kind of in all with the kind of experience 335 00:21:35,800 --> 00:21:38,159 Speaker 3: that they can it can provide. So I think there 336 00:21:38,160 --> 00:21:41,639 Speaker 3: are a lot of on nine digital games out there, 337 00:21:42,000 --> 00:21:47,000 Speaker 3: and the kind of the size of the catalog makes 338 00:21:47,560 --> 00:21:50,200 Speaker 3: kids and who are trying to go through what kind 339 00:21:50,200 --> 00:21:54,200 Speaker 3: of play options are there end up choosing a game 340 00:21:54,320 --> 00:21:55,600 Speaker 3: or two from from. 341 00:21:55,400 --> 00:21:59,200 Speaker 2: That they can select what they want to do right exactly. 342 00:21:59,000 --> 00:22:01,560 Speaker 3: But but they're are but but also like if you 343 00:22:01,600 --> 00:22:03,320 Speaker 3: think of a game, game is not just one thing. 344 00:22:03,320 --> 00:22:05,680 Speaker 3: They are like sandbox games, they're like a building game. 345 00:22:05,720 --> 00:22:09,159 Speaker 3: There are like a coss game, the story based games. 346 00:22:09,280 --> 00:22:12,280 Speaker 3: There are different type of games that you can choose from, 347 00:22:12,560 --> 00:22:14,840 Speaker 3: So depending on your preference and what you find the 348 00:22:14,880 --> 00:22:18,200 Speaker 3: most engagement, you can choose different games. 349 00:22:18,440 --> 00:22:22,960 Speaker 2: So let's stay with kids. With children and in particular students. 350 00:22:23,359 --> 00:22:25,520 Speaker 2: There's been a lot of concern about the impact of 351 00:22:25,640 --> 00:22:31,280 Speaker 2: AI on education, on learning, on training people to get 352 00:22:31,359 --> 00:22:34,000 Speaker 2: jobs in the real world. There's a quote of years 353 00:22:34,200 --> 00:22:37,840 Speaker 2: I was kind of intrigued with, rather than competing with AI, 354 00:22:38,160 --> 00:22:43,360 Speaker 2: students should be prepared to leverage unuquely human capabilities. Explain 355 00:22:43,400 --> 00:22:45,800 Speaker 2: what that means in terms of the real world. 356 00:22:46,080 --> 00:22:49,320 Speaker 3: I think it's if you think about education, our education 357 00:22:50,160 --> 00:22:54,680 Speaker 3: has been over the last couple of one hundred years, 358 00:22:54,960 --> 00:22:59,000 Speaker 3: it has been optimized for delivering knowledge, and I think 359 00:22:59,119 --> 00:23:04,359 Speaker 3: weird witnessing that the knowledge delivery and memorization is rapidly 360 00:23:04,600 --> 00:23:09,280 Speaker 3: being commoditized. And is that what students have to spend 361 00:23:09,320 --> 00:23:13,120 Speaker 3: all their time sitting in a classroom to to kind 362 00:23:13,160 --> 00:23:15,840 Speaker 3: of to to kind of a polish I think it's 363 00:23:16,440 --> 00:23:18,879 Speaker 3: what we What our next generation needs is more of 364 00:23:18,880 --> 00:23:21,520 Speaker 3: the creativity and the problem solving skills, and I have 365 00:23:21,560 --> 00:23:25,000 Speaker 3: to think about if we can redesign the classroom to 366 00:23:25,080 --> 00:23:30,480 Speaker 3: really enhance those skills instead of helping them acquire one 367 00:23:30,520 --> 00:23:31,320 Speaker 3: more knowledge. 368 00:23:31,920 --> 00:23:36,360 Speaker 2: So there's a very different set of targets acquiring skills 369 00:23:36,480 --> 00:23:41,040 Speaker 2: or just learning things or memorizing things. I'm I'm a 370 00:23:41,040 --> 00:23:45,000 Speaker 2: big fan of teaching children how to problem solve. When 371 00:23:45,040 --> 00:23:47,960 Speaker 2: you say skills, let's let's think of that from an 372 00:23:48,000 --> 00:23:54,679 Speaker 2: institutional level. How should schools be using AI to teach 373 00:23:55,080 --> 00:24:00,760 Speaker 2: children new skills, developing and expertise developing problems? So what's 374 00:24:00,800 --> 00:24:05,160 Speaker 2: the proper role of AI for educational institutions? 375 00:24:05,800 --> 00:24:08,760 Speaker 3: I mean, I think what I what I would like 376 00:24:08,840 --> 00:24:13,320 Speaker 3: to say is that we have to educate and prepare 377 00:24:13,359 --> 00:24:17,240 Speaker 3: our students to thrive in the world. Where AI is more, 378 00:24:17,280 --> 00:24:21,359 Speaker 3: it's going to be more prevalent. AI is not. But 379 00:24:21,800 --> 00:24:25,359 Speaker 3: the solution to that is not just AI. It's many 380 00:24:25,560 --> 00:24:31,439 Speaker 3: different kind of It could be redesigning the curriculum. It 381 00:24:31,480 --> 00:24:34,240 Speaker 3: could be redesigning the kind of school system, the kind 382 00:24:34,240 --> 00:24:37,880 Speaker 3: of thinking about how we evaluate their achievement and how 383 00:24:37,960 --> 00:24:42,800 Speaker 3: to retrain our teachers. AI could be a tool for 384 00:24:43,000 --> 00:24:46,600 Speaker 3: doing that, but it's not a solution for everything. So 385 00:24:46,640 --> 00:24:48,320 Speaker 3: I think there is a huge difference there. 