1 00:00:00,500 --> 00:00:01,964 [SQUEAKING] 2 00:00:01,964 --> 00:00:04,419 [RUSTLING] 3 00:00:04,419 --> 00:00:05,892 [CLICKING] 4 00:00:09,843 --> 00:00:11,260 GARY GENSLER: On Monday, we talked 5 00:00:11,260 --> 00:00:13,540 about financial technology as a whole, 6 00:00:13,540 --> 00:00:16,210 and what we're going to do in these next four classes 7 00:00:16,210 --> 00:00:20,590 is really talk about three major technologies. 8 00:00:20,590 --> 00:00:24,250 The broad subject of artificial intelligence, machine learning, 9 00:00:24,250 --> 00:00:27,100 and deep learning, we'll talk about today, 10 00:00:27,100 --> 00:00:30,760 and next Monday, we'll move on and talk about the marketing 11 00:00:30,760 --> 00:00:35,080 channels, what I'll broadly call OpenAPI and some 12 00:00:35,080 --> 00:00:39,640 of those conversational agents and the relationship that 13 00:00:39,640 --> 00:00:41,210 is happening there. 14 00:00:41,210 --> 00:00:45,930 And then next Wednesday's class, I might be off. 15 00:00:45,930 --> 00:00:47,830 It's the following Monday's class, 16 00:00:47,830 --> 00:00:50,680 we'll talk about blockchain technology and cryptocurrency. 17 00:00:50,680 --> 00:00:53,920 So we're taking three big slices of technology, 18 00:00:53,920 --> 00:00:56,190 and then we're going to go into the sectors. 19 00:00:56,190 --> 00:01:00,430 And so what I thought of about in structuring this class, 20 00:01:00,430 --> 00:01:03,790 as we've already discussed, is I take a broad view 21 00:01:03,790 --> 00:01:08,470 of what the subject of fintech is, that it's technologies 22 00:01:08,470 --> 00:01:16,080 of our time that are potentially materially affecting finance. 23 00:01:16,080 --> 00:01:18,525 So in any decade, any five year, it 24 00:01:18,525 --> 00:01:20,010 might be different technologies. 25 00:01:20,010 --> 00:01:22,680 It's technologies of a certain time that are materially 26 00:01:22,680 --> 00:01:25,800 affecting finance, and it's the broad-- 27 00:01:25,800 --> 00:01:27,750 any competitor can use fintech. 28 00:01:27,750 --> 00:01:30,810 Now, I recognize that a lot of people 29 00:01:30,810 --> 00:01:33,270 use the subject or fintech, and they narrow it down, 30 00:01:33,270 --> 00:01:36,420 and they just use those terms for the disruptors, 31 00:01:36,420 --> 00:01:38,190 the startups. 32 00:01:38,190 --> 00:01:39,690 And there were really good questions 33 00:01:39,690 --> 00:01:42,990 of why I think of it more broadly. 34 00:01:42,990 --> 00:01:47,010 I just think that the incumbents are very much involved 35 00:01:47,010 --> 00:01:49,260 in this fintech wave, and to ignore 36 00:01:49,260 --> 00:01:54,640 that would be at your peril if you're starting a new business, 37 00:01:54,640 --> 00:01:56,520 if you're an entrepreneur. 38 00:01:56,520 --> 00:02:01,170 And of course, if you're Jamie Dimon running JPMorgan Chase, 39 00:02:01,170 --> 00:02:03,877 or you're running a big bank in China or in Europe, 40 00:02:03,877 --> 00:02:06,210 you would be at your peril if you didn't think about it. 41 00:02:06,210 --> 00:02:11,020 And big tech, of course, has found their way in here. 42 00:02:11,020 --> 00:02:15,403 The Alibaba example's in China, but all around the globe, 43 00:02:15,403 --> 00:02:17,070 whether it's Brazil, whether it's India, 44 00:02:17,070 --> 00:02:21,780 whether it's-- your big tech has found their way in here. 45 00:02:21,780 --> 00:02:24,300 And so I've also decided to sort of structure it 46 00:02:24,300 --> 00:02:28,470 around these three big thematic technologies, 47 00:02:28,470 --> 00:02:32,040 and then we'll start to dive into the sectors themselves 48 00:02:32,040 --> 00:02:34,680 and take a look at four or five sectors 49 00:02:34,680 --> 00:02:40,410 before we call it a day at the end of the half semester. 50 00:02:40,410 --> 00:02:42,780 So I'm about to pull up some slides, but, Romain, 51 00:02:42,780 --> 00:02:45,802 are there any broad questions? 52 00:02:45,802 --> 00:02:47,010 ROMAIN: Nothing so far, Gary. 53 00:02:47,010 --> 00:02:48,177 GARY GENSLER: OK, thank you. 54 00:02:51,070 --> 00:02:55,480 So just give me a second to make sure 55 00:02:55,480 --> 00:03:01,585 I pull up the right set of slides. 56 00:03:08,031 --> 00:03:24,880 I think that-- all right, so two readings today. 57 00:03:24,880 --> 00:03:27,980 I don't know if you were able to take a look at them. 58 00:03:27,980 --> 00:03:31,240 I hope you had the time to take a look at them. 59 00:03:31,240 --> 00:03:33,400 And these were really just a chance 60 00:03:33,400 --> 00:03:38,710 to sort of grab hold of a broad discussion of what's going on. 61 00:03:38,710 --> 00:03:41,080 Again, I went to the Financial Stability Board. 62 00:03:41,080 --> 00:03:44,220 It's a little dated because it's 2017, 63 00:03:44,220 --> 00:03:46,220 but I thought the executive summaries 64 00:03:46,220 --> 00:03:48,530 and these various sections were helpful. 65 00:03:48,530 --> 00:03:52,750 And then a shorter medium post really on six examples, 66 00:03:52,750 --> 00:03:55,210 and today, we're going to talk about 67 00:03:55,210 --> 00:03:58,780 what is artificial intelligence, what is machine learning 68 00:03:58,780 --> 00:04:03,820 and deep learning, and then what are the eight or 10 major areas 69 00:04:03,820 --> 00:04:06,505 in finance that it's being used. 70 00:04:06,505 --> 00:04:07,880 And then next Monday, we're going 71 00:04:07,880 --> 00:04:11,350 to talk about a lot of the challenges 72 00:04:11,350 --> 00:04:16,450 and go a little deeper with regard to this. 73 00:04:16,450 --> 00:04:18,910 Now, I said that I had study questions. 74 00:04:18,910 --> 00:04:21,149 We're going to see if we can get this to work, 75 00:04:21,149 --> 00:04:23,620 and I'm going to ask for volunteers to speak up. 76 00:04:23,620 --> 00:04:27,230 If not, Romain might just unmute everybody, 77 00:04:27,230 --> 00:04:31,690 and then we'll see who I can cold call on or something. 78 00:04:31,690 --> 00:04:34,430 But who would like to take a crack? 79 00:04:34,430 --> 00:04:38,140 And remember, we're all pass emergency and no credit 80 00:04:38,140 --> 00:04:40,180 emergency, so this is just about trying 81 00:04:40,180 --> 00:04:42,340 to get the conversation going. 82 00:04:42,340 --> 00:04:44,410 But who would want to answer the question, what 83 00:04:44,410 --> 00:04:47,300 is artificial intelligence, machine learning, deep 84 00:04:47,300 --> 00:04:47,800 learning? 85 00:04:47,800 --> 00:04:50,290 And you don't need to be a computer scientist, 86 00:04:50,290 --> 00:04:52,570 but these terms are really important 87 00:04:52,570 --> 00:04:55,450 if you're going to be an entrepreneur and do a startup, 88 00:04:55,450 --> 00:04:58,030 or if you're going to be in a big incumbent or big tech 89 00:04:58,030 --> 00:05:01,690 company, just to have this conceptual framework 90 00:05:01,690 --> 00:05:05,963 and understanding of artificial intelligence, machine 91 00:05:05,963 --> 00:05:07,130 learning, and deep learning. 92 00:05:07,130 --> 00:05:10,690 So I'm going to wait for Romain to either find 93 00:05:10,690 --> 00:05:14,080 somebody who's raised their blue hand, or, Romain, you 94 00:05:14,080 --> 00:05:17,440 get to help me cold call if you wish. 95 00:05:17,440 --> 00:05:21,970 But hopefully, somebody wants to just dive into this. 96 00:05:21,970 --> 00:05:24,550 ROMAIN: I'm still waiting let's see who will be-- ah, 97 00:05:24,550 --> 00:05:26,200 we have our first volunteer of the day. 98 00:05:26,200 --> 00:05:28,493 Thank you, Michael. 99 00:05:28,493 --> 00:05:29,410 GARY GENSLER: Michael. 100 00:05:32,530 --> 00:05:35,140 STUDENT: Sorry, I forgot to unmute. 101 00:05:35,140 --> 00:05:39,970 So my understanding-- artificial intelligence just is more 102 00:05:39,970 --> 00:05:43,980 of an over encompassing term, being-- 103 00:05:43,980 --> 00:05:47,130 just computer is kind of mimicking human behavior 104 00:05:47,130 --> 00:05:50,327 and thought, so that's more-- 105 00:05:50,327 --> 00:05:51,660 GARY GENSLER: Let's pause there. 106 00:05:51,660 --> 00:05:55,540 And a very good answer, so computers 107 00:05:55,540 --> 00:05:57,490 mimicking human behavior. 108 00:05:57,490 --> 00:05:58,750 And, Michael, just a sense-- 109 00:05:58,750 --> 00:06:01,000 do you have a sense of how-- when did this come about? 110 00:06:01,000 --> 00:06:04,720 Was this in the last five years, or was it 111 00:06:04,720 --> 00:06:08,050 a longer time ago that somebody came up with this term 112 00:06:08,050 --> 00:06:10,000 "artificial intelligence?" 113 00:06:10,000 --> 00:06:14,210 STUDENT: More like the early to mid 1900s. 114 00:06:14,210 --> 00:06:15,340 It's been a while. 115 00:06:15,340 --> 00:06:16,798 GARY GENSLER: So it's been a while. 116 00:06:16,798 --> 00:06:18,820 It's actually way back in the 1950s, 117 00:06:18,820 --> 00:06:21,610 "artificial intelligence--" the concept 118 00:06:21,610 --> 00:06:24,580 of a computer mimicking humans. 119 00:06:24,580 --> 00:06:29,140 In fact, at MIT, we have the Computer Science and Artificial 120 00:06:29,140 --> 00:06:31,210 Intelligence Lab. 121 00:06:31,210 --> 00:06:34,950 It's not a creature that was just invented 122 00:06:34,950 --> 00:06:36,940 in the 20-teens or 20-naughts. 123 00:06:36,940 --> 00:06:38,680 It goes back decades. 124 00:06:38,680 --> 00:06:40,760 It was a merger of two earlier labs, 125 00:06:40,760 --> 00:06:43,420 but we've had an artificial intelligence lab at MIT 126 00:06:43,420 --> 00:06:46,810 for multiple decades. 127 00:06:46,810 --> 00:06:50,880 So who wants to say what machine learning is? 128 00:06:50,880 --> 00:06:51,500 Romain? 129 00:06:51,500 --> 00:06:53,018 I hand it off you to find another-- 130 00:06:53,018 --> 00:06:55,060 ROMAIN: Who will be the next volunteer for today? 131 00:06:58,870 --> 00:07:00,670 STUDENT: I think the machine [INAUDIBLE],, 132 00:07:00,670 --> 00:07:05,860 there's very limited or even no intervention with human, 133 00:07:05,860 --> 00:07:09,370 and learning are you solve problems step 134 00:07:09,370 --> 00:07:11,680 by step in a sequential manner. 135 00:07:11,680 --> 00:07:13,960 So that is like the machine solve 136 00:07:13,960 --> 00:07:18,185 problems step by step without much intervention by humans. 137 00:07:18,185 --> 00:07:19,810 GARY GENSLER: So I think you're raising 138 00:07:19,810 --> 00:07:22,320 two points, is the machine solving 139 00:07:22,320 --> 00:07:25,510 a problem without intervention, which is good. 140 00:07:25,510 --> 00:07:27,550 And then you said briefly, learning. 141 00:07:27,550 --> 00:07:29,020 Do you want to say something more, 142 00:07:29,020 --> 00:07:30,728 or anybody want to say, what does it mean 143 00:07:30,728 --> 00:07:32,455 that the machine is learning? 144 00:07:39,090 --> 00:07:41,892 ROMAIN: We have Rakan who raised his hand. 145 00:07:41,892 --> 00:07:43,350 STUDENT: Yes, well, essentially, it 146 00:07:43,350 --> 00:07:48,330 means that you're feeding the machine data, 147 00:07:48,330 --> 00:07:50,415 and as you feed the machine data, 148 00:07:50,415 --> 00:07:53,370 the machine learns not to do specific tasks. 149 00:07:53,370 --> 00:07:55,770 And you get better results as you feed 150 00:07:55,770 --> 00:07:58,050 it more data and more and more. 151 00:07:58,050 --> 00:08:00,870 GARY GENSLER: All right, so the concept of machine learning, 152 00:08:00,870 --> 00:08:02,670 again, is not that new. 153 00:08:02,670 --> 00:08:07,200 It was first written about in the 1980s and 1990s. 154 00:08:07,200 --> 00:08:08,740 It had a little bit of a wave-- 155 00:08:08,740 --> 00:08:12,590 you might call it a boomlet, a little bit of a hype cycle-- 156 00:08:12,590 --> 00:08:14,660 and then it sort of tamped down. 157 00:08:14,660 --> 00:08:20,170 But the conceptual framework is that the computer, machines, 158 00:08:20,170 --> 00:08:25,250 with data, are actually learning, that they actually-- 159 00:08:25,250 --> 00:08:29,270 whatever algorithms or decision making or pattern recognition 160 00:08:29,270 --> 00:08:31,730 that they have gets better. 