386 00:24:48,800 --> 00:24:50,800 Speaker 2: All right, So let's bring this now out to the 387 00:24:50,840 --> 00:24:56,600 Speaker 2: world of the economy and business. The probably the biggest 388 00:24:56,680 --> 00:25:01,520 Speaker 2: narrative I've been following about AI outside of it's a 389 00:25:01,560 --> 00:25:03,840 Speaker 2: bubble or we're all going to lose our jobs, and 390 00:25:03,880 --> 00:25:08,040 Speaker 2: I kind of ignore both of those extremes, is that 391 00:25:08,400 --> 00:25:12,560 Speaker 2: successful companies have wide moats and we're starting to see 392 00:25:12,680 --> 00:25:18,440 Speaker 2: AI compress those motes over time. Think about industries like lawyers, 393 00:25:19,040 --> 00:25:22,639 Speaker 2: tax prepared as accountants. There's a lot of stuff AI 394 00:25:22,720 --> 00:25:24,760 Speaker 2: can do in a fraction of the time and with 395 00:25:24,840 --> 00:25:29,280 Speaker 2: a greater accuracy level. Everybody knows about reading X rays 396 00:25:29,280 --> 00:25:34,639 Speaker 2: and MRIs. So if we know our motes are going 397 00:25:34,680 --> 00:25:39,399 Speaker 2: to get compressed, how should companies be using AI either 398 00:25:39,640 --> 00:25:43,760 Speaker 2: to protect and expand those motes or somehow use AI 399 00:25:43,960 --> 00:25:48,040 Speaker 2: to expand their competitive advantages while they last. 400 00:25:51,800 --> 00:25:56,439 Speaker 3: I mean, I think that's a very interesting because I 401 00:25:56,440 --> 00:25:59,480 Speaker 3: think there are some of the industries and profession that 402 00:25:59,600 --> 00:26:04,960 Speaker 3: will be they'll become much more productive and probably need 403 00:26:04,960 --> 00:26:08,800 Speaker 3: a lot less profession to solve certain well defined problems. 404 00:26:09,320 --> 00:26:12,639 Speaker 3: But that does mean that human as humanity as a 405 00:26:12,680 --> 00:26:16,679 Speaker 3: society is not is left with no problem to solve. 406 00:26:17,000 --> 00:26:18,919 Speaker 3: I think we have, like so many other problems to 407 00:26:18,960 --> 00:26:22,520 Speaker 3: solve that AI cannot address for example, how were I mean, 408 00:26:22,560 --> 00:26:29,720 Speaker 3: I think they are like very old problem of the 409 00:26:29,720 --> 00:26:34,439 Speaker 3: politics is how we are gonna redistribute the resources? What 410 00:26:34,600 --> 00:26:39,160 Speaker 3: is our societal priority in solving in enhancing the agency 411 00:26:39,200 --> 00:26:41,720 Speaker 3: of everyone and kind of helping them to achieve their 412 00:26:41,760 --> 00:26:44,080 Speaker 3: full potential. I think those are the things that we 413 00:26:44,119 --> 00:26:47,720 Speaker 3: don't have a good solution for. While AI can take 414 00:26:47,760 --> 00:26:50,639 Speaker 3: care of some of the things of a well defined workforce, 415 00:26:51,160 --> 00:26:55,240 Speaker 3: providing that expertise to make it more efficient, way will 416 00:26:55,320 --> 00:27:00,680 Speaker 3: have time to work on other problems to progress or 417 00:27:00,720 --> 00:27:04,119 Speaker 3: humanity forward. And I think that's we have prepared to 418 00:27:04,200 --> 00:27:06,720 Speaker 3: accept your type of roles and your type of kind 419 00:27:06,720 --> 00:27:11,840 Speaker 3: of professions that will exist enabled by this technology advencement. 420 00:27:12,320 --> 00:27:14,760 Speaker 2: So I think we're all in agreement. It's going to 421 00:27:14,760 --> 00:27:21,080 Speaker 2: be a very disruptive technology. Am I hearing you say? Essentially, Hey, 422 00:27:21,080 --> 00:27:23,640 Speaker 2: it's up to everybody to learn how to use these 423 00:27:23,680 --> 00:27:27,600 Speaker 2: tools and adapt. But the change is coming. You have 424 00:27:27,680 --> 00:27:28,000 Speaker 2: to be. 425 00:27:27,960 --> 00:27:29,840 Speaker 3: Prepared, yes, right, exactly. 426 00:27:30,119 --> 00:27:36,720 Speaker 2: Yeah, So you've operated at the intersection of artificial intelligence, gaming, telecommunication, 427 00:27:36,880 --> 00:27:40,040 Speaker 2: and social platforms. That's a great conversion of a lot 428 00:27:40,040 --> 00:27:45,720 Speaker 2: of different technologies. How is that evolving and how are 429 00:27:45,840 --> 00:27:51,600 Speaker 2: both consumers and institutions really adapting to that sort of 430 00:27:51,600 --> 00:27:52,960 Speaker 2: the AI driven economy. 431 00:27:56,440 --> 00:27:58,840 Speaker 3: I mean, I think it's a lot of people recognize 432 00:27:58,880 --> 00:28:02,080 Speaker 3: that this is one of the greatest platform shifts in 433 00:28:02,119 --> 00:28:05,439 Speaker 3: our in our lifetime, and there's a lot of excitement 434 00:28:05,840 --> 00:28:11,399 Speaker 3: uh and uh but as as many people say, I 435 00:28:11,440 --> 00:28:14,480 Speaker 3: mean the camp of uh seeing and as a kind 436 00:28:14,480 --> 00:28:18,280 Speaker 3: of we're at the very early inning of how its 437 00:28:18,600 --> 00:28:22,040 Speaker 3: kind of how it's gonna fully pan out. We don't 438 00:28:22,040 --> 00:28:24,639 Speaker 3: even know what's gonna what is coming in uh in 439 00:28:25,000 --> 00:28:28,920 Speaker 3: uh in the next three to five years and uh 440 00:28:29,040 --> 00:28:31,919 Speaker 3: and really excited to see all this kind of the 441 00:28:32,040 --> 00:28:37,000 Speaker 3: use case and application of technology fully using based on 442 00:28:37,040 --> 00:28:40,920 Speaker 3: the creativity of this AI native generation. And I think 443 00:28:40,960 --> 00:28:43,800 Speaker 3: the people who think with AI as part of their 444 00:28:44,280 --> 00:28:47,520 Speaker 3: tool and at their finger tip will come up with 445 00:28:47,560 --> 00:28:50,560 Speaker 3: the different ideas and and and kind of app like 446 00:28:50,600 --> 00:28:52,880 Speaker 3: their creativity. And I think that that that's what I'm 447 00:28:52,880 --> 00:28:53,760 Speaker 3: really excited about. 448 00:28:54,160 --> 00:28:58,640 Speaker 2: So you've founded Chameleon as a corporate venture arm and 449 00:28:58,640 --> 00:29:02,880 Speaker 2: and now you run a fully independent early stage venture funds. 