161 00:08:31,730 --> 00:08:35,490 So machine learning is a subset of artificial intelligence. 162 00:08:35,490 --> 00:08:37,520 Artificial intelligence is this concept 163 00:08:37,520 --> 00:08:41,380 of computers, a form of machines, 164 00:08:41,380 --> 00:08:45,790 mimicking human intelligence, but it could mimic it 165 00:08:45,790 --> 00:08:46,870 without learning. 166 00:08:46,870 --> 00:08:51,460 It could just replicate or automate what we are doing. 167 00:08:51,460 --> 00:08:54,960 Machine learning was a new concept-- 168 00:08:54,960 --> 00:08:56,040 it didn't take off. 169 00:08:56,040 --> 00:08:59,250 It didn't dramatically change our lives at first-- 170 00:08:59,250 --> 00:09:02,670 where, actually, the computers could adapt 171 00:09:02,670 --> 00:09:06,960 and change based upon analysis of the data, 172 00:09:06,960 --> 00:09:09,900 and that their algorithms and their pattern recognition 173 00:09:09,900 --> 00:09:11,830 could shift. 174 00:09:11,830 --> 00:09:14,430 Does anybody want to take a crack at deep learning? 175 00:09:20,830 --> 00:09:22,112 ROMAIN: Anyone? 176 00:09:22,112 --> 00:09:23,070 Albert raised his hand. 177 00:09:25,590 --> 00:09:29,670 STUDENT: So deep learning involves 178 00:09:29,670 --> 00:09:34,170 using large neural networks to perform machine learning, 179 00:09:34,170 --> 00:09:36,120 so it's sort of a subset of machine learning. 180 00:09:36,120 --> 00:09:39,127 But it can be very powerful, and most of the time, 181 00:09:39,127 --> 00:09:41,460 it just gets better and better as you feed it more data. 182 00:09:41,460 --> 00:09:42,835 Other machine learning algorithms 183 00:09:42,835 --> 00:09:46,680 tend to have sort of a plateau and can't improve no matter 184 00:09:46,680 --> 00:09:48,690 how much data you feed them. 185 00:09:48,690 --> 00:09:53,700 GARY GENSLER: So important concept there in a phrase 186 00:09:53,700 --> 00:09:57,510 was "neural networks." 187 00:09:57,510 --> 00:09:58,890 Think of our brain's-- 188 00:09:58,890 --> 00:10:01,290 neural, neurology. 189 00:10:01,290 --> 00:10:04,380 It's about our brains. 190 00:10:04,380 --> 00:10:06,360 Early computer scientists started 191 00:10:06,360 --> 00:10:09,780 to think can we learn something from how the human brain 192 00:10:09,780 --> 00:10:16,560 works, which is in essence, a biological computer that takes 193 00:10:16,560 --> 00:10:21,120 electrical pulses and stores data, analyzes data, 194 00:10:21,120 --> 00:10:23,760 recognizes patterns. 195 00:10:23,760 --> 00:10:25,830 Even all of us right now are recognizing 196 00:10:25,830 --> 00:10:27,960 voice and visual patterns. 197 00:10:27,960 --> 00:10:29,960 That's in our human brain. 198 00:10:29,960 --> 00:10:33,000 So when looking at the brain, the conceptual framework 199 00:10:33,000 --> 00:10:36,000 is could we build a network similar to the brain, 200 00:10:36,000 --> 00:10:40,380 and thus, using these words, neural networks. 201 00:10:40,380 --> 00:10:43,560 In deep learning and machine learning, 202 00:10:43,560 --> 00:10:45,930 there's pattern recognition, but deep learning 203 00:10:45,930 --> 00:10:48,990 is a subset of machine learning that has 204 00:10:48,990 --> 00:10:53,440 multiple layers of connections. 205 00:10:53,440 --> 00:10:58,950 And in machine learning, you can take a base layer of data 206 00:10:58,950 --> 00:11:02,190 and try to find patterns, but deep learning 207 00:11:02,190 --> 00:11:04,290 finds patterns in the patterns. 208 00:11:04,290 --> 00:11:08,760 And you can think of it as putting it through layers. 209 00:11:08,760 --> 00:11:12,120 Now, if you're in Sloan, and you're deeply involved 210 00:11:12,120 --> 00:11:17,340 in computer science, and you also enjoy the topic, 211 00:11:17,340 --> 00:11:18,780 you can go much further. 212 00:11:18,780 --> 00:11:22,530 But in this class, we're not trying to go there. 213 00:11:22,530 --> 00:11:28,530 The importance of deep learning is that it can find and extract 214 00:11:28,530 --> 00:11:32,490 patterns even better than machine learning, 215 00:11:32,490 --> 00:11:36,960 but it takes more computational power and often more data. 216 00:11:36,960 --> 00:11:41,460 Deep learning is more an innovation of the 20-teens, 217 00:11:41,460 --> 00:11:46,390 and by 2011 and 2012, it really started to change things. 218 00:11:46,390 --> 00:11:48,570 And in the last eight years, we've 219 00:11:48,570 --> 00:11:51,590 seen dramatic advancements even in deep learning. 220 00:11:51,590 --> 00:11:56,070 It's a conceptual framework of taking a pool of data, 221 00:11:56,070 --> 00:11:59,630 looking at its patterns, and going a little higher. 222 00:11:59,630 --> 00:12:01,760 I'm going to talk about an example a little bit 223 00:12:01,760 --> 00:12:04,850 later in this discussion, but please 224 00:12:04,850 --> 00:12:11,760 bring me back to talk about deep learning in facial recognition, 225 00:12:11,760 --> 00:12:14,730 deep learning in autonomous vehicles 226 00:12:14,730 --> 00:12:16,230 and just thinking about it there, 227 00:12:16,230 --> 00:12:18,000 and then we'll pull it back to finance. 228 00:12:18,000 --> 00:12:18,992 ROMAIN: Gary? 229 00:12:18,992 --> 00:12:20,200 GARY GENSLER: Yes, questions? 230 00:12:20,200 --> 00:12:23,050 ROMAIN: We have two questions-- one from Rosana, who's asking, 231 00:12:23,050 --> 00:12:25,230 what is the difference between representation 232 00:12:25,230 --> 00:12:27,150 learning and deep learning? 233 00:12:27,150 --> 00:12:29,430 And then we'll give the floor to Pablo. 234 00:12:29,430 --> 00:12:30,930 GARY GENSLER: So very good question. 235 00:12:30,930 --> 00:12:33,510 Representation learning, you can think almost as 236 00:12:33,510 --> 00:12:35,580 in between machine learning and deep learning. 237 00:12:35,580 --> 00:12:41,010 If we said, this is almost like those Russian dolls. 238 00:12:41,010 --> 00:12:43,770 The AI is the big vessel. 239 00:12:43,770 --> 00:12:46,890 Machine learning is a subset, and deep learning 240 00:12:46,890 --> 00:12:48,420 was a subset of machine learning. 241 00:12:48,420 --> 00:12:52,320 Representational learning is a subset, in this context, 242 00:12:52,320 --> 00:12:58,560 of machine learning, and it's basically extracting features. 243 00:12:58,560 --> 00:13:01,830 So think of machine learning that's looking for patterns. 244 00:13:01,830 --> 00:13:05,340 It's extracting features out of data. 245 00:13:05,340 --> 00:13:11,130 So photo recognition, our standard Facebook photo 246 00:13:11,130 --> 00:13:15,750 recognition that might recognize Kelly versus Romain 247 00:13:15,750 --> 00:13:19,910 is extracting certain features of Kelly's face. 248 00:13:19,910 --> 00:13:21,530 I'm sorry if I'm picking on you Kelly. 249 00:13:21,530 --> 00:13:24,610 You just happen to be on my screen. 250 00:13:24,610 --> 00:13:27,650 But it's extracting features, so some people 251 00:13:27,650 --> 00:13:31,820 call it feature learning or representational learning, 252 00:13:31,820 --> 00:13:36,570 but extracting features as opposed to specific data. 253 00:13:36,570 --> 00:13:38,390 And then deep learning is generally 254 00:13:38,390 --> 00:13:40,670 thought of as a subset even though there 255 00:13:40,670 --> 00:13:46,150 is some debate as to how you would categorize these. 256 00:13:46,150 --> 00:13:48,920 And you said there was another question? 257 00:13:48,920 --> 00:13:49,920 STUDENT: Yeah, hi, Gary. 258 00:13:49,920 --> 00:13:52,070 This is Pablo. 259 00:13:52,070 --> 00:13:54,490 So I just wanted to verify, because my understanding 260 00:13:54,490 --> 00:13:57,220 is that, also, one of the big differences between machine 261 00:13:57,220 --> 00:13:59,920 learning and deep learning, or artificial intelligence 262 00:13:59,920 --> 00:14:01,390 in general besides deep learning, 263 00:14:01,390 --> 00:14:04,300 is that deep learning used unstructured data, 264 00:14:04,300 --> 00:14:06,610 whereas typically, for machine learning models, 265 00:14:06,610 --> 00:14:10,390 you need basically tables of data that you have organized 266 00:14:10,390 --> 00:14:12,645 and labeled and everything. 267 00:14:12,645 --> 00:14:14,020 And just the check whether that's 268 00:14:14,020 --> 00:14:18,640 the difference besides the additional complexity 269 00:14:18,640 --> 00:14:20,270 and multiple layers. 270 00:14:20,270 --> 00:14:23,940 GARY GENSLER: So let me step back and share 271 00:14:23,940 --> 00:14:26,620 the important vocabulary, and again this 272 00:14:26,620 --> 00:14:29,470 is important vocabulary well beyond being a computer 273 00:14:29,470 --> 00:14:30,150 scientist. 274 00:14:30,150 --> 00:14:33,400 It's important vocabulary if you're running a business, 275 00:14:33,400 --> 00:14:38,290 and you're trying to get the most out of data and your data 276 00:14:38,290 --> 00:14:39,710 analytics team. 277 00:14:39,710 --> 00:14:42,700 So I'm going to assume we're talking as if you are now going 278 00:14:42,700 --> 00:14:45,610 to be in a C-suite, and you want to get the most out 279 00:14:45,610 --> 00:14:48,340 of your data analytics teams and so forth-- 280 00:14:48,340 --> 00:14:53,620 is this concept of structured data versus unstructured data, 281 00:14:53,620 --> 00:14:57,490 and then I'll go to that specific question. 282 00:14:57,490 --> 00:15:06,360 Data that we see all the time with our eyes, that we read, 283 00:15:06,360 --> 00:15:08,200 that comes into us, you can think 284 00:15:08,200 --> 00:15:11,020 of as unstructured sometimes because it 285 00:15:11,020 --> 00:15:13,910 doesn't have a label on it. 286 00:15:13,910 --> 00:15:17,200 But if it has a label on it, all of a sudden, 287 00:15:17,200 --> 00:15:20,270 people call it structured data. 288 00:15:20,270 --> 00:15:22,300 So machine learning, conceptually, 289 00:15:22,300 --> 00:15:30,120 is that the machines are getting better at recognizing patterns. 290 00:15:30,120 --> 00:15:32,980 It's primarily about pattern recognition, 291 00:15:32,980 --> 00:15:37,990 that the machines are getting better recognizing patterns off 292 00:15:37,990 --> 00:15:39,110 of the data. 293 00:15:39,110 --> 00:15:41,410 And the question here is, is machine learning 294 00:15:41,410 --> 00:15:44,050 always structured learning? 295 00:15:44,050 --> 00:15:46,390 And structured learning means that the data 296 00:15:46,390 --> 00:15:49,510 is labeled, that you have a whole data set, 297 00:15:49,510 --> 00:15:52,420 and it's labeled. 298 00:15:52,420 --> 00:15:54,580 An example of labeling I will give you, 299 00:15:54,580 --> 00:15:58,720 that we all live with in our daily lives-- how many people-- 300 00:15:58,720 --> 00:16:01,390 just use the blue hands if you wish. 301 00:16:01,390 --> 00:16:06,190 How many people have ever been asked in a computer, 302 00:16:06,190 --> 00:16:08,830 will you please look at this picture 303 00:16:08,830 --> 00:16:13,210 and tell us whether there are any traffic 304 00:16:13,210 --> 00:16:14,243 lights in the picture? 305 00:16:14,243 --> 00:16:15,910 We want to make sure you're not a robot. 306 00:16:20,020 --> 00:16:23,680 All right, so here, I'm just going to-- 307 00:16:23,680 --> 00:16:26,290 Romain, just cold call somebody that put their hand up. 308 00:16:26,290 --> 00:16:28,510 I want to ask them a question. 309 00:16:28,510 --> 00:16:30,500 ROMAIN: Sophia, please? 310 00:16:30,500 --> 00:16:33,100 GARY GENSLER: All right, Sophia, are you are unmuted? 311 00:16:33,100 --> 00:16:34,160 STUDENT: Yes I'm on. 312 00:16:34,160 --> 00:16:36,327 GARY GENSLER: All right, so, Sophia, when you go in, 313 00:16:36,327 --> 00:16:38,270 and you tell the computer that you're 314 00:16:38,270 --> 00:16:43,160 not a machine, why do you think they ask you that question? 315 00:16:43,160 --> 00:16:45,380 STUDENT: To make sure that there aren't any bots who 316 00:16:45,380 --> 00:16:48,090 are trying to take advantage of any service 317 00:16:48,090 --> 00:16:50,840 that the machine is trying to offer. 318 00:16:50,840 --> 00:16:52,280 GARY GENSLER: Really good answer. 319 00:16:52,280 --> 00:16:55,550 You're correct, but you're not completely correct. 320 00:16:55,550 --> 00:16:57,200 What's the other reason that they're 321 00:16:57,200 --> 00:17:00,755 asking you whether that's a traffic light or not? 322 00:17:00,755 --> 00:17:02,130 STUDENT: And also to collect data 323 00:17:02,130 --> 00:17:04,270 so that they can use that data for labeling 324 00:17:04,270 --> 00:17:06,020 for future purposes as well. 325 00:17:06,020 --> 00:17:08,450 GARY GENSLER: So they are labeling data. 326 00:17:08,450 --> 00:17:14,310 They're using Sophia, if I might say, as free labor. 327 00:17:14,310 --> 00:17:20,940 Sophia is training the data that Google or whomever 328 00:17:20,940 --> 00:17:22,569 is putting together there. 329 00:17:22,569 --> 00:17:23,670 And thank you, Sofia. 330 00:17:23,670 --> 00:17:26,940 You're labeling that data so that autonomous vehicles 331 00:17:26,940 --> 00:17:28,680 will work better in the future. 332 00:17:28,680 --> 00:17:31,280 You're also frankly labeling data 333 00:17:31,280 --> 00:17:34,530 to put millions of people out of jobs as truckers, 334 00:17:34,530 --> 00:17:36,030 but I don't want you to get sort of 335 00:17:36,030 --> 00:17:41,040 wrapped up into those sort of social and public policy 336 00:17:41,040 --> 00:17:43,300 debates, but that's what's happening. 337 00:17:43,300 --> 00:17:45,270 So back to the earlier question. 338 00:17:45,270 --> 00:17:47,820 Data can be labeled by us. 339 00:17:47,820 --> 00:17:50,490 An earlier form of labeling was labeling 340 00:17:50,490 --> 00:17:53,220 what is an A, what is a B, what is a C, what 341 00:17:53,220 --> 00:17:56,280 are all the letters of the alphabet 342 00:17:56,280 --> 00:17:58,680 so that our postal services now can 343 00:17:58,680 --> 00:18:04,140 use a form of machine learning to read all of our written 344 00:18:04,140 --> 00:18:06,090 scratch on envelopes. 345 00:18:06,090 --> 00:18:08,520 If we address an envelope, it can all 346 00:18:08,520 --> 00:18:11,490 be read by computers rather than humans. 347 00:18:11,490 --> 00:18:12,750 To your earlier question-- 348 00:18:12,750 --> 00:18:16,230 I went a long way around this, but machine learning 349 00:18:16,230 --> 00:18:19,740 can be both unstructured and structured. 350 00:18:19,740 --> 00:18:23,100 The question was, is machine learning always labeled? 351 00:18:23,100 --> 00:18:25,290 Is machine learning always structured? 352 00:18:25,290 --> 00:18:26,880 And the answer's no. 353 00:18:26,880 --> 00:18:31,670 Machine learning can also be unstructured and unlabeled. 354 00:18:31,670 --> 00:18:34,640 Deep learning can be both labeled, which 355 00:18:34,640 --> 00:18:38,230 is structured, or unlabeled. 356 00:18:38,230 --> 00:18:41,350 Some of the economics and some of the computer science 357 00:18:41,350 --> 00:18:43,570 are worthwhile to understand. 358 00:18:43,570 --> 00:18:47,530 Labeled data can be trained faster. 359 00:18:47,530 --> 00:18:54,580 Label data can, in many regards, lower your error rates faster 360 00:18:54,580 --> 00:18:56,620 and have a certain-- 361 00:18:56,620 --> 00:19:02,800 extract correlations better, but it comes with a cost. 362 00:19:02,800 --> 00:19:05,170 You need to label the data, and so there's 363 00:19:05,170 --> 00:19:10,900 some tradeoff of getting Sophia to label data or other humans 364 00:19:10,900 --> 00:19:16,660 to label data versus unlabeled data. 365 00:19:16,660 --> 00:19:20,320 Think about radiology in the practice of medicine, 366 00:19:20,320 --> 00:19:25,030 and looking at body scans or mammograms 367 00:19:25,030 --> 00:19:28,960 or any form of radiology to identify 368 00:19:28,960 --> 00:19:30,910 whether there's an anomaly. 369 00:19:30,910 --> 00:19:34,330 There's something that needs to have further investigation 370 00:19:34,330 --> 00:19:38,890 to see whether it's a tumor or not. 371 00:19:38,890 --> 00:19:42,130 Radiology is dramatically changing in the last three 372 00:19:42,130 --> 00:19:46,040 or five years based upon machine learning and deep learning. 373 00:19:46,040 --> 00:19:47,470 Remember, deep learning just means 374 00:19:47,470 --> 00:19:50,640 there's multiple layers of pattern recognition 375 00:19:50,640 --> 00:19:52,810 in these neural networks. 376 00:19:52,810 --> 00:19:57,520 Labeled radiology, labeled mammograms, or labeled 377 00:19:57,520 --> 00:20:04,730 MRIs will train the machines faster, 378 00:20:04,730 --> 00:20:11,170 but unstructured, unlabeled data can also be used. 379 00:20:11,170 --> 00:20:13,930 You need bigger data sets. 380 00:20:13,930 --> 00:20:15,550 So I went I went off a little bit, 381 00:20:15,550 --> 00:20:17,050 but I hope that that's helpful. 382 00:20:19,580 --> 00:20:21,080 Other questions, Romain? 383 00:20:21,080 --> 00:20:22,910 ROMAIN: Yes, we have one from Victor. 384 00:20:22,910 --> 00:20:24,164 GARY GENSLER: Please. 385 00:20:24,164 --> 00:20:25,460 STUDENT: Hi, professor. 386 00:20:25,460 --> 00:20:27,746 I just wanted to double click on the-- 387 00:20:27,746 --> 00:20:30,530 take a step back in the initial definitions between machine 388 00:20:30,530 --> 00:20:32,780 learning and deep learning, because we were discussing 389 00:20:32,780 --> 00:20:35,960 that deep learning had the feature that it 390 00:20:35,960 --> 00:20:38,970 keeps learning despite the data growing exponentially. 391 00:20:38,970 --> 00:20:40,310 It doesn't plateau. 392 00:20:40,310 --> 00:20:44,900 But I don't fully understood the differentiation between two 393 00:20:44,900 --> 00:20:46,670 concepts beyond that. 394 00:20:46,670 --> 00:20:50,930 GARY GENSLER: So I didn't disagree or agree 395 00:20:50,930 --> 00:20:53,030 with that comment, and I apologize, 396 00:20:53,030 --> 00:20:55,790 I can't remember who said that deep learning keeps 397 00:20:55,790 --> 00:20:57,770 to grow exponentially. 398 00:20:57,770 --> 00:21:02,000 I think both machine learning and deep learning-- 399 00:21:02,000 --> 00:21:05,720 both machine learning and deep learning learn from data, 400 00:21:05,720 --> 00:21:10,943 and this word "learn" should be explored a little bit more. 401 00:21:10,943 --> 00:21:12,860 What machine learning and deep learning can do 402 00:21:12,860 --> 00:21:15,770 is extract correlations. 403 00:21:15,770 --> 00:21:18,200 I hope nearly everybody in this class 404 00:21:18,200 --> 00:21:20,150 has taken some form of statistics 405 00:21:20,150 --> 00:21:21,510 at some point in time. 406 00:21:21,510 --> 00:21:25,270 You might have hated statistics, but we all took it, 407 00:21:25,270 --> 00:21:27,530 and some of us took more advanced statistics 408 00:21:27,530 --> 00:21:30,420 where you use linear algebra and the like. 409 00:21:30,420 --> 00:21:34,450 But just thinking about a standard regression analysis-- 410 00:21:34,450 --> 00:21:38,420 a standard regression analysis finds a pattern, 411 00:21:38,420 --> 00:21:42,170 generally a linear pattern, or a quadratic pattern 412 00:21:42,170 --> 00:21:45,200 if you move on. 413 00:21:45,200 --> 00:21:47,810 Machine learning and deep learning find pattern, 414 00:21:47,810 --> 00:21:52,850 and they're really remarkable tools to extract correlations. 415 00:21:52,850 --> 00:21:55,370 And one of the features of both machine learning 416 00:21:55,370 --> 00:21:58,130 and deep learning is they look at error rates, 417 00:21:58,130 --> 00:22:03,410 particularly versus data sets that have been labeled. 418 00:22:03,410 --> 00:22:07,790 And traditionally, what you do is you have a big data set-- 419 00:22:07,790 --> 00:22:11,690 maybe it's millions of pieces of data that you're training on, 420 00:22:11,690 --> 00:22:14,300 and you take a random sample of it 421 00:22:14,300 --> 00:22:17,810 and put it to the side, a random sample on the side 422 00:22:17,810 --> 00:22:19,730 that you label. 423 00:22:19,730 --> 00:22:22,580 And then you compare what comes out 424 00:22:22,580 --> 00:22:27,740 of the machine learning with the test data on the side 425 00:22:27,740 --> 00:22:29,240 and see what's the error rate. 426 00:22:29,240 --> 00:22:32,960 And this labeled set on this side might say, these are men, 427 00:22:32,960 --> 00:22:35,600 these are women, this is a stoplight, 428 00:22:35,600 --> 00:22:40,700 this is a traffic light, whatever the labeled data is, 429 00:22:40,700 --> 00:22:45,620 and you see the predictive model, what's the error rate. 430 00:22:45,620 --> 00:22:48,020 Both machine learning and deep learning 431 00:22:48,020 --> 00:22:52,130 continue to do quite well. 432 00:22:52,130 --> 00:22:55,080 And I apologize I cannot remember who said it earlier, 433 00:22:55,080 --> 00:22:59,390 which student said, deep learning continues to grow 434 00:22:59,390 --> 00:23:01,830 further than machine learning. 435 00:23:01,830 --> 00:23:05,720 It can, but I wouldn't accept that machine learning can't get 436 00:23:05,720 --> 00:23:08,180 better and lower error rates. 437 00:23:08,180 --> 00:23:10,730 Now, once you get down to very low error rates, 438 00:23:10,730 --> 00:23:15,645 that's another circumstance altogether. 439 00:23:15,645 --> 00:23:18,020 The difference between deep learning and machine learning 440 00:23:18,020 --> 00:23:20,540 is that-- 441 00:23:20,540 --> 00:23:26,340 I'm going to use photo recognition software. 442 00:23:26,340 --> 00:23:29,850 If you put a photograph into a computer, where does it start? 443 00:23:29,850 --> 00:23:34,070 Does anybody-- what does it see at the very beginning, 444 00:23:34,070 --> 00:23:35,650 it's base level of data? 445 00:23:39,050 --> 00:23:41,550 STUDENT: The pixel? 446 00:23:41,550 --> 00:23:43,470 GARY GENSLER: What did I hear? 447 00:23:43,470 --> 00:23:44,720 STUDENT: I said, the pixels. 448 00:23:44,720 --> 00:23:45,950 GARY GENSLER: Pixels. 449 00:23:45,950 --> 00:23:48,700 So the only thing a computer can read is pixels. 450 00:23:48,700 --> 00:23:51,840 It has to start with the pixels and build up, 451 00:23:51,840 --> 00:23:54,400 and the next layer at most-- 452 00:23:54,400 --> 00:23:58,120 and again, I once pretended to know something 453 00:23:58,120 --> 00:23:59,420 about computer science. 454 00:23:59,420 --> 00:24:03,820 But I programmed in Fortran and APL years ago, 455 00:24:03,820 --> 00:24:07,440 and that was before many of you were born. 456 00:24:07,440 --> 00:24:11,860 But I guess I used to know how to program something. 457 00:24:11,860 --> 00:24:14,320 But if you read the pixels, the next layer 458 00:24:14,320 --> 00:24:17,080 up to find a pattern in the pixels 459 00:24:17,080 --> 00:24:20,850 is just small changes of shade, and then 460 00:24:20,850 --> 00:24:23,700 you can think of the next layer up from those little-- 461 00:24:23,700 --> 00:24:26,130 you can see edges. 462 00:24:26,130 --> 00:24:30,060 So the computer has to sort of go through layers 463 00:24:30,060 --> 00:24:34,620 from the pixels up to this is a traffic light versus a stop 464 00:24:34,620 --> 00:24:40,920 sign, and so deep learning interposes multiple layers 465 00:24:40,920 --> 00:24:44,000 of pattern recognition. 466 00:24:44,000 --> 00:24:46,430 Some would say that you need many layers, 467 00:24:46,430 --> 00:24:48,590 and then other research shows that, no, 468 00:24:48,590 --> 00:24:50,950 once you get to about three layers, 469 00:24:50,950 --> 00:24:55,250 there's less and less return on this. 470 00:24:55,250 --> 00:24:56,570 So let me sort of-- 471 00:24:56,570 --> 00:24:59,285 unless, Romain, is there other questions, or can I-- 472 00:24:59,285 --> 00:25:00,410 ROMAIN: No, we're all good. 473 00:25:00,410 --> 00:25:03,650 GARY GENSLER: All right, so again, this 474 00:25:03,650 --> 00:25:05,150 is just a broad thing, and we're not 475 00:25:05,150 --> 00:25:06,410 going to spend as much time. 476 00:25:06,410 --> 00:25:09,020 But what's natural language processing? 477 00:25:09,020 --> 00:25:11,550 Does anybody want to-- 478 00:25:11,550 --> 00:25:14,390 just what are these words broadly mean? 479 00:25:20,242 --> 00:25:21,200 ROMAIN: Any volunteers? 