450 00:29:03,360 --> 00:29:06,720 Speaker 2: I'm curious what are the differences between being part of 451 00:29:06,720 --> 00:29:10,440 Speaker 2: a corporate venture fund or an independent What are the 452 00:29:10,480 --> 00:29:14,000 Speaker 2: strengths and blind spots that you end up with in each. 453 00:29:18,480 --> 00:29:22,560 Speaker 3: I think we work with the strategic investors LP, then 454 00:29:22,600 --> 00:29:28,280 Speaker 3: I think there is a certain I think the objective 455 00:29:28,320 --> 00:29:30,320 Speaker 3: is a different depend I mean the kind of as 456 00:29:30,320 --> 00:29:32,640 Speaker 3: a GP we work for our LPs and the kind 457 00:29:32,680 --> 00:29:34,800 Speaker 3: of who is providing a capital and like what is 458 00:29:34,800 --> 00:29:36,920 Speaker 3: an objective with the fund and I think it's those 459 00:29:36,960 --> 00:29:39,760 Speaker 3: are very different. So at the PvP, I think we 460 00:29:39,960 --> 00:29:45,200 Speaker 3: focus more on a type of investors would like to 461 00:29:46,080 --> 00:29:48,520 Speaker 3: be at the forefront of the innovation and then capture 462 00:29:48,560 --> 00:29:52,920 Speaker 3: the value being created. I mean, regardless of the kind 463 00:29:52,920 --> 00:29:54,960 Speaker 3: of the area doesn't have to be confined in like 464 00:29:55,120 --> 00:29:58,360 Speaker 3: entertainment and consumer space. I think we were able to 465 00:29:58,640 --> 00:30:00,360 Speaker 3: kind of look more broadly. 466 00:30:00,200 --> 00:30:05,080 Speaker 2: So corporate is pure strategic and independent is strictly ROI. 467 00:30:05,480 --> 00:30:08,720 Speaker 2: So let's talk about some of the companies you've backed together, 468 00:30:09,000 --> 00:30:13,200 Speaker 2: dot Ai, Cartesia, Sesame. These all seem to be pretty 469 00:30:13,240 --> 00:30:17,000 Speaker 2: core infrastructure plays. Tell us a little bit about those. 470 00:30:17,760 --> 00:30:20,240 Speaker 2: What was it about each of those that made them 471 00:30:20,240 --> 00:30:22,120 Speaker 2: so appeal on I mean, I. 472 00:30:22,040 --> 00:30:25,600 Speaker 3: Think it's a it's a really tricky time to make 473 00:30:25,640 --> 00:30:27,960 Speaker 3: an investment because I mean, as we say, I think 474 00:30:27,960 --> 00:30:31,160 Speaker 3: there is a lot of kind of excitement about this technology. 475 00:30:31,200 --> 00:30:33,560 Speaker 3: There is a kind of the rushing mentality, so I 476 00:30:33,680 --> 00:30:38,120 Speaker 3: try to invest in companies uh that that's going to 477 00:30:38,160 --> 00:30:41,320 Speaker 3: be durable in the coming decades. So I really like 478 00:30:41,320 --> 00:30:46,160 Speaker 3: the companies that are building infrastructure technology that's that has 479 00:30:46,400 --> 00:30:50,280 Speaker 3: multi purpose us as this platform and technology evolves. So 480 00:30:50,320 --> 00:30:54,160 Speaker 3: I think Together and together in Cartesia are both of 481 00:30:54,200 --> 00:30:57,360 Speaker 3: them have a great founders, have a have a vision 482 00:30:57,400 --> 00:31:01,680 Speaker 3: of a building infrastructure and aunditional technology that's going to 483 00:31:01,680 --> 00:31:05,120 Speaker 3: be used uh in many different agie platform companies that 484 00:31:05,160 --> 00:31:07,680 Speaker 3: are gonna be building. And I think Sesame was an 485 00:31:07,680 --> 00:31:12,520 Speaker 3: interesting case because it's it's building the voice applications that 486 00:31:12,600 --> 00:31:15,520 Speaker 3: I know from my gaming experience. It's kind of importance 487 00:31:15,560 --> 00:31:19,719 Speaker 3: of it and importance of focusing on certain features that 488 00:31:19,720 --> 00:31:23,400 Speaker 3: that will provide certain experiences uh to to the users. 489 00:31:23,400 --> 00:31:26,920 Speaker 3: And I think the founders got what was important, and 490 00:31:27,000 --> 00:31:31,120 Speaker 3: I think they're kind of their capabilities and talent was 491 00:31:32,080 --> 00:31:36,400 Speaker 3: uh singularly focused on making that technology push. So I 492 00:31:36,440 --> 00:31:38,800 Speaker 3: really like what they were doing, and that's one of 493 00:31:38,880 --> 00:31:41,320 Speaker 3: the reasons and one of the reasons I ended up 494 00:31:41,360 --> 00:31:44,600 Speaker 3: investing in Sesame as a company. But I think there 495 00:31:44,600 --> 00:31:47,160 Speaker 3: are other type of companies as well that we're excited 496 00:31:47,200 --> 00:31:52,200 Speaker 3: about in this very in this kind of like sending 497 00:31:52,200 --> 00:31:53,560 Speaker 3: on this kind of shifting grounds. 498 00:31:53,800 --> 00:31:53,920 Speaker 2: Uh. 499 00:31:55,200 --> 00:31:57,800 Speaker 3: And I think those are the companies who are in 500 00:31:57,840 --> 00:32:00,880 Speaker 3: a position of building this data fly I will, because 501 00:32:01,360 --> 00:32:07,120 Speaker 3: one of the non deniable characteristics of the companies that 502 00:32:07,240 --> 00:32:10,280 Speaker 3: will be durable in this environment are the ones who 503 00:32:10,360 --> 00:32:15,200 Speaker 3: have appropriate access to data, the data, understanding customers and 504 00:32:15,240 --> 00:32:18,840 Speaker 3: consumers and the business and build a technology, unique technology 505 00:32:18,880 --> 00:32:22,040 Speaker 3: on top of that. So we also are investing companies 506 00:32:22,040 --> 00:32:25,680 Speaker 3: that are building this data fly will that over time 507 00:32:26,040 --> 00:32:27,880 Speaker 3: building very defensible modes. 508 00:32:28,280 --> 00:32:31,800 Speaker 2: Really really interesting. Coming up, we continue our conversation with 509 00:32:31,920 --> 00:32:36,880 Speaker 2: Sony Yun, co founder and managing partner at Principal Ventures, 510 00:32:36,880 --> 00:32:42,080 Speaker 2: discussing the state of venture investing into artificial intelligence. Today, 511 00:32:42,560 --> 00:32:45,920 Speaker 2: I'm Barry Richholds. You're listening to Masters and Business on 512 00:32:46,120 --> 00:33:02,640 Speaker 2: Bloomberg Right now, Barryerhelts, you're listening to Masters and Business 513 00:33:02,640 --> 00:33:06,320 Speaker 2: on Bloomberg Radio. My extra special guest today is Samya Yun. 514 00:33:06,800 --> 00:33:11,880 Speaker 2: She is the founder and managing partner at Principal Mentored Partners, 515 00:33:12,320 --> 00:33:17,920 Speaker 2: an AI focused VC. So that's a fascinating phrase. You 516 00:33:17,920 --> 00:33:20,800 Speaker 2: don't really hear a lot of that up until recently, 517 00:33:21,560 --> 00:33:26,640 Speaker 2: what is the key problem Principal Venture Partners is trying 518 00:33:26,680 --> 00:33:28,600 Speaker 2: to solve in the world of AI today. 