480 00:25:26,920 --> 00:25:29,500 GARY GENSLER: I'll volunteer, then. 481 00:25:29,500 --> 00:25:33,650 So natural language processing is just 482 00:25:33,650 --> 00:25:39,550 simply taking human language, natural language, 483 00:25:39,550 --> 00:25:43,630 and processing it down to computer language, 484 00:25:43,630 --> 00:25:47,170 all the way down to machine readable code, 485 00:25:47,170 --> 00:25:50,050 or going the other direction. 486 00:25:50,050 --> 00:25:53,490 So you can almost think of it as input and output 487 00:25:53,490 --> 00:25:59,080 to the computer, or we have many, many languages 488 00:25:59,080 --> 00:26:05,170 represented on this call right here with 99 participants. 489 00:26:05,170 --> 00:26:06,910 You can think about it as translating 490 00:26:06,910 --> 00:26:09,640 French to German and German to French, 491 00:26:09,640 --> 00:26:13,060 but instead it's natural language, what 492 00:26:13,060 --> 00:26:16,420 we do, down to the computer. 493 00:26:16,420 --> 00:26:19,680 And this is really important in terms of user interface 494 00:26:19,680 --> 00:26:25,050 and user experiences, and we'll get to that a little bit more. 495 00:26:25,050 --> 00:26:27,300 And then we're going to talk a lot about which sectors 496 00:26:27,300 --> 00:26:30,240 in financial services are being most affected 497 00:26:30,240 --> 00:26:31,220 at this point in time. 498 00:26:31,220 --> 00:26:35,100 So I won't call on the class right now because I want 499 00:26:35,100 --> 00:26:38,260 to keep moving to go forward. 500 00:26:38,260 --> 00:26:41,490 So we're in to talk about the financial world and fintech, 501 00:26:41,490 --> 00:26:43,480 and then-- oops, I didn't change this. 502 00:26:43,480 --> 00:26:46,680 This slide will-- I'll shift, because that's 503 00:26:46,680 --> 00:26:48,840 from the other day. 504 00:26:48,840 --> 00:26:51,360 So we looked at this slide the other day, 505 00:26:51,360 --> 00:26:53,100 and it just helps us-- 506 00:26:53,100 --> 00:26:54,960 what is AI machine learning? 507 00:26:54,960 --> 00:26:58,590 Extracting useful patterns from the data, 508 00:26:58,590 --> 00:27:00,870 using neural networks that we talked about, 509 00:27:00,870 --> 00:27:04,040 optimizing to lower error rates. 510 00:27:04,040 --> 00:27:09,930 Optimizing so that you actually say with 99% or 99 1/2% 511 00:27:09,930 --> 00:27:13,800 of the time, this is a traffic light, this is a stop sign. 512 00:27:13,800 --> 00:27:15,810 Actually optimizing. 513 00:27:15,810 --> 00:27:18,750 There's lots of programs that you can use. 514 00:27:18,750 --> 00:27:20,940 Google has TensorFlow, and-- 515 00:27:20,940 --> 00:27:23,070 I don't know. 516 00:27:23,070 --> 00:27:25,260 Have many of you ever used TensorFlow? 517 00:27:25,260 --> 00:27:27,210 I don't know all of your backgrounds, 518 00:27:27,210 --> 00:27:32,197 but any show blue hands? 519 00:27:32,197 --> 00:27:34,030 I'm not going to call on you to describe it. 520 00:27:34,030 --> 00:27:36,280 I'm just kind of curious if there's many people that 521 00:27:36,280 --> 00:27:40,240 have used TensorFlow or not. 522 00:27:40,240 --> 00:27:42,556 ROMAIN: How about we go with Devin? 523 00:27:42,556 --> 00:27:44,380 GARY GENSLER: So Devin's actually used it. 524 00:27:44,380 --> 00:27:45,880 I wasn't going to pick on them up, 525 00:27:45,880 --> 00:27:48,993 but if you wanted to say anything about it, Devin. 526 00:27:48,993 --> 00:27:50,410 STUDENT: Yeah, I can very briefly. 527 00:27:50,410 --> 00:27:54,310 So it essentially gives like a plug-and-play method 528 00:27:54,310 --> 00:27:58,858 to do machine learning and build neural nets in Python. 529 00:27:58,858 --> 00:28:01,150 You don't necessarily have to have a full understanding 530 00:28:01,150 --> 00:28:03,070 of how under the hood works. 531 00:28:03,070 --> 00:28:05,200 You can just add bits as you want, take bits away 532 00:28:05,200 --> 00:28:08,133 as you want, and it speeds up the whole process. 533 00:28:08,133 --> 00:28:10,300 GARY GENSLER: And so what's important about that is, 534 00:28:10,300 --> 00:28:15,040 just as somebody that came of age in the 1970s 535 00:28:15,040 --> 00:28:18,250 and '80s didn't have to learn how to computer code all 536 00:28:18,250 --> 00:28:20,800 the way down to machine readable code, 537 00:28:20,800 --> 00:28:25,000 they could learn how to use-- 538 00:28:25,000 --> 00:28:30,170 by the 1990s, C++, or C And C++, and later, Python. 539 00:28:30,170 --> 00:28:34,640 And many people in this class know how to use Python. 540 00:28:34,640 --> 00:28:36,740 In the machine learning area, there's 541 00:28:36,740 --> 00:28:41,360 been plug and play programs like TensorFlow, 542 00:28:41,360 --> 00:28:43,250 where you don't need to actually know 543 00:28:43,250 --> 00:28:47,290 how to build the data and things like that. 544 00:28:47,290 --> 00:28:50,080 Most importantly, and this is if you're 545 00:28:50,080 --> 00:28:53,380 thinking about being a data analyst 546 00:28:53,380 --> 00:28:55,930 or actually building a business around it, 547 00:28:55,930 --> 00:29:00,730 it's the data and the questions you train on the data. 548 00:29:00,730 --> 00:29:03,790 And most studies have shown, as of 2019, 549 00:29:03,790 --> 00:29:08,560 that 90%, 95% of the cost of data analytics 550 00:29:08,560 --> 00:29:11,930 is what some people might call cleaning up the data, 551 00:29:11,930 --> 00:29:13,660 making sure the data's well labeled. 552 00:29:13,660 --> 00:29:16,480 We talked about structured versus unstructured data 553 00:29:16,480 --> 00:29:17,440 earlier. 554 00:29:17,440 --> 00:29:21,070 Really important about that labeling, the cost. 555 00:29:21,070 --> 00:29:26,430 If you have 1,000 people working in a machine learning shop, 556 00:29:26,430 --> 00:29:30,420 or 500 people or five, it is quite likely 557 00:29:30,420 --> 00:29:36,480 that a big bulk of their time is standardizing the data, so 558 00:29:36,480 --> 00:29:39,840 to speak cleaning up the data, making sure that the fields are 559 00:29:39,840 --> 00:29:46,610 filled, and ensuring that you can then train on this data, 560 00:29:46,610 --> 00:29:51,507 meaning it's labeled, or enough of it's labeled. 561 00:29:51,507 --> 00:29:53,840 And then thinking about what the questions you're really 562 00:29:53,840 --> 00:29:57,230 trying to achieve, what you're trying to extract. 563 00:29:57,230 --> 00:29:58,490 Why is it happening now? 564 00:29:58,490 --> 00:30:00,980 This is off of Lex Freeman's slide again, but why? 565 00:30:00,980 --> 00:30:04,520 Because the hardware, the tools, the analytics-- a lot 566 00:30:04,520 --> 00:30:08,960 has shifted in the last five or eight years. 567 00:30:08,960 --> 00:30:12,710 To give you a sense of what it's being used for, all of these, 568 00:30:12,710 --> 00:30:13,430 we know. 569 00:30:13,430 --> 00:30:14,540 We know already. 570 00:30:14,540 --> 00:30:18,090 It's dramatically changing our lives. 571 00:30:18,090 --> 00:30:20,940 When we're sitting at home, sheltering at a home, 572 00:30:20,940 --> 00:30:22,680 and you're thinking about the next movie, 573 00:30:22,680 --> 00:30:25,050 and you're on Netflix, Netflix is telling us 574 00:30:25,050 --> 00:30:27,900 what they think the next thing we should watch. 575 00:30:27,900 --> 00:30:31,170 That's training off of not just the knowledge 576 00:30:31,170 --> 00:30:33,750 of what each of us has been watching, 577 00:30:33,750 --> 00:30:37,080 but it's about what others are watching. 578 00:30:37,080 --> 00:30:39,270 It's the Postal Service that no longer 579 00:30:39,270 --> 00:30:43,260 has to have a human reading the text, 580 00:30:43,260 --> 00:30:46,020 our scrawl on the envelope. 581 00:30:46,020 --> 00:30:50,070 It's Facebook with the facial recognition programs and so 582 00:30:50,070 --> 00:30:55,380 forth, and autonomous vehicles that 583 00:30:55,380 --> 00:30:57,570 are now being tested on the roads 584 00:30:57,570 --> 00:31:00,240 but are very likely part of our future. 585 00:31:00,240 --> 00:31:04,800 Now, will they be rolled out in a dramatic way in five years, 586 00:31:04,800 --> 00:31:06,360 or will it be 15 years? 587 00:31:06,360 --> 00:31:08,100 But I would feel comfortable that we 588 00:31:08,100 --> 00:31:11,670 will have autonomous vehicles on the road 589 00:31:11,670 --> 00:31:16,080 sometime at least by the 2030, but maybe others 590 00:31:16,080 --> 00:31:17,610 would be more optimistic. 591 00:31:17,610 --> 00:31:21,540 So it's changing a lot in many, many fields. 592 00:31:21,540 --> 00:31:26,370 The question is now, how is it shifting this field of finance? 593 00:31:26,370 --> 00:31:32,295 Why do I put it at the center of what we're doing here? 594 00:31:35,500 --> 00:31:37,065 So we talked about this. 595 00:31:37,065 --> 00:31:39,570 This was just my little attempt, and so now you 596 00:31:39,570 --> 00:31:43,080 have a slide that does what we chatted about before. 597 00:31:45,690 --> 00:31:49,895 One important thing also is happening is, in finance-- 598 00:31:49,895 --> 00:31:51,270 [CLEARING HIS THROAT] excuse me-- 599 00:31:51,270 --> 00:31:55,680 people are grabbing alternative data, using alternative data. 600 00:31:55,680 --> 00:31:59,490 I said earlier that the most important questions 601 00:31:59,490 --> 00:32:02,340 are what's the good data? 602 00:32:02,340 --> 00:32:06,870 So then we think about, in finance, what type of data 603 00:32:06,870 --> 00:32:09,480 do we want to grab? 604 00:32:09,480 --> 00:32:12,810 Data analytics in finance goes back centuries. 605 00:32:12,810 --> 00:32:14,730 That is not new. 606 00:32:14,730 --> 00:32:17,040 The Medicis, when they had to figure out 607 00:32:17,040 --> 00:32:21,300 to whom to land in Renaissance era 608 00:32:21,300 --> 00:32:24,840 had to figure out who is a good credit or not. 609 00:32:24,840 --> 00:32:28,110 And two data scientists from Stanford started a company 610 00:32:28,110 --> 00:32:29,790 called Fair Isaac-- 611 00:32:29,790 --> 00:32:32,940 those were their last names, Fair and Isaac-- 612 00:32:32,940 --> 00:32:36,030 and that became the FICO company, the Fair Isaac 613 00:32:36,030 --> 00:32:37,670 Company. 614 00:32:37,670 --> 00:32:41,530 So the data analytics in finance and the consumer side 615 00:32:41,530 --> 00:32:45,540 has certainly been around since the 1950s and 1960s, 616 00:32:45,540 --> 00:32:47,470 but where we are now is to say what 617 00:32:47,470 --> 00:32:50,380 is the additional types of data that we might take, 618 00:32:50,380 --> 00:32:54,970 and not just banking and checking and so forth? 619 00:32:54,970 --> 00:32:59,540 But Alibaba can look at a company, look at a company 620 00:32:59,540 --> 00:33:03,440 very closely, and do a full cash flow underwriting. 621 00:33:03,440 --> 00:33:06,740 Alibaba, because they have AliPay, 622 00:33:06,740 --> 00:33:10,550 can see what that small business is spending 623 00:33:10,550 --> 00:33:12,500 and what that small business is receiving. 624 00:33:12,500 --> 00:33:15,440 Amazon Prime can't quite do it as much, 625 00:33:15,440 --> 00:33:18,570 but they can do a bit of it as well. 626 00:33:18,570 --> 00:33:21,150 And even Toast, which is a fintech company 627 00:33:21,150 --> 00:33:25,440 in the restaurant business until this Corona shut down, 628 00:33:25,440 --> 00:33:27,600 Toast could see a lot about what's 629 00:33:27,600 --> 00:33:31,590 happening restaurant by restaurant in their cash flows. 630 00:33:31,590 --> 00:33:37,600 They had the revenue side more than the expenditure side. 631 00:33:37,600 --> 00:33:41,700 Alibaba, much better data sets than Toast, 632 00:33:41,700 --> 00:33:47,840 but I wouldn't put at rest a Toast started earlier 633 00:33:47,840 --> 00:33:53,120 in the last year to do credit extension to the restaurant 634 00:33:53,120 --> 00:33:54,470 business. 635 00:33:54,470 --> 00:33:56,990 Now, do you need deep learning and machine learning 636 00:33:56,990 --> 00:33:57,680 to do that? 637 00:33:57,680 --> 00:33:58,940 Not necessarily. 638 00:33:58,940 --> 00:34:01,790 You can still use plain old regression 639 00:34:01,790 --> 00:34:07,160 and linear statistics, linear regression analysis, 640 00:34:07,160 --> 00:34:12,975 but machine learning and deep learning help you go further. 641 00:34:12,975 --> 00:34:14,600 And then there's, of course, everything 642 00:34:14,600 --> 00:34:18,350 about our usage, our browser history, our email receipts, 643 00:34:18,350 --> 00:34:19,429 and so forth. 644 00:34:19,429 --> 00:34:22,310 If we look at China, they've stood up 645 00:34:22,310 --> 00:34:28,670 a broader social credit system, and in that system, 646 00:34:28,670 --> 00:34:32,989 they can tap into data about users 647 00:34:32,989 --> 00:34:34,370 in many different platforms. 648 00:34:34,370 --> 00:34:36,460 Romain, do I see-- are you waving at me? 649 00:34:36,460 --> 00:34:38,406 Is there a question? 650 00:34:38,406 --> 00:34:39,239 ROMAIN: No, I'm not. 651 00:34:39,239 --> 00:34:40,440 Sorry for that. 652 00:34:40,440 --> 00:34:41,732 GARY GENSLER: That's all right. 653 00:34:43,510 --> 00:34:47,080 So natural language processing, I mentioned. 654 00:34:47,080 --> 00:34:50,139 I just want to say a few more words about it. 655 00:34:50,139 --> 00:34:55,120 Think of it as computers input and output interpretation. 656 00:34:55,120 --> 00:34:58,430 This sort of going from German to French, 657 00:34:58,430 --> 00:35:03,760 or going from computer language to human language and back 658 00:35:03,760 --> 00:35:04,780 again. 659 00:35:04,780 --> 00:35:08,710 That's this important back and forth, 660 00:35:08,710 --> 00:35:10,960 and so it's natural language understanding, 661 00:35:10,960 --> 00:35:14,500 meaning a computer understands something, and also 662 00:35:14,500 --> 00:35:17,570 natural language generation. 663 00:35:17,570 --> 00:35:20,260 So it can be audio, image, video, 664 00:35:20,260 --> 00:35:24,590 any form of communication-- even a gesture. 665 00:35:24,590 --> 00:35:27,480 This hand wave can be interpreted-- 666 00:35:27,480 --> 00:35:32,220 if not now in 2020, within a few years will be interpreted. 667 00:35:32,220 --> 00:35:37,210 A movement of your face will be interpreted as well. 668 00:35:39,840 --> 00:35:44,340 And so how it's being used is really quite interesting, 669 00:35:44,340 --> 00:35:48,740 but we all know about chat bots and voice assistance already. 670 00:35:48,740 --> 00:35:52,250 That's shifting our worlds. 671 00:35:52,250 --> 00:35:52,790 Yeah? 672 00:35:52,790 --> 00:35:54,650 ROMAIN: We have a question from Nadia. 673 00:35:54,650 --> 00:35:56,600 GARY GENSLER: Nadia, please. 674 00:35:56,600 --> 00:35:59,030 STUDENT: I have one question related to the chat bots. 675 00:35:59,030 --> 00:36:00,470 What kind of factors do you think 676 00:36:00,470 --> 00:36:02,300 will encourage people to use chat bots? 677 00:36:02,300 --> 00:36:04,910 Because now, I do think people prefer 678 00:36:04,910 --> 00:36:09,280 to talk to a person rather than chat bots. 679 00:36:09,280 --> 00:36:12,700 GARY GENSLER: Well, I think, Nadia, 680 00:36:12,700 --> 00:36:15,280 we might still prefer to talk to a person, 681 00:36:15,280 --> 00:36:20,950 but there's a certain efficiency in-- 682 00:36:20,950 --> 00:36:23,860 and I'm just going to stay in finance for a minute. 683 00:36:23,860 --> 00:36:29,750 But there's a certain efficiency that financial service firms 684 00:36:29,750 --> 00:36:34,010 find that they can use chat bots instead of putting 685 00:36:34,010 --> 00:36:35,250 a human on the phone. 686 00:36:35,250 --> 00:36:38,930 So even when you and I call up to a Bank of America, 687 00:36:38,930 --> 00:36:43,800 and we want to check in on something on our credit cards, 688 00:36:43,800 --> 00:36:47,720 we're put through a various series of push one 689 00:36:47,720 --> 00:36:50,150 if you want this, push two if you want that, 690 00:36:50,150 --> 00:36:51,870 and we're pushing buttons. 691 00:36:51,870 --> 00:36:55,610 That's not high technology, by the way, 692 00:36:55,610 --> 00:36:58,820 but that's an efficiency that Bank of America 693 00:36:58,820 --> 00:37:02,060 has interposed into the system instead of having 694 00:37:02,060 --> 00:37:04,640 a call center of humans. 695 00:37:04,640 --> 00:37:08,750 And so if they can move from a cost center of humans 696 00:37:08,750 --> 00:37:14,050 to an automated call center of chat bots, 697 00:37:14,050 --> 00:37:19,150 they can provide services at a lower cost and to more people. 698 00:37:19,150 --> 00:37:22,930 Now, you and I might still want a human on the other side, 699 00:37:22,930 --> 00:37:28,710 but business is interposing an automation, 700 00:37:28,710 --> 00:37:36,050 and that automation means, often, quicker response time. 701 00:37:36,050 --> 00:37:39,890 So many of us, you go into a website today, 702 00:37:39,890 --> 00:37:44,870 and there's a little bot window that comes up. 703 00:37:44,870 --> 00:37:49,940 And the first thing that comes up on so many websites-- 704 00:37:49,940 --> 00:37:52,460 and this is true if it's a financial site. 705 00:37:52,460 --> 00:37:57,290 It's true if it's a commercial website where 706 00:37:57,290 --> 00:37:58,850 you're buying something online. 707 00:37:58,850 --> 00:38:00,920 It's probably true of dating websites, 708 00:38:00,920 --> 00:38:03,410 that there's some little bot window that comes up and says, 709 00:38:03,410 --> 00:38:05,060 can we help you. 710 00:38:05,060 --> 00:38:09,620 That's not a human, but it does give us greater service. 711 00:38:09,620 --> 00:38:12,920 It gives us an immediate recognition somebody's 712 00:38:12,920 --> 00:38:15,280 answering a question. 713 00:38:15,280 --> 00:38:17,530 Nadia, do I sense you'd prefer not to have 714 00:38:17,530 --> 00:38:18,970 the chat bots interposed? 715 00:38:26,410 --> 00:38:28,000 Is Nadia still there? 716 00:38:28,000 --> 00:38:29,760 STUDENT: Oh, yeah. 717 00:38:29,760 --> 00:38:31,810 Yeah, because I do think sometimes, 718 00:38:31,810 --> 00:38:34,420 chat bots, we ask a question, but their answer 719 00:38:34,420 --> 00:38:38,320 is not really related to our questions. 720 00:38:38,320 --> 00:38:40,720 GARY GENSLER: So you would prefer a human 721 00:38:40,720 --> 00:38:44,950 because you think the human will interpret your question 722 00:38:44,950 --> 00:38:46,920 and be able to answer it better? 723 00:38:46,920 --> 00:38:48,840 STUDENT: Yes. 724 00:38:48,840 --> 00:38:52,290 GARY GENSLER: So if the chat bot could answer as well 725 00:38:52,290 --> 00:38:53,700 as Kelly could answer-- 726 00:38:53,700 --> 00:38:54,840 again, I'm sorry, Kelly. 727 00:38:54,840 --> 00:38:56,280 You're on my screen. 728 00:38:56,280 --> 00:39:01,380 But if the chat bot could answer as well as Kelly or Camillo 729 00:39:01,380 --> 00:39:03,480 or others in this class could answer, 730 00:39:03,480 --> 00:39:05,590 you'd be all right with that? 731 00:39:05,590 --> 00:39:08,213 STUDENT: Yeah, it's faster. 732 00:39:08,213 --> 00:39:09,630 GARY GENSLER: See, if it's faster, 733 00:39:09,630 --> 00:39:11,320 and it can answer as well. 734 00:39:11,320 --> 00:39:15,084 So those are some of the commercial challenges. 735 00:39:15,084 --> 00:39:18,230 ROMAIN: I think Ivy would like to contribute as well. 736 00:39:18,230 --> 00:39:20,340 GARY GENSLER: Sure. 737 00:39:20,340 --> 00:39:24,660 STUDENT: Yeah, I just wanted to offer a little bit of some 738 00:39:24,660 --> 00:39:26,490 of the consumer studies that we did 739 00:39:26,490 --> 00:39:28,610 when I was working for a startup where 740 00:39:28,610 --> 00:39:30,180 we were building chat bot. 741 00:39:30,180 --> 00:39:34,230 And interesting enough, I think most people are actually 742 00:39:34,230 --> 00:39:34,740 pretty-- 743 00:39:34,740 --> 00:39:37,410 and we did a pretty large survey, 744 00:39:37,410 --> 00:39:40,490 and most people were pretty open to the idea 745 00:39:40,490 --> 00:39:42,240 of working with a chat bot because I think 746 00:39:42,240 --> 00:39:44,310 that's become so pervasive. 747 00:39:44,310 --> 00:39:47,040 But then there's this idea, they want 748 00:39:47,040 --> 00:39:50,490 to know that it is a chat bot, that the company is very 749 00:39:50,490 --> 00:39:52,890 transparent about that, because people change 750 00:39:52,890 --> 00:39:55,110 their behavior when they are speaking to the chat 751 00:39:55,110 --> 00:39:58,110 bot or some kind of virtual assistant, 752 00:39:58,110 --> 00:40:00,340 as long as they know, just to build that trust. 753 00:40:00,340 --> 00:40:02,370 And also, I guess we try to be more 754 00:40:02,370 --> 00:40:05,190 explicit in our wording, both verbally 755 00:40:05,190 --> 00:40:07,500 as well as written texts. 756 00:40:07,500 --> 00:40:09,660 And then secondly, I think because-- 757 00:40:09,660 --> 00:40:12,900 I guess I wanted to pose this as a question, too. 758 00:40:12,900 --> 00:40:16,300 When I think about chat bots and things like that, 759 00:40:16,300 --> 00:40:20,540 the technology is not necessarily there, 760 00:40:20,540 --> 00:40:22,920 or it takes a long time. 761 00:40:22,920 --> 00:40:28,110 And so I'm just curious what your thoughts are in terms of-- 762 00:40:28,110 --> 00:40:35,360 for me, I see it as like AI gets us 80% of the way there, 763 00:40:35,360 --> 00:40:38,210 but we need the human touch 20% of the way there. 764 00:40:38,210 --> 00:40:41,280 And so I actually see a lot of companies either having 765 00:40:41,280 --> 00:40:43,030 that human at the end of the-- 766 00:40:43,030 --> 00:40:46,260 you ultimately still need a human at the end of the day. 767 00:40:46,260 --> 00:40:48,825 So I mean, I just wanted to explore that a little. 768 00:40:48,825 --> 00:40:51,950 GARY GENSLER: I think what Nadia and Ivy are raising, 769 00:40:51,950 --> 00:40:53,900 and I'm sure we're all grappling with this. 770 00:40:53,900 --> 00:40:57,480 We are living in a very exciting time, 771 00:40:57,480 --> 00:41:01,460 and I'm not talking about this corona crisis. 772 00:41:01,460 --> 00:41:04,580 That's a different type of challenge. 773 00:41:04,580 --> 00:41:06,320 But we're living in an exciting time 774 00:41:06,320 --> 00:41:10,190 where we can automate a lot of things that humans have done. 775 00:41:10,190 --> 00:41:12,230 We've automated so many things that humans 776 00:41:12,230 --> 00:41:14,420 have done for centuries, but we're now 777 00:41:14,420 --> 00:41:19,070 automating this interface, through chat 778 00:41:19,070 --> 00:41:24,310 bots and conversational interfaces, voice assistance. 779 00:41:24,310 --> 00:41:28,550 I would dare say that, of the 90-plus people in this class 780 00:41:28,550 --> 00:41:32,570 now, that most of us, if not all of us, at some point in time, 781 00:41:32,570 --> 00:41:34,540 have used Siri. 782 00:41:34,540 --> 00:41:36,460 I mean, if we're driving along a highway, 783 00:41:36,460 --> 00:41:38,800 and we're supposed to be hands free, 784 00:41:38,800 --> 00:41:45,210 we might talk and start up an app or something legally, 785 00:41:45,210 --> 00:41:46,240 legally. 786 00:41:46,240 --> 00:41:51,270 And there's a lot of automation that's going on. 787 00:41:51,270 --> 00:41:56,280 How many of us have called to arrange a reservation 788 00:41:56,280 --> 00:41:58,140 at a restaurant, and we're not quite sure 789 00:41:58,140 --> 00:42:04,050 if we're talking to a human or a conversational agent? 790 00:42:04,050 --> 00:42:05,940 But I think that what Ivy's saying 791 00:42:05,940 --> 00:42:12,690 is that there might always need to be a human somewhere there. 792 00:42:12,690 --> 00:42:15,630 I don't know, Ivy, if that's correct. 793 00:42:15,630 --> 00:42:18,390 That's where we are in 2020. 794 00:42:18,390 --> 00:42:20,310 Let's think about autonomous vehicles. 795 00:42:20,310 --> 00:42:22,760 Right now, we're not comfortable enough. 796 00:42:22,760 --> 00:42:25,080 The manufacturers aren't comfortable enough. 