519 00:33:29,240 --> 00:33:32,960 Speaker 3: So we've started to back AI native companies, and I 520 00:33:33,000 --> 00:33:36,560 Speaker 3: think when we started, when we first talked about AI 521 00:33:36,720 --> 00:33:39,720 Speaker 3: native companies, that was not a very common phrase. People 522 00:33:39,760 --> 00:33:42,440 Speaker 3: ask me like, what do you mean by AI native companies? 523 00:33:42,480 --> 00:33:46,920 Speaker 3: And I had to explain what it meant. And I 524 00:33:46,960 --> 00:33:51,240 Speaker 3: think it's these days, I think it's more widely used term, 525 00:33:51,640 --> 00:33:55,040 Speaker 3: and we would like to build We'd like to back 526 00:33:55,080 --> 00:34:00,360 Speaker 3: companies who are fully embracing the technology of today and tomorrow, 527 00:34:01,880 --> 00:34:05,560 Speaker 3: led by founders who understand the technology and limitations of 528 00:34:05,600 --> 00:34:08,160 Speaker 3: it and able to come up with org design that 529 00:34:08,280 --> 00:34:11,600 Speaker 3: reflect the importance of this. So I think the in 530 00:34:11,680 --> 00:34:13,880 Speaker 3: terms of the size of the department, I think it 531 00:34:14,280 --> 00:34:19,719 Speaker 3: will be very different from companies built upon last generation 532 00:34:19,960 --> 00:34:23,800 Speaker 3: technology tech stack, And I think a type of leaders 533 00:34:23,840 --> 00:34:26,239 Speaker 3: and talents who are going to lead all these departments 534 00:34:26,280 --> 00:34:28,200 Speaker 3: are going to be different in terms of the use 535 00:34:28,200 --> 00:34:32,359 Speaker 3: of technology and their vision and solving the problems that's 536 00:34:32,480 --> 00:34:35,799 Speaker 3: relevant in the AI native era. Are the companies that 537 00:34:35,840 --> 00:34:37,720 Speaker 3: are really excite us, and then those are the companies 538 00:34:38,080 --> 00:34:39,840 Speaker 3: were focused and investing in. 539 00:34:40,200 --> 00:34:45,160 Speaker 2: So every time there's a new technology, everybody just kind 540 00:34:45,160 --> 00:34:47,520 Speaker 2: of sprinkles a little bit on it to catch a 541 00:34:47,520 --> 00:34:49,680 Speaker 2: little bit of the buzz. We had it with the 542 00:34:49,760 --> 00:34:52,040 Speaker 2: dot coms, we had it with the metaverse, we had 543 00:34:52,080 --> 00:34:56,640 Speaker 2: it with crypto, and now everybody's claiming they're an AI company, 544 00:34:56,719 --> 00:34:58,200 Speaker 2: or at least a lot of companies. How do you 545 00:34:58,280 --> 00:35:02,719 Speaker 2: distinguish between what is truly AI native and what is 546 00:35:02,920 --> 00:35:05,920 Speaker 2: just let's put a little dash of AI salt on this. 547 00:35:06,719 --> 00:35:08,719 Speaker 3: That's a good that's a very good question. I think 548 00:35:08,760 --> 00:35:13,480 Speaker 3: I have unfaired advantage by like working in a gaming company, 549 00:35:13,520 --> 00:35:17,760 Speaker 3: and I think gaming company actually gave me an experience 550 00:35:17,800 --> 00:35:22,600 Speaker 3: in gaming industry is like having a lens into the future, right, 551 00:35:22,760 --> 00:35:26,400 Speaker 3: because a lot of the technology and innovation happens in 552 00:35:26,480 --> 00:35:30,200 Speaker 3: gaming first, and it gives us kind of sense of 553 00:35:30,520 --> 00:35:36,480 Speaker 3: like is this type of technology adoptable and will the 554 00:35:36,480 --> 00:35:39,839 Speaker 3: consumers accept this type of application? So like in terms 555 00:35:39,880 --> 00:35:44,560 Speaker 3: of application and platform, I think that that's really an 556 00:35:44,560 --> 00:35:48,280 Speaker 3: interesting guiding north star for me. And then in addition 557 00:35:48,360 --> 00:35:50,680 Speaker 3: to that, I think that companies that are fully AI 558 00:35:50,800 --> 00:35:55,680 Speaker 3: native are built around that text egg. Whereas if you 559 00:35:55,760 --> 00:35:58,440 Speaker 3: are trying to splinkling AI. I think you kind of 560 00:35:58,480 --> 00:36:02,000 Speaker 3: ask questions like can you do the same thing without AI? 561 00:36:02,680 --> 00:36:05,359 Speaker 3: Like why do you need it? Why is it dispensable? 562 00:36:05,400 --> 00:36:08,880 Speaker 3: I think there are kind of the businesses and business 563 00:36:08,880 --> 00:36:13,600 Speaker 3: operations that they're using, for example, like agent technology, but 564 00:36:13,760 --> 00:36:16,080 Speaker 3: like a lot of the applications and you don't need 565 00:36:16,120 --> 00:36:18,279 Speaker 3: the agents, you just need some kind of good, good 566 00:36:18,360 --> 00:36:21,080 Speaker 3: data analytics. So I think it's a There are many 567 00:36:21,280 --> 00:36:25,200 Speaker 3: ways that we tried to understand how businesses that were 568 00:36:25,239 --> 00:36:29,600 Speaker 3: operating and see their kind of full potential like and 569 00:36:29,680 --> 00:36:30,400 Speaker 3: their strategy. 