797 00:42:25,080 --> 00:42:27,840 The computer scientist aren't comfortable enough. 798 00:42:27,840 --> 00:42:29,550 The regulators aren't comfortable enough. 799 00:42:29,550 --> 00:42:31,080 The public's not comfortable enough 800 00:42:31,080 --> 00:42:35,910 to have autonomous vehicles on the road with no humans 801 00:42:35,910 --> 00:42:41,280 whatsoever, but that's not really necessarily 802 00:42:41,280 --> 00:42:43,710 where we'll be in 2030. 803 00:42:43,710 --> 00:42:45,090 Or take radiology. 804 00:42:45,090 --> 00:42:48,930 Right now, at least in advanced economies 805 00:42:48,930 --> 00:42:51,840 like in Europe and the US and elsewhere, 806 00:42:51,840 --> 00:42:53,580 in advanced economies, we say we still 807 00:42:53,580 --> 00:42:57,100 want a doctor's eyes on a radiologist report. 808 00:42:57,100 --> 00:43:01,920 So the mammogram might be read by some artificial intelligence 809 00:43:01,920 --> 00:43:07,380 machine learning trained data, but we still have a human. 810 00:43:07,380 --> 00:43:11,370 But is that really the tradeoff we'll make in a few years? 811 00:43:11,370 --> 00:43:13,140 And is it the right tradeoff to be 812 00:43:13,140 --> 00:43:16,140 made in less developed countries, where 813 00:43:16,140 --> 00:43:19,170 they don't have the resources to have the doctors? 814 00:43:19,170 --> 00:43:22,440 And now, we even look in the middle of this crisis, 815 00:43:22,440 --> 00:43:24,870 the corona crisis, if-- 816 00:43:24,870 --> 00:43:27,300 this is sort of God willing. 817 00:43:27,300 --> 00:43:31,380 If the Baidus of China and the Googles of the US 818 00:43:31,380 --> 00:43:35,280 and others sharp analytic AI shops 819 00:43:35,280 --> 00:43:39,540 come up with a way to extract patterns and develop 820 00:43:39,540 --> 00:43:43,770 some recognition as to who's most vulnerable, 821 00:43:43,770 --> 00:43:46,140 are we going to rely on that, or are we 822 00:43:46,140 --> 00:43:49,380 going to say a human has to also interpret it and be involved 823 00:43:49,380 --> 00:43:50,820 in it? 824 00:43:50,820 --> 00:43:52,410 And I don't know. 825 00:43:52,410 --> 00:43:54,270 So I think we're at an exciting time 826 00:43:54,270 --> 00:43:57,150 where we're automating more and more. 827 00:43:57,150 --> 00:43:59,250 I do agree with you, Ivy, there's always 828 00:43:59,250 --> 00:44:00,540 going to be a role for humans. 829 00:44:00,540 --> 00:44:06,360 I'm not terribly worried that we'll all be put out of a job. 830 00:44:06,360 --> 00:44:10,130 200 years ago, our ancestors, all of our ancestors, 831 00:44:10,130 --> 00:44:12,155 were, by and large, working on farms. 832 00:44:15,420 --> 00:44:16,950 That's the economies. 833 00:44:16,950 --> 00:44:20,340 And we have found other things to fill those roles 834 00:44:20,340 --> 00:44:21,930 and those needs. 835 00:44:21,930 --> 00:44:28,370 I think we'll still have the humans, but not in every task. 836 00:44:32,190 --> 00:44:34,200 So let me go through a little bit-- 837 00:44:34,200 --> 00:44:36,000 so the Financial Stability Board, this is 838 00:44:36,000 --> 00:44:38,520 their definitions I'm going to pass on, 839 00:44:38,520 --> 00:44:40,890 but this was in that paper that you all 840 00:44:40,890 --> 00:44:43,950 read about what big data and machine learning was. 841 00:44:43,950 --> 00:44:46,560 When I show this page to computer scientists, 842 00:44:46,560 --> 00:44:50,970 when I show it to colleagues of mine at MIT 843 00:44:50,970 --> 00:44:54,375 from the College of Computing, they look at it, 844 00:44:54,375 --> 00:44:56,250 and they say, jeez, that's funny that a bunch 845 00:44:56,250 --> 00:44:59,010 of financial treasury secretaries 846 00:44:59,010 --> 00:45:01,680 and central bankers and their staffs 847 00:45:01,680 --> 00:45:03,930 define big data machine learning this way. 848 00:45:03,930 --> 00:45:06,090 So I partly put it up because this 849 00:45:06,090 --> 00:45:10,170 is kind of what the regulators define as to what it is. 850 00:45:10,170 --> 00:45:12,810 Machine learning may be defined as a method of designing 851 00:45:12,810 --> 00:45:16,200 sequence of actions to solve a problem, known as algorithms, 852 00:45:16,200 --> 00:45:18,090 and so forth. 853 00:45:18,090 --> 00:45:22,440 Computer scientists would name it a little bit differently. 854 00:45:22,440 --> 00:45:24,230 So I said to you the other day that I 855 00:45:24,230 --> 00:45:28,220 think of financial technology as history is building 856 00:45:28,220 --> 00:45:32,570 on these things, but machine learning and deep learning 857 00:45:32,570 --> 00:45:34,480 is at this top level. 858 00:45:34,480 --> 00:45:37,620 In the customer interface, it's the chat box 859 00:45:37,620 --> 00:45:41,960 we were just talking about, and on the risk management side, 860 00:45:41,960 --> 00:45:45,420 it's extracting patterns to make better risk decisions. 861 00:45:45,420 --> 00:45:48,770 So it's in these two broad fields, 862 00:45:48,770 --> 00:45:54,040 I sort of think of it is the customer interface 863 00:45:54,040 --> 00:45:57,970 and then lowering risk and extracting patterns. 864 00:45:57,970 --> 00:45:59,800 And sometimes, it's not just lowering risk. 865 00:45:59,800 --> 00:46:02,020 It's enhancing returns. 866 00:46:02,020 --> 00:46:04,600 And so I was going to go through and chat 867 00:46:04,600 --> 00:46:09,700 about each of these eight areas, and not 868 00:46:09,700 --> 00:46:12,190 all of the areas we're going to talk about are as robust. 869 00:46:12,190 --> 00:46:15,390 Asset management, right now-- 870 00:46:15,390 --> 00:46:18,480 asset management from hedge funds all the way 871 00:46:18,480 --> 00:46:21,990 to the BlackRock and Fidelity are exploring the use 872 00:46:21,990 --> 00:46:24,530 of machine learning and AI. 873 00:46:24,530 --> 00:46:27,710 By and large, most high frequency trading shops, 874 00:46:27,710 --> 00:46:31,520 most hedge funds today, most asset managers today, 875 00:46:31,520 --> 00:46:35,740 are not using much machine learning and AI. 876 00:46:35,740 --> 00:46:40,150 I view that as an opportunity in the 2020s. 877 00:46:40,150 --> 00:46:45,640 I view that as a real possibility of a shift, a very 878 00:46:45,640 --> 00:46:46,630 significant shift. 879 00:46:46,630 --> 00:46:49,180 But where is it being used so far? 880 00:46:49,180 --> 00:46:51,730 So BlackRock has been announcing and saying 881 00:46:51,730 --> 00:46:53,320 that they're already using BlackRock 882 00:46:53,320 --> 00:46:55,960 as one of the world's, if not the world's, largest 883 00:46:55,960 --> 00:46:57,610 asset manager. 884 00:46:57,610 --> 00:47:00,820 Before this corona crisis, probably $6 or $7 trillion 885 00:47:00,820 --> 00:47:01,940 of assets. 886 00:47:01,940 --> 00:47:05,470 It's, of course, a lower number now. 887 00:47:05,470 --> 00:47:06,970 And BlackRock and others have been 888 00:47:06,970 --> 00:47:11,470 saying we're using machine learning to actually listen 889 00:47:11,470 --> 00:47:16,930 to all of the audio files, all of the audio files 890 00:47:16,930 --> 00:47:18,910 of the major companies when they announce 891 00:47:18,910 --> 00:47:22,080 their quarterly earnings. 892 00:47:22,080 --> 00:47:29,210 And they're also putting in news articles, digital news 893 00:47:29,210 --> 00:47:32,330 or articles about those announcements, 894 00:47:32,330 --> 00:47:37,790 and also feeding in some of the actual financial statements 895 00:47:37,790 --> 00:47:39,920 that are released. 896 00:47:39,920 --> 00:47:44,480 And that takes a little bit of natural language processing. 897 00:47:44,480 --> 00:47:47,780 You need some form of taking the audio files, 898 00:47:47,780 --> 00:47:52,580 taking the written files, and interpreting that. 899 00:47:52,580 --> 00:47:55,490 But with that data they're looking for sentiment. 900 00:47:55,490 --> 00:47:59,510 They're trying to interpret the sentiments 901 00:47:59,510 --> 00:48:01,550 and see if the stock price are moving 902 00:48:01,550 --> 00:48:05,700 based on all of that data. 903 00:48:05,700 --> 00:48:08,790 Now, that's BlackRock with $6 or $7 trillion of assets. 904 00:48:08,790 --> 00:48:11,650 They're deeply resourced, fidelity, and so forth. 905 00:48:11,650 --> 00:48:13,530 But if you go down the value chain, 906 00:48:13,530 --> 00:48:15,890 if you go down to smaller asset managers, 907 00:48:15,890 --> 00:48:21,920 they're not doing a lot, I would say, yet. 908 00:48:21,920 --> 00:48:24,530 But there are hedge funds that are specifically saying, 909 00:48:24,530 --> 00:48:26,720 we are data analytic hedge funds. 910 00:48:26,720 --> 00:48:28,130 We want to move a little further. 911 00:48:28,130 --> 00:48:31,130 We want to try to use this machine learning because it's 912 00:48:31,130 --> 00:48:34,225 a better way to extract correlations, 913 00:48:34,225 --> 00:48:35,600 and that's what it's looking for. 914 00:48:35,600 --> 00:48:37,580 It's pattern recognition. 915 00:48:37,580 --> 00:48:40,790 I think that you're going to see more and more high 916 00:48:40,790 --> 00:48:44,510 frequency trading shops and hedge funds exploring this. 917 00:48:44,510 --> 00:48:47,030 But one conversation I had in the last couple of months 918 00:48:47,030 --> 00:48:48,800 with the high frequency trading shop-- 919 00:48:48,800 --> 00:48:51,170 and it was a shop that had about 100 employees. 920 00:48:51,170 --> 00:48:54,100 It was not big, but it was big enough. 921 00:48:54,100 --> 00:48:57,770 It was certainly making money in those days. 922 00:48:57,770 --> 00:49:00,620 I don't know how it's doing now. 923 00:49:00,620 --> 00:49:04,430 But they said, look, we feel pretty good about what we do, 924 00:49:04,430 --> 00:49:09,500 and when we look for algorithms, when we look-- 925 00:49:09,500 --> 00:49:14,270 our algorithmic trading doesn't need all of that expenditure 926 00:49:14,270 --> 00:49:17,300 and all that resource intensive thing of machine 927 00:49:17,300 --> 00:49:21,180 learning and cleaning up the data and finding the patterns. 928 00:49:21,180 --> 00:49:26,930 And in fact, we think that is not flexible enough yet for us. 929 00:49:26,930 --> 00:49:30,950 We're looking for short term opportunities, 930 00:49:30,950 --> 00:49:35,110 and we think that regression analysis and our classic linear 931 00:49:35,110 --> 00:49:39,000 and correlation analyses are enough at this moment. 932 00:49:39,000 --> 00:49:41,480 What's going to come in 2023 is a different thing, 933 00:49:41,480 --> 00:49:45,290 but what they're doing now, [INAUDIBLE] not really needed. 934 00:49:45,290 --> 00:49:48,260 Questions about asset management just before I 935 00:49:48,260 --> 00:49:49,930 go to a couple other fields? 936 00:49:52,792 --> 00:49:55,250 ROMAIN: Now is the time to raise your hand if you have any. 937 00:49:57,760 --> 00:49:59,433 I don't see any, Gary. 938 00:49:59,433 --> 00:50:01,002 GARY GENSLER: So cost-- 939 00:50:01,002 --> 00:50:03,880 STUDENT: Sorry, Gary, I raised it in the last second, 940 00:50:03,880 --> 00:50:05,030 so Romain didn't see it. 941 00:50:05,030 --> 00:50:07,210 GARY GENSLER: [INAUDIBLE]. 942 00:50:07,210 --> 00:50:08,530 STUDENT: My question is-- 943 00:50:08,530 --> 00:50:11,320 so I fully understand their claims 944 00:50:11,320 --> 00:50:13,960 that this is not the claims of the asset manager, 945 00:50:13,960 --> 00:50:18,840 you were talking of high speed trading. 946 00:50:18,840 --> 00:50:22,320 I don't see why machine learning or any algorithm 947 00:50:22,320 --> 00:50:25,922 besides linear regression doesn't have enough flexibility 948 00:50:25,922 --> 00:50:27,630 to actually replicate what they're doing, 949 00:50:27,630 --> 00:50:31,860 but with a bit more of accuracy or computing 950 00:50:31,860 --> 00:50:35,250 performance, et cetera, because in the end, 951 00:50:35,250 --> 00:50:37,770 my understanding is that you need the same-- 952 00:50:37,770 --> 00:50:39,300 except when you go to deep learning. 