570 00:36:30,800 --> 00:36:34,080 Speaker 2: So on the one hand, I know AI has been 571 00:36:34,120 --> 00:36:37,799 Speaker 2: around a long time. There was you know when Deep 572 00:36:37,840 --> 00:36:41,360 Speaker 2: Blue beat Casparov that was a big deal, and then 573 00:36:41,719 --> 00:36:46,480 Speaker 2: I forgot the name of the AI app that ended 574 00:36:46,560 --> 00:36:49,920 Speaker 2: up winning Jeopardy. These are like ten and twenty years ago, 575 00:36:50,080 --> 00:36:54,840 Speaker 2: so it's not a brand new technology. However, it feels 576 00:36:54,880 --> 00:37:01,880 Speaker 2: like we took another level jump with Chacchi and go 577 00:37:01,960 --> 00:37:07,920 Speaker 2: down the list Claude notebook, perplexity, whatever. How do you 578 00:37:08,080 --> 00:37:12,319 Speaker 2: think about this moment in time? Is this similar to 579 00:37:12,520 --> 00:37:17,480 Speaker 2: early broadbands, early smartphones, early cloud use? Like for someone 580 00:37:17,520 --> 00:37:20,360 Speaker 2: who's a tech investor, they want to know, hey, is 581 00:37:20,400 --> 00:37:23,040 Speaker 2: it early Is it late? Is it so early that 582 00:37:23,600 --> 00:37:25,520 Speaker 2: nine out of ten of the startups we're going to 583 00:37:25,520 --> 00:37:27,160 Speaker 2: see you are going to go bell? Yeah, Like, how 584 00:37:27,160 --> 00:37:29,720 Speaker 2: do you think about this moment where we are today? 585 00:37:30,360 --> 00:37:34,920 Speaker 3: That's great? Actually it's older than that. Do you remember, 586 00:37:35,080 --> 00:37:38,960 Speaker 3: like in sixties there is a like a location called 587 00:37:39,000 --> 00:37:40,680 Speaker 3: Eliza that. 588 00:37:40,719 --> 00:37:43,560 Speaker 2: Sounds very familiar, right, So telephone. 589 00:37:43,120 --> 00:37:49,480 Speaker 3: Or Eliza was was it a kind of very early 590 00:37:49,480 --> 00:37:53,960 Speaker 3: incarnation of checkbot? And I think there was even a 591 00:37:54,000 --> 00:37:59,279 Speaker 3: newspaper headline at the end of psychotherapist because it was 592 00:37:59,320 --> 00:38:02,839 Speaker 3: doing so bad well in terms of like rephrasing what 593 00:38:02,840 --> 00:38:06,040 Speaker 3: we're asking and kind of giving them comfort when you're 594 00:38:06,080 --> 00:38:09,200 Speaker 3: asking about very personal questions and this, oh, we don't 595 00:38:09,239 --> 00:38:12,200 Speaker 3: need a psychotherapists anymore because it's doing good enough. And 596 00:38:12,239 --> 00:38:14,480 Speaker 3: I think since then there was a lot of kind 597 00:38:14,520 --> 00:38:18,920 Speaker 3: of AI winters and summer ups and downs, and I 598 00:38:18,960 --> 00:38:22,080 Speaker 3: think there's a lot of hype in terms of like 599 00:38:22,320 --> 00:38:26,120 Speaker 3: comparing where we are in terms of the kind of 600 00:38:27,200 --> 00:38:31,879 Speaker 3: platform shift, and I think we're I think we're what's 601 00:38:31,920 --> 00:38:36,960 Speaker 3: surprising to many people about this time and movement is 602 00:38:37,000 --> 00:38:42,320 Speaker 3: that the the AI shift is closer to the introduction 603 00:38:42,400 --> 00:38:47,160 Speaker 3: of the railroad than the introduction of the PC or Internet, 604 00:38:47,920 --> 00:38:52,680 Speaker 3: because the biggest breakthrough that that allowed us to come 605 00:38:52,719 --> 00:38:57,360 Speaker 3: here was actually the scale. So it's not just a 606 00:38:57,440 --> 00:39:01,680 Speaker 3: new it's not new algorithm, it's not new software, kind 607 00:39:01,719 --> 00:39:04,359 Speaker 3: of new way of doing things, but it was a 608 00:39:04,400 --> 00:39:07,160 Speaker 3: scale let's do it like kind of pouring a lot 609 00:39:07,200 --> 00:39:11,040 Speaker 3: of resources to make it really big, and that's that 610 00:39:11,239 --> 00:39:15,080 Speaker 3: that's where we saw the tremendous jump in terms jump 611 00:39:15,160 --> 00:39:19,759 Speaker 3: in terms of the capability of AI. So with that, 612 00:39:19,920 --> 00:39:23,799 Speaker 3: I think it's kind of rolling out that that as 613 00:39:23,840 --> 00:39:28,799 Speaker 3: an infrastructure has been the focus of the last two 614 00:39:28,840 --> 00:39:34,560 Speaker 3: three years, and on this exciting new railroad that's built out. 615 00:39:34,760 --> 00:39:37,480 Speaker 3: I think there will be interesting new businesses that will 616 00:39:37,520 --> 00:39:40,080 Speaker 3: emerge out of it. So yes, I think we're very 617 00:39:40,080 --> 00:39:45,000 Speaker 3: early in terms of kind of fully appreciating what's possible 618 00:39:45,080 --> 00:39:45,759 Speaker 3: on top of this. 619 00:39:47,080 --> 00:39:50,560 Speaker 2: So I love the idea of interesting new businesses. And 620 00:39:50,600 --> 00:39:54,560 Speaker 2: I'm always fascinated with what do the public markets know? 621 00:39:55,200 --> 00:39:57,880 Speaker 2: You know, it's they're more or less kind of eventually 622 00:39:57,920 --> 00:40:02,320 Speaker 2: efficient and very often though, when a new technology comes along, 623 00:40:02,840 --> 00:40:07,080 Speaker 2: they very much underestimate where it can go. So what's 624 00:40:07,080 --> 00:40:10,839 Speaker 2: the sort of use case that the public markets might 625 00:40:10,920 --> 00:40:14,800 Speaker 2: be underestimating where might this go? You look at dozens 626 00:40:14,840 --> 00:40:19,640 Speaker 2: and dozens of new companies, what what direction is just 627 00:40:19,760 --> 00:40:22,360 Speaker 2: mind blowing that nobody is really anticipating. 628 00:40:25,000 --> 00:40:27,600 Speaker 3: So there are I think it's a there are a 629 00:40:27,640 --> 00:40:32,680 Speaker 3: lot of things happening. One is that one interesting thing 630 00:40:32,719 --> 00:40:40,719 Speaker 3: about this technology is that while it has been kind 631 00:40:40,760 --> 00:40:44,520 Speaker 3: of it, it beat many people's expectations about what it 632 00:40:44,560 --> 00:40:47,879 Speaker 3: can do. However, I think there is a lot more 633 00:40:47,920 --> 00:40:52,799 Speaker 3: innovation is coming along because in terms of this architecture 634 00:40:52,880 --> 00:40:57,080 Speaker 3: design and fundamental design of the framework, there is a 635 00:40:57,120 --> 00:41:00,720 Speaker 3: lot more innovation coming along. So we're not done with 636 00:41:01,200 --> 00:41:04,439 Speaker 3: what is the most efficient railroad design. I think there 637 00:41:04,480 --> 00:41:08,360 Speaker 3: could be other type of railroad that can come along 638 00:41:08,440 --> 00:41:12,279 Speaker 3: online that will allow faster and more kind of more 639 00:41:12,320 --> 00:41:16,279 Speaker 3: comfortable right experience. And once there is a railroad, I 640 00:41:16,320 --> 00:41:20,280 Speaker 3: think there was an interesting business that emerged, like for example, 641 00:41:20,360 --> 00:41:24,799 Speaker 3: like mail order or like people. It's really hard to 642 00:41:24,840 --> 00:41:27,520 Speaker 3: make a connection, but that type of new businesses was 643 00:41:28,120 --> 00:41:32,360 Speaker 3: made possible because railroad was in place. But when I 644 00:41:32,360 --> 00:41:33,640 Speaker 3: don't think, I don't think it was the kind of 645 00:41:33,640 --> 00:41:36,320 Speaker 3: first thing that came in mind. We're kind of rolling 646 00:41:36,320 --> 00:41:41,040 Speaker 3: out the railroad as a kind of cross across different states. 647 00:41:40,640 --> 00:41:43,839 Speaker 2: Well, well, broadband and fiber optic led to so many 648 00:41:43,960 --> 00:41:46,960 Speaker 2: different everything from YouTube exactly, really the build out of 649 00:41:47,120 --> 00:41:48,920 Speaker 2: Amazon Web services and. 650 00:41:48,880 --> 00:41:53,040 Speaker 3: Right on nine games on nine games right exactly. I 651 00:41:53,040 --> 00:41:56,000 Speaker 3: think so like So that's why I'm really excited about 652 00:41:56,360 --> 00:41:59,080 Speaker 3: AI native generations and creativity what they're going to build 653 00:41:59,080 --> 00:42:00,840 Speaker 3: on top of this. So I think there will be 654 00:42:01,120 --> 00:42:04,120 Speaker 3: new type of businesses that we don't we don't comprehend 655 00:42:04,120 --> 00:42:06,920 Speaker 3: today that will be enabled by this infrastructure. 656 00:42:07,320 --> 00:42:10,440 Speaker 2: So when you're sitting with the founder of a company 657 00:42:10,480 --> 00:42:14,440 Speaker 2: that's looking for financing, what sort of questions do you ask? 658 00:42:14,680 --> 00:42:17,600 Speaker 2: What are you trying to figure out about their model, 659 00:42:17,680 --> 00:42:23,040 Speaker 2: about their direction, about their team. It's so so unique 660 00:42:23,080 --> 00:42:24,080 Speaker 2: and cutting edge. 661 00:42:24,200 --> 00:42:26,960 Speaker 3: I mean, I think there it depends depending on what 662 00:42:27,040 --> 00:42:30,080 Speaker 3: they're building. I think it's the set of questions that 663 00:42:30,120 --> 00:42:33,440 Speaker 3: I ask when they're building more of the infrastructure. Technology 664 00:42:33,920 --> 00:42:38,960 Speaker 3: versus kind of business applications are different, but especially when 665 00:42:38,960 --> 00:42:44,799 Speaker 3: they're building business applications or bodical applications, I always try 666 00:42:44,840 --> 00:42:48,080 Speaker 3: to ask what is the real value that is going 667 00:42:48,160 --> 00:42:53,200 Speaker 3: to bring to the end users and not we are 668 00:42:53,200 --> 00:42:56,719 Speaker 3: not investing in companies that are building amazing kind of 669 00:42:56,880 --> 00:43:00,920 Speaker 3: tech demonstrations and the features. We're trying to find the 670 00:43:00,960 --> 00:43:04,040 Speaker 3: companies who are built kind of solving real business, real 671 00:43:04,080 --> 00:43:06,879 Speaker 3: world business problems and do it in a way that's 672 00:43:06,920 --> 00:43:10,839 Speaker 3: sustainable and more efficient and using other type of technology. 673 00:43:11,320 --> 00:43:16,080 Speaker 2: So you're looking at infrastructure type companies. What other type 674 00:43:16,080 --> 00:43:18,680 Speaker 2: of AI applications are you looking at. 675 00:43:19,000 --> 00:43:22,640 Speaker 3: We're looking at the companies that are building vertical applications 676 00:43:23,360 --> 00:43:26,320 Speaker 3: by developing data kind of flyable and data mode. 677 00:43:28,040 --> 00:43:31,760 Speaker 2: So so there's been a little bit of a lightning rod. 678 00:43:32,000 --> 00:43:36,120 Speaker 2: From a regulatory standpoint, there's you know, all the lllms 679 00:43:36,640 --> 00:43:41,080 Speaker 2: have copyright complaints and issues. You sit in a really 680 00:43:41,120 --> 00:43:45,600 Speaker 2: interesting intersection when when you look at a term sheet today, 681 00:43:45,640 --> 00:43:50,080 Speaker 2: how do you think about the regulatory risks, the litigation risks. 682 00:43:50,160 --> 00:43:54,360 Speaker 2: I mean that when llms first came out, I didn't 683 00:43:54,400 --> 00:43:57,200 Speaker 2: for a moment think oh no, oh, that was stolen 684 00:43:57,400 --> 00:44:01,319 Speaker 2: content and there that that they were crolling over and 685 00:44:01,440 --> 00:44:04,360 Speaker 2: using as their training tools. How do you think about 686 00:44:04,880 --> 00:44:09,080 Speaker 2: regulatory framework and geopolitics. It seems like there's a lot 687 00:44:09,120 --> 00:44:11,040 Speaker 2: of novel moving parts. 688 00:44:11,239 --> 00:44:14,080 Speaker 3: Yeah, I think that's a really great question, and I 689 00:44:14,080 --> 00:44:19,200 Speaker 3: think that more than ever, understanding how the regulatory body 690 00:44:19,280 --> 00:44:22,600 Speaker 3: thinks and the policy is gonna gonna evolve over time 691 00:44:23,560 --> 00:44:28,400 Speaker 3: is important at making this decisions, especially in the venture space. 