953 00:50:39,300 --> 00:50:41,448 But you need the same data sources 954 00:50:41,448 --> 00:50:43,740 that you currently have, and the only thing that you're 955 00:50:43,740 --> 00:50:49,640 going to do is, instead of just having linear models predicting 956 00:50:49,640 --> 00:50:51,990 the different traits that you want to do, 957 00:50:51,990 --> 00:50:55,260 you have other models that can find 958 00:50:55,260 --> 00:50:58,330 alternative basically trends and et cetera. 959 00:50:58,330 --> 00:51:00,850 So I don't see the limitation there of using it. 960 00:51:00,850 --> 00:51:03,790 GARY GENSLER: So I think you raise a really good point. 961 00:51:03,790 --> 00:51:06,110 We're in a period of transition. 962 00:51:06,110 --> 00:51:09,570 I personally don't think that machine learning 963 00:51:09,570 --> 00:51:13,440 and deep learning is the answer to all these pattern 964 00:51:13,440 --> 00:51:16,350 recognition challenges in finance. 965 00:51:16,350 --> 00:51:23,280 But you will find I'm more to the sort of center maximalist 966 00:51:23,280 --> 00:51:25,480 than center minimalist, and those of you 967 00:51:25,480 --> 00:51:28,470 that know me, that when we talk about blockchain technology, 968 00:51:28,470 --> 00:51:31,740 you'll find I'm more to the center minimalist side. 969 00:51:31,740 --> 00:51:35,010 But what I mean by that is I think that the pattern 970 00:51:35,010 --> 00:51:37,900 recognition out of deep learning, machine learning-- 971 00:51:37,900 --> 00:51:41,340 and by pattern recognition I mean the ability to extract, 972 00:51:41,340 --> 00:51:47,810 with remarkable ability, correlations and then create 973 00:51:47,810 --> 00:51:51,650 certain decision sets based on those correlations-- 974 00:51:51,650 --> 00:51:56,660 is better than classic linear algebra, classic regression 975 00:51:56,660 --> 00:51:58,370 analysis. 976 00:51:58,370 --> 00:52:03,380 But it comes with a cost, and that's the tradeoff in 2020. 977 00:52:03,380 --> 00:52:06,190 Maybe in a handful of years, it'll be less cost, 978 00:52:06,190 --> 00:52:10,460 but I'm going to use an example, which 979 00:52:10,460 --> 00:52:14,060 is not about asset management, but it's about lending. 980 00:52:14,060 --> 00:52:18,740 I had this conversation just a few weeks ago with the CEO 981 00:52:18,740 --> 00:52:21,000 of a major peer to peer lending company, 982 00:52:21,000 --> 00:52:22,940 and as this is being recorded, I'm 983 00:52:22,940 --> 00:52:27,530 just going to maybe not say his name. 984 00:52:27,530 --> 00:52:31,820 And I said, do you use machine learning and deep 985 00:52:31,820 --> 00:52:36,410 learning to do your credit decisions, 986 00:52:36,410 --> 00:52:38,150 all your credit decisions? 987 00:52:38,150 --> 00:52:41,720 And he said, yes, we use a lot of alternative data. 988 00:52:41,720 --> 00:52:44,150 We run it into the decision sets, 989 00:52:44,150 --> 00:52:46,650 and we see what patterns emerge. 990 00:52:46,650 --> 00:52:49,040 And I said, so you extend credit then based on that? 991 00:52:49,040 --> 00:52:51,290 He said, well, not exactly. 992 00:52:51,290 --> 00:52:55,280 He said, what we do is, we look for patterns, 993 00:52:55,280 --> 00:52:57,920 and then when we find them, we then just 994 00:52:57,920 --> 00:53:04,300 use classic algebra and linear flagging when we actually 995 00:53:04,300 --> 00:53:05,770 extend the credit. 996 00:53:05,770 --> 00:53:09,927 So we use it to look for patterns, 997 00:53:09,927 --> 00:53:12,385 but then we sort of use the traditional way, and I ask why. 998 00:53:12,385 --> 00:53:13,843 He said, well, there's two reasons. 999 00:53:13,843 --> 00:53:17,800 It's less costly than running the whole thing all the time, 1000 00:53:17,800 --> 00:53:21,940 and two, they can explain it better to regulators, 1001 00:53:21,940 --> 00:53:24,850 and they can explain it better to the public. 1002 00:53:24,850 --> 00:53:26,890 And at least in consumer finance, 1003 00:53:26,890 --> 00:53:30,070 there were laws passed about 50 years ago-- in the US, 1004 00:53:30,070 --> 00:53:32,620 it's called the Fair Credit Reporting Act. 1005 00:53:32,620 --> 00:53:34,240 These laws were passed and said if you 1006 00:53:34,240 --> 00:53:36,610 deny somebody credit you have to be able to explain 1007 00:53:36,610 --> 00:53:39,430 why you're denying them credit. 1008 00:53:39,430 --> 00:53:41,950 And of course, there are other laws in many other countries 1009 00:53:41,950 --> 00:53:43,630 similar to that, and there are also 1010 00:53:43,630 --> 00:53:46,450 laws about avoiding biases. 1011 00:53:46,450 --> 00:53:49,720 In our country, we call it the Equal Credit Opportunity 1012 00:53:49,720 --> 00:53:52,030 Act or ECOA, and in other countries, 1013 00:53:52,030 --> 00:53:54,580 similar things about avoiding biases 1014 00:53:54,580 --> 00:54:00,440 for gender and race and background and the like. 1015 00:54:00,440 --> 00:54:04,420 So I'm just saying that, whether it's asset management, 1016 00:54:04,420 --> 00:54:06,280 whether it's consumer credit, there's 1017 00:54:06,280 --> 00:54:11,260 some tradeoffs of these new data analytic tools. 1018 00:54:11,260 --> 00:54:17,580 And those tradeoffs, I think, in the next handful of years, 1019 00:54:17,580 --> 00:54:22,800 will keep tipping in the way towards using deep learning. 1020 00:54:22,800 --> 00:54:23,910 But they don't come-- 1021 00:54:23,910 --> 00:54:28,300 they're not cost free is what I'm saying. 1022 00:54:28,300 --> 00:54:30,040 ROMAIN: José has his hand up. 1023 00:54:30,040 --> 00:54:31,390 GARY GENSLER: Please. 1024 00:54:31,390 --> 00:54:34,150 STUDENT: So you talked a bit about how 1025 00:54:34,150 --> 00:54:37,390 this is impacting the high frequency trading shops. 1026 00:54:37,390 --> 00:54:39,730 Is there something similar happening 1027 00:54:39,730 --> 00:54:43,450 on more like value investing long term or in the hedge 1028 00:54:43,450 --> 00:54:44,240 funds? 1029 00:54:44,240 --> 00:54:48,370 So I heard some of them are using satellite image 1030 00:54:48,370 --> 00:54:50,188 to predict the number of cars-- 1031 00:54:50,188 --> 00:54:51,730 to see the number of cars [INAUDIBLE] 1032 00:54:51,730 --> 00:54:54,200 and predict the sales for that year, things like that. 1033 00:54:54,200 --> 00:54:58,600 But I don't know, do you see a lot of headroom in this area? 1034 00:54:58,600 --> 00:55:02,680 GARY GENSLER: I think there is headroom where I'm saying, 1035 00:55:02,680 --> 00:55:06,100 I've mentioned two areas, but I think you're right. 1036 00:55:06,100 --> 00:55:07,150 There's a third area. 1037 00:55:07,150 --> 00:55:10,450 The two areas are sentiment analysis, just 1038 00:55:10,450 --> 00:55:15,460 sentiment analysis around either an individual company 1039 00:55:15,460 --> 00:55:17,800 or the overall markets and seeing 1040 00:55:17,800 --> 00:55:21,730 what the sentiment, the mood, the sense of the crowd 1041 00:55:21,730 --> 00:55:29,300 is off of words, images, or the like. 1042 00:55:29,300 --> 00:55:33,410 And then also the high frequency traders 1043 00:55:33,410 --> 00:55:35,950 and so forth just looking for the patterns in the short term 1044 00:55:35,950 --> 00:55:37,060 trading. 1045 00:55:37,060 --> 00:55:40,360 You're talking about more on the broader value orientation, 1046 00:55:40,360 --> 00:55:42,880 and I absolutely share your view that we'll 1047 00:55:42,880 --> 00:55:44,980 see more of that develop. 1048 00:55:44,980 --> 00:55:48,670 I haven't heard a lot of it, but you're right about sector 1049 00:55:48,670 --> 00:55:51,450 after sector that you might be able to analyze. 1050 00:55:51,450 --> 00:55:55,050 But again, it has to have some ability 1051 00:55:55,050 --> 00:56:01,000 to extract a pattern better than classic linear regressions 1052 00:56:01,000 --> 00:56:02,200 in analysis. 1053 00:56:02,200 --> 00:56:04,200 So let me just try to hit a couple more of these 1054 00:56:04,200 --> 00:56:05,825 because we're going to run out of time, 1055 00:56:05,825 --> 00:56:08,010 but we are going to talk Monday more about these. 1056 00:56:08,010 --> 00:56:11,040 We talked about call centers, chat bots, robo-advising, 1057 00:56:11,040 --> 00:56:13,710 and so forth, so I think you've got that. 1058 00:56:13,710 --> 00:56:16,770 A lot of that is not just efficiency, 1059 00:56:16,770 --> 00:56:22,150 but it's also inclusion. 1060 00:56:22,150 --> 00:56:24,940 You can cover many, many more people 1061 00:56:24,940 --> 00:56:27,760 by automating some of these tasks. 1062 00:56:27,760 --> 00:56:30,940 It's just reality, it's the tradeoff 1063 00:56:30,940 --> 00:56:38,620 of efficiency and inclusion, that you cover far more people. 1064 00:56:38,620 --> 00:56:40,660 You can also be more targeted. 1065 00:56:40,660 --> 00:56:44,280 You can be more targeted with advice and so forth 1066 00:56:44,280 --> 00:56:45,760 as these are automated. 1067 00:56:45,760 --> 00:56:48,500 It comes with the tradeoff that Ivy and Nadia 1068 00:56:48,500 --> 00:56:52,120 were talking about earlier. 1069 00:56:52,120 --> 00:56:54,880 Credit and insurance-- this is the concept 1070 00:56:54,880 --> 00:56:57,550 of basically how you allocate or extend 1071 00:56:57,550 --> 00:57:03,550 or price, either alone or price insurance. 1072 00:57:03,550 --> 00:57:06,960 To date, the insurance companies and insurance underwriting 1073 00:57:06,960 --> 00:57:09,510 are starting to grapple with this, starting to move on, 1074 00:57:09,510 --> 00:57:11,610 and we'll talk about some fintech companies 1075 00:57:11,610 --> 00:57:13,830 in this space. 1076 00:57:13,830 --> 00:57:17,760 I think that insurance companies have been a little bit slower 1077 00:57:17,760 --> 00:57:20,250 to do it than, let's say, the credit card 1078 00:57:20,250 --> 00:57:23,340 companies, but the allocation-- 1079 00:57:23,340 --> 00:57:26,940 I think this will be dramatically shifting. 1080 00:57:26,940 --> 00:57:29,480 I think if you look at what's going on, again, 1081 00:57:29,480 --> 00:57:32,970 in consumer credit and small business credit in China, 1082 00:57:32,970 --> 00:57:35,430 through WeChat Pay and AliPay, they're 1083 00:57:35,430 --> 00:57:40,640 much further along than we are, frankly, here in the US. 1084 00:57:40,640 --> 00:57:47,190 We're still largely reliant on a 30 or 40-year-old architecture 1085 00:57:47,190 --> 00:57:50,790 around the Fair Isaac Company, the FICO scores 1086 00:57:50,790 --> 00:57:54,510 that are used in about 30 countries around the globe. 1087 00:57:54,510 --> 00:57:58,500 These are still quite limited, but FICO itself, 1088 00:57:58,500 --> 00:58:01,590 FICO itself rolls out new versions of FICO 1089 00:58:01,590 --> 00:58:02,760 every few years. 1090 00:58:02,760 --> 00:58:07,650 I think they're rolling out FICO 10.0 this summer, 1091 00:58:07,650 --> 00:58:10,270 or were before the corona crisis. 1092 00:58:10,270 --> 00:58:13,290 I think, if you look at the end of the 2020s, 1093 00:58:13,290 --> 00:58:16,650 either FICO will not exist at all, 1094 00:58:16,650 --> 00:58:21,600 or FICO 14.0 or 15.0 will look a lot more 1095 00:58:21,600 --> 00:58:26,250 like a machine learning, deep learning type of model. 1096 00:58:26,250 --> 00:58:29,280 What it is right now is pretty rudimentary 1097 00:58:29,280 --> 00:58:34,490 compared to what it could be in 5 or 10 years. 1098 00:58:34,490 --> 00:58:36,890 This is an area that's being used a lot-- 1099 00:58:36,890 --> 00:58:38,850 fraud detection and prevention. 1100 00:58:38,850 --> 00:58:40,940 The credit card companies, if you 1101 00:58:40,940 --> 00:58:46,640 look at Cap 1 and Bank of America, Discover, American 1102 00:58:46,640 --> 00:58:50,690 Express, they're deeply now using machine learning tools 1103 00:58:50,690 --> 00:58:52,520 for fraud detection. 1104 00:58:52,520 --> 00:58:57,230 Many of us probably remember just a year or two ago, 1105 00:58:57,230 --> 00:58:59,360 you would still call up your credit card company, 1106 00:58:59,360 --> 00:59:01,610 and you would say I'm traveling to France, 1107 00:59:01,610 --> 00:59:04,430 I'm traveling to Italy, wherever I'm traveling. 