692 00:44:28,640 --> 00:44:33,560 Speaker 3: We are making a bet, We're making investment that should 693 00:44:34,360 --> 00:44:37,040 Speaker 3: last over a decade, right. 694 00:44:37,400 --> 00:44:37,640 Speaker 1: So. 695 00:44:39,160 --> 00:44:42,600 Speaker 3: I think it comes from the kind of the belief 696 00:44:42,600 --> 00:44:49,680 Speaker 3: and understanding that the innovation and the research is really 697 00:44:49,840 --> 00:44:54,200 Speaker 3: kind of the is very precious for all of us 698 00:44:54,640 --> 00:45:04,120 Speaker 3: as a humanity. And the kind of the this kind 699 00:45:04,120 --> 00:45:07,120 Speaker 3: of peer reviewed, the kind of the tradition of the 700 00:45:07,160 --> 00:45:13,479 Speaker 3: peer review and open forum has has really propelled us 701 00:45:13,680 --> 00:45:16,120 Speaker 3: to where we are today and it's going to continue. 702 00:45:16,800 --> 00:45:20,560 Speaker 3: And I think the collaboration and openness will better serve 703 00:45:20,640 --> 00:45:25,440 Speaker 3: our end customers for so the transparency is important. And 704 00:45:25,480 --> 00:45:27,640 Speaker 3: I think with all of those beliefs, I mean, I 705 00:45:27,640 --> 00:45:29,879 Speaker 3: think we don't have we don't have the crystal ball 706 00:45:30,000 --> 00:45:32,319 Speaker 3: to say, like what's gonna what the police frameworkwork is 707 00:45:32,360 --> 00:45:34,560 Speaker 3: gonna be, Like what is a coate geopolitical tension it's 708 00:45:34,600 --> 00:45:37,239 Speaker 3: going to be like in the next one or two years. 709 00:45:37,560 --> 00:45:41,719 Speaker 3: But we have the belief that human kind of our 710 00:45:41,840 --> 00:45:46,160 Speaker 3: collective work will converge in a direction that serves our 711 00:45:46,239 --> 00:45:49,560 Speaker 3: humanity in a positive direction. And I think that's uh, 712 00:45:49,680 --> 00:45:52,200 Speaker 3: that's kind of where I'd like to uh to see 713 00:45:52,600 --> 00:45:56,040 Speaker 3: that that's ultimately that's gonna what's going to be implemented 714 00:45:56,080 --> 00:46:01,600 Speaker 3: and incorporated and under under that type of the world, 715 00:46:02,160 --> 00:46:06,680 Speaker 3: whether these companies and founders can test a pressure test 716 00:46:07,160 --> 00:46:09,560 Speaker 3: to sustain and I think that's how I see and 717 00:46:10,200 --> 00:46:13,160 Speaker 3: when I evaluate the companies, how it's going to sustain 718 00:46:13,200 --> 00:46:14,480 Speaker 3: those kind of policy changes. 719 00:46:15,280 --> 00:46:17,440 Speaker 2: All right, So before we get to our speed round, 720 00:46:17,520 --> 00:46:20,640 Speaker 2: let me ask you one last question, which is what 721 00:46:20,680 --> 00:46:24,319 Speaker 2: do you think in investors in the AI space are 722 00:46:24,400 --> 00:46:27,520 Speaker 2: either not thinking about or not talking about, but is 723 00:46:27,560 --> 00:46:30,600 Speaker 2: important and perhaps they really should be paying attention to. 724 00:46:34,160 --> 00:46:37,120 Speaker 3: I think that the saying that says we are very 725 00:46:37,160 --> 00:46:42,520 Speaker 3: at the very early inning means a lot. I hear 726 00:46:42,680 --> 00:46:44,759 Speaker 3: someone even saying that we are still in the car 727 00:46:44,960 --> 00:46:47,920 Speaker 3: getting to the stadium. You're not even in the first inning. 728 00:46:49,239 --> 00:46:53,719 Speaker 3: That means all this kind of the models, models and 729 00:46:53,800 --> 00:47:01,080 Speaker 3: structures can change significantly, can evolve over time, and nothing 730 00:47:01,400 --> 00:47:07,960 Speaker 3: cannot be seen as kind of engraved in a stone. 731 00:47:08,520 --> 00:47:11,480 Speaker 3: So with that, I think a lot of the investment 732 00:47:11,520 --> 00:47:18,160 Speaker 3: decisions has to be reflecting the fact that there it 733 00:47:18,440 --> 00:47:20,680 Speaker 3: kind of has to remain like a nimble and flexible 734 00:47:20,719 --> 00:47:24,120 Speaker 3: because we should be able to adjust to when those 735 00:47:24,239 --> 00:47:26,360 Speaker 3: changes and new breakthroughs come around. 736 00:47:27,000 --> 00:47:29,400 Speaker 2: All right, So I only have you for a few minutes, 737 00:47:29,760 --> 00:47:32,320 Speaker 2: so we'll click through these really quickly. Our speed round. 738 00:47:32,760 --> 00:47:36,839 Speaker 2: Starting with who your early mentors who helped to shape. 739 00:47:36,560 --> 00:47:40,120 Speaker 3: Your career, I would say, I mean, I think I 740 00:47:40,160 --> 00:47:44,200 Speaker 3: was fortunate enough to have a lot of mentors when 741 00:47:44,239 --> 00:47:47,759 Speaker 3: I was a student, But one person that stands out 742 00:47:47,880 --> 00:47:52,560 Speaker 3: is Dominic Barton, who was the global menting partner at 743 00:47:52,880 --> 00:47:56,560 Speaker 3: McKinsey when I first started out as an associate of McKinsey. 744 00:47:57,000 --> 00:48:00,359 Speaker 3: He had his office was right next to mine, so 745 00:48:01,080 --> 00:48:04,879 Speaker 3: he was literally my mentor. And I think I learned 746 00:48:04,920 --> 00:48:07,400 Speaker 3: a lot from him as a as a leader and 747 00:48:09,800 --> 00:48:14,920 Speaker 3: as a as a mentor. And I think it's still today. 748 00:48:14,960 --> 00:48:16,719 Speaker 3: I reach out to him if I have to make 749 00:48:16,840 --> 00:48:19,600 Speaker 3: tough decisions, and I think he has been always very 750 00:48:19,600 --> 00:48:21,920 Speaker 3: generous with his time, so I really appreciate. 751 00:48:22,360 --> 00:48:24,480 Speaker 2: Let's talk about books. What are some of your favorites? 752 00:48:24,520 --> 00:48:25,720 Speaker 2: What are you reading right now? 753 00:48:26,160 --> 00:48:28,920 Speaker 3: Oh? So I read a lot of books, but I 754 00:48:28,920 --> 00:48:32,080 Speaker 3: think I kind of read many books of the type 755 00:48:32,080 --> 00:48:35,279 Speaker 3: that I read many books that kind of simultaneously one 756 00:48:35,360 --> 00:48:38,359 Speaker 3: chapter here and then I go to another book. But 757 00:48:38,680 --> 00:48:41,120 Speaker 3: the book that I recommend to everyone these days is 758 00:48:41,200 --> 00:48:46,520 Speaker 3: our two but one is the Empire of AI and 759 00:48:46,600 --> 00:48:49,440 Speaker 3: the other is Power and Progress. And I think that 760 00:48:49,840 --> 00:48:53,080 Speaker 3: those books kind of help us understand the dynamics of 761 00:48:53,080 --> 00:48:55,600 Speaker 3: what's happening and what we need to think about as 762 00:48:55,680 --> 00:48:56,640 Speaker 3: a society. 763 00:48:56,960 --> 00:48:59,480 Speaker 2: So let's talk about streaming. What are you either listening 764 00:48:59,480 --> 00:49:00,960 Speaker 2: to or watching these days? 765 00:49:01,960 --> 00:49:07,480 Speaker 3: So, I mean I listened to I listened to music 766 00:49:07,560 --> 00:49:11,719 Speaker 3: through like Spotify a lot. My son is a big 767 00:49:11,760 --> 00:49:15,560 Speaker 3: fan of Taylor Swift. I have to listen to Taylor 768 00:49:15,600 --> 00:49:20,040 Speaker 3: Swift like when I'm in a car, like driving a lot. 769 00:49:21,200 --> 00:49:23,720 Speaker 3: Also I watched k Drama on Netflix. 770 00:49:23,880 --> 00:49:27,759 Speaker 2: Really really interesting. Our final two questions, what sort of 771 00:49:27,800 --> 00:49:31,480 Speaker 2: advice would you give to a recent college graduate interest 772 00:49:31,560 --> 00:49:36,600 Speaker 2: in the career in either artificial intelligence, fest thing or gaming. 773 00:49:38,840 --> 00:49:45,480 Speaker 3: I mean, I think it kids just graduating today. I mean, 774 00:49:45,480 --> 00:49:49,080 Speaker 3: I think it's a one thing that's not gonna change 775 00:49:49,160 --> 00:49:51,319 Speaker 3: is it's gonna be very bumpy, and it's gonna be 776 00:49:51,400 --> 00:49:55,600 Speaker 3: like there's disruptive, and the world they're gonna be working 777 00:49:55,600 --> 00:49:59,239 Speaker 3: in is not gonna look like the world today. Like 778 00:49:59,320 --> 00:50:03,640 Speaker 3: that's that's the constant, right, And I mean I think 779 00:50:03,680 --> 00:50:05,359 Speaker 3: what I would like to remind them is like, don't 780 00:50:05,400 --> 00:50:11,480 Speaker 3: try to find that kind of follow the trend. Really 781 00:50:11,520 --> 00:50:13,960 Speaker 3: have to stick to what you're I mean it sounds 782 00:50:13,960 --> 00:50:16,799 Speaker 3: like a cliche, like what you're passionate about. I mean, 783 00:50:16,840 --> 00:50:20,440 Speaker 3: you remember in the seventies, the most popular major to 784 00:50:20,520 --> 00:50:22,520 Speaker 3: go in was what was that was like kind of 785 00:50:22,520 --> 00:50:25,680 Speaker 3: material science, and then like chemical engineering, and then like 786 00:50:25,760 --> 00:50:29,920 Speaker 3: electric engineering and computer like just to see that those 787 00:50:30,120 --> 00:50:34,359 Speaker 3: those the popularity of those major kind of plummeting, Like 788 00:50:34,400 --> 00:50:37,600 Speaker 3: we have witnesses like so many of those cases. So 789 00:50:37,960 --> 00:50:41,040 Speaker 3: it doesn't really I don't think it serves well to 790 00:50:41,120 --> 00:50:44,719 Speaker 3: find what's try to kind of follow that that kind 791 00:50:44,719 --> 00:50:45,800 Speaker 3: of fashion or trend. 792 00:50:46,000 --> 00:50:49,000 Speaker 2: Really be a generalist and be flexible. 793 00:50:50,880 --> 00:50:52,399 Speaker 3: Could be yeah right, yeah, all right. 794 00:50:52,440 --> 00:50:54,920 Speaker 2: And our final question, what do you know about the 795 00:50:54,960 --> 00:51:00,400 Speaker 2: world of venture investing and artificial intelligence today? Might have 796 00:51:00,480 --> 00:51:02,200 Speaker 2: been useful to know twenty years. 797 00:51:01,960 --> 00:51:06,799 Speaker 3: Ago, I mean, I think the patients so I mean 798 00:51:07,080 --> 00:51:10,880 Speaker 3: it's also the power of compounding is not just in finance, 799 00:51:10,920 --> 00:51:15,239 Speaker 3: but also in like human capital and our understanding technology 800 00:51:16,360 --> 00:51:21,520 Speaker 3: and also the relations too. Like it seems very slow today, 801 00:51:21,680 --> 00:51:26,000 Speaker 3: but like if it you're persistent and for twenty years, 802 00:51:26,040 --> 00:51:28,799 Speaker 3: I think that what you can achieve is really tremendous. 803 00:51:29,280 --> 00:51:31,920 Speaker 2: Well, thank you song Ye for being so generous with 804 00:51:31,960 --> 00:51:34,879 Speaker 2: your time. We have been speaking with Sung Yi yun, 805 00:51:35,960 --> 00:51:40,759 Speaker 2: founder and managing partner at Principal Venture Partners. If you 806 00:51:41,040 --> 00:51:44,160 Speaker 2: enjoyed this conversation, well check out any of the six 807 00:51:44,239 --> 00:51:48,480 Speaker 2: hundred plus interviews we've done over the past twelve years. 808 00:51:48,840 --> 00:51:54,799 Speaker 2: You can find those at iTunes, podcasts, Spotify, YouTube, Bloomberg, 809 00:51:55,000 --> 00:51:59,000 Speaker 2: wherever you find your favorite podcasts. I would be remiss 810 00:51:59,000 --> 00:52:01,000 Speaker 2: if I didn't thank the Craft team that helps us 811 00:52:01,000 --> 00:52:06,279 Speaker 2: put these conversations together each week. Alexis Noriega is my 812 00:52:06,440 --> 00:52:11,240 Speaker 2: video producer. Anna Luke is my podcast producer. Jean Russo 813 00:52:11,360 --> 00:52:14,680 Speaker 2: is my head of research. I'm Barry Rittolts. You've been 814 00:52:14,719 --> 00:52:24,440 Speaker 2: listening to Masters and Business on Bloomberg Radio.