1108 00:59:04,430 --> 00:59:06,350 I want to put a flag on there that I 1109 00:59:06,350 --> 00:59:09,320 might be using my cell phone-- 1110 00:59:09,320 --> 00:59:13,470 using my credit card in one of these countries. 1111 00:59:13,470 --> 00:59:15,560 Well, most companies now don't ask you 1112 00:59:15,560 --> 00:59:17,270 to put a travel alert on it. 1113 00:59:17,270 --> 00:59:18,800 Now, part of that is because they 1114 00:59:18,800 --> 00:59:21,230 know where we're traveling because we're walking around 1115 00:59:21,230 --> 00:59:24,870 with these location devices. 1116 00:59:24,870 --> 00:59:28,285 The banks no longer need us to call them to tell them 1117 00:59:28,285 --> 00:59:29,910 we're going to be in Paris because they 1118 00:59:29,910 --> 00:59:31,130 know we're in Paris-- 1119 00:59:31,130 --> 00:59:36,300 this location tracking company device. 1120 00:59:36,300 --> 00:59:40,410 But in addition to that location device, 1121 00:59:40,410 --> 00:59:42,360 they are also using machine learning 1122 00:59:42,360 --> 00:59:45,570 to do fraud detection and prevention in the credit card 1123 00:59:45,570 --> 00:59:47,910 space. 1124 00:59:47,910 --> 00:59:50,390 Similarly, they're using it to track 1125 00:59:50,390 --> 00:59:53,000 and try to comply with laws called anti money laundering. 1126 00:59:53,000 --> 00:59:56,520 These two areas, fraud detection and any money laundering, 1127 00:59:56,520 --> 00:59:59,630 which I might call compliance broadly, 1128 00:59:59,630 --> 01:00:03,860 are two of the areas most developed right now in 2020. 1129 01:00:03,860 --> 01:00:06,380 That doesn't mean they'll be the most developed later. 1130 01:00:06,380 --> 01:00:09,560 I think a lot more will happen in the underwriting space. 1131 01:00:09,560 --> 01:00:11,945 A lot more will happen in the asset management space. 1132 01:00:15,350 --> 01:00:17,780 Robotic process automation-- I want 1133 01:00:17,780 --> 01:00:19,490 to just pause for a minute. 1134 01:00:19,490 --> 01:00:23,060 Does anybody have a sense of what these three words together 1135 01:00:23,060 --> 01:00:26,290 mean-- robotic process automation? 1136 01:00:26,290 --> 01:00:29,905 Romain, you get to see if there's any blue hands up. 1137 01:00:29,905 --> 01:00:32,001 ROMAIN: Andrea. 1138 01:00:32,001 --> 01:00:33,830 STUDENT: Hi. 1139 01:00:33,830 --> 01:00:36,600 So robotic process automation is very simple. 1140 01:00:36,600 --> 01:00:39,280 You have a lot of manual processes 1141 01:00:39,280 --> 01:00:41,080 or manual work done in, for example, 1142 01:00:41,080 --> 01:00:43,020 back office of the banks. 1143 01:00:43,020 --> 01:00:46,030 And the idea here is, instead of people 1144 01:00:46,030 --> 01:00:48,550 doing that, low skilled work or workforce, 1145 01:00:48,550 --> 01:00:53,540 you can actually teach robots or the algorithms with the PC 1146 01:00:53,540 --> 01:00:55,330 to do it instead of you. 1147 01:00:55,330 --> 01:00:56,890 So for example, it can be anything 1148 01:00:56,890 --> 01:00:59,050 as simple as just going through the forms 1149 01:00:59,050 --> 01:01:03,490 and copying or overwriting and rewriting 1150 01:01:03,490 --> 01:01:09,257 some of the words or parts of the forms to some other place. 1151 01:01:09,257 --> 01:01:10,090 GARY GENSLER: Right. 1152 01:01:10,090 --> 01:01:14,320 So robotic process automation can be as simple 1153 01:01:14,320 --> 01:01:19,330 as you're giving your permission to a startup 1154 01:01:19,330 --> 01:01:26,320 company, a fintech company, to access your bank account. 1155 01:01:26,320 --> 01:01:31,000 And one of us gives a fintech company-- maybe Credit Karma. 1156 01:01:31,000 --> 01:01:33,400 We give Credit Karma the right to go in and look 1157 01:01:33,400 --> 01:01:35,020 at a bank account of ours. 1158 01:01:35,020 --> 01:01:39,040 Credit Karma might not have permission from Bank of America 1159 01:01:39,040 --> 01:01:40,490 go in, but they have my-- 1160 01:01:40,490 --> 01:01:41,880 I'm permissioned them. 1161 01:01:41,880 --> 01:01:46,777 I've given them my password and my user ID, and they go in, 1162 01:01:46,777 --> 01:01:47,860 and they want to automate. 1163 01:01:47,860 --> 01:01:49,360 Credit Karma wants to automate. 1164 01:01:49,360 --> 01:01:52,150 They don't want to have a human actually have to type that all 1165 01:01:52,150 --> 01:01:52,680 in. 1166 01:01:52,680 --> 01:01:56,880 It can be as simple as just automating inserting the user 1167 01:01:56,880 --> 01:01:59,830 name and the password and the like, 1168 01:01:59,830 --> 01:02:03,250 but then you go further than it can navigate the web page. 1169 01:02:03,250 --> 01:02:05,830 It can navigate and click the right buttons 1170 01:02:05,830 --> 01:02:08,270 and get the right data and so forth. 1171 01:02:08,270 --> 01:02:10,090 So robotic process automation can 1172 01:02:10,090 --> 01:02:16,170 be helping a startup company say Gary Gensler, 1173 01:02:16,170 --> 01:02:18,780 we want you as a client of Credit Karma. 1174 01:02:18,780 --> 01:02:20,640 We, Credit Karma, will figure out 1175 01:02:20,640 --> 01:02:23,310 how to interface with Bank of America 1176 01:02:23,310 --> 01:02:26,670 and with Chase and the others. 1177 01:02:26,670 --> 01:02:30,870 But also, the banks are using robotic process automation 1178 01:02:30,870 --> 01:02:36,030 to automate so much of their both back office and their data 1179 01:02:36,030 --> 01:02:37,170 entry. 1180 01:02:37,170 --> 01:02:40,470 And one form, just to say, is many 1181 01:02:40,470 --> 01:02:43,320 of us now feel very comfortable to deposit 1182 01:02:43,320 --> 01:02:45,810 a check from our cell phone. 1183 01:02:45,810 --> 01:02:49,560 So you can take your cell phone, take a picture of a check, 1184 01:02:49,560 --> 01:02:51,750 and somehow, that gives an instruction, 1185 01:02:51,750 --> 01:02:56,550 a digital instruction, to move money. 1186 01:02:56,550 --> 01:02:59,580 Well, part of that's natural language processing, 1187 01:02:59,580 --> 01:03:03,880 that the cell phone could take a picture, read all that data, 1188 01:03:03,880 --> 01:03:07,920 put it into computer language, and actually move a digital. 1189 01:03:07,920 --> 01:03:10,650 Part of that is robotic process automation. 1190 01:03:10,650 --> 01:03:11,910 Romain, was there a question? 1191 01:03:11,910 --> 01:03:14,298 I think I saw some flashing chat rooms. 1192 01:03:14,298 --> 01:03:15,090 ROMAIN: Yes, sorry. 1193 01:03:15,090 --> 01:03:16,820 You have 15 minutes left. 1194 01:03:16,820 --> 01:03:17,890 GARY GENSLER: Oh, OK. 1195 01:03:17,890 --> 01:03:20,710 Then was there a question or no? 1196 01:03:20,710 --> 01:03:22,340 ROMAIN: No question at this point. 1197 01:03:22,340 --> 01:03:23,632 GARY GENSLER: So now, trading-- 1198 01:03:23,632 --> 01:03:27,280 trading is an area I spent a lot of time with Goldman Sachs. 1199 01:03:27,280 --> 01:03:30,640 And in that trading of the day, we 1200 01:03:30,640 --> 01:03:32,680 were automating everything we could automate, 1201 01:03:32,680 --> 01:03:35,820 and this is in the 1990s. 1202 01:03:35,820 --> 01:03:38,010 And ever since, anything is-- 1203 01:03:38,010 --> 01:03:41,220 a trading floor in 2020 looks very different than a trading 1204 01:03:41,220 --> 01:03:45,630 floor in the 1990s in terms of the day to day trading, 1205 01:03:45,630 --> 01:03:49,290 and this is trading at the center of the markets, 1206 01:03:49,290 --> 01:03:55,860 the platforms themselves, and of course, the high frequency 1207 01:03:55,860 --> 01:03:58,020 traders on the other end of the market 1208 01:03:58,020 --> 01:04:00,390 is basically just like asset management. 1209 01:04:00,390 --> 01:04:02,010 What patterns can you see? 1210 01:04:02,010 --> 01:04:04,200 Now, this is less about value investing. 1211 01:04:04,200 --> 01:04:09,300 This is the patterns right in the nanoseconds 1212 01:04:09,300 --> 01:04:11,640 and milliseconds and so forth. 1213 01:04:11,640 --> 01:04:14,580 I'll make one note in terms of trading, which is not 1214 01:04:14,580 --> 01:04:16,830 related to machine learning, but just related 1215 01:04:16,830 --> 01:04:20,730 to the corona crisis that we are all living in. 1216 01:04:20,730 --> 01:04:23,970 I have an overall belief that this coronavirus 1217 01:04:23,970 --> 01:04:28,680 crisis will accentuate trends that we've already seen. 1218 01:04:28,680 --> 01:04:32,700 In industry after industry, if we're locked down 1219 01:04:32,700 --> 01:04:37,320 for two, three, or four months, or God forbid, for 18 or 24 1220 01:04:37,320 --> 01:04:38,610 months-- 1221 01:04:38,610 --> 01:04:41,130 if we're locked down for that long, 1222 01:04:41,130 --> 01:04:44,820 we're going to find new ways to engage 1223 01:04:44,820 --> 01:04:47,460 with each other in economic activity 1224 01:04:47,460 --> 01:04:49,620 and social activity and the like. 1225 01:04:49,620 --> 01:04:52,890 And we've already had some trends, deep trends-- 1226 01:04:52,890 --> 01:04:55,770 we talked about them Monday-- 1227 01:04:55,770 --> 01:05:00,270 that we're unlikely to be using many paper money and coinage 1228 01:05:00,270 --> 01:05:01,880 money. 1229 01:05:01,880 --> 01:05:07,250 Three months from now, nearly 70% or 80% the world 1230 01:05:07,250 --> 01:05:11,390 will have forgotten how to use paper money and coinage. 1231 01:05:11,390 --> 01:05:14,810 In fact, it will even be viewed as a disease delivery device. 1232 01:05:14,810 --> 01:05:15,820 It might be dirty. 1233 01:05:15,820 --> 01:05:17,570 It might be something we don't want to use 1234 01:05:17,570 --> 01:05:21,900 because it could be a problem. 1235 01:05:21,900 --> 01:05:23,810 Well, let me talk about trading for a second. 1236 01:05:23,810 --> 01:05:28,340 The New York Stock Exchange and the world's largest stock 1237 01:05:28,340 --> 01:05:30,620 exchanges are now electronic. 1238 01:05:30,620 --> 01:05:34,050 They could have done that two years ago. 1239 01:05:34,050 --> 01:05:35,880 When the Intercontinental Exchange, 1240 01:05:35,880 --> 01:05:38,580 which is a big public company, bought the New York Stock 1241 01:05:38,580 --> 01:05:43,110 Exchange a handful of years ago, five or so years ago, 1242 01:05:43,110 --> 01:05:45,360 Jeff Sprecher, the entrepreneur who 1243 01:05:45,360 --> 01:05:49,800 started the Intercontinental Exchange in 1998 or '99, 1244 01:05:49,800 --> 01:05:53,280 he's always been an entrepreneur, an innovator, 1245 01:05:53,280 --> 01:05:55,098 to do electronic trading. 1246 01:05:55,098 --> 01:05:57,390 They could have taken the New York Stock Exchange fully 1247 01:05:57,390 --> 01:05:59,040 electronic, but guess what? 1248 01:05:59,040 --> 01:06:02,310 That's what happened in the last three weeks. 1249 01:06:02,310 --> 01:06:05,750 So after we get out of this lockdown period, 1250 01:06:05,750 --> 01:06:11,370 will we bring back the floor of these London and New 1251 01:06:11,370 --> 01:06:16,200 York and Shanghai and Mumbai and so forth? 1252 01:06:16,200 --> 01:06:18,420 Will we bring back the floors? 1253 01:06:18,420 --> 01:06:21,711 I think quite possibly not. 1254 01:06:21,711 --> 01:06:29,420 Not sure, but there's a lot that's shifting on, I think. 1255 01:06:29,420 --> 01:06:31,460 Natural language processing-- we talked 1256 01:06:31,460 --> 01:06:35,360 about in customer service, process automation, 1257 01:06:35,360 --> 01:06:36,770 and sentiment analysis. 1258 01:06:36,770 --> 01:06:39,860 These are sort of the slices that I think about 1259 01:06:39,860 --> 01:06:42,690 in these fields. 1260 01:06:42,690 --> 01:06:47,570 So I was curious how many people have ever used Siri? 1261 01:06:47,570 --> 01:06:50,210 Probably almost every hand would go up. 1262 01:06:50,210 --> 01:06:53,225 But how many people have ever used Erica? 1263 01:06:57,610 --> 01:07:00,572 ROMAIN: So perhaps we have Shaheryar who'd 1264 01:07:00,572 --> 01:07:01,780 like to share his experience. 1265 01:07:01,780 --> 01:07:05,880 Sorry for your name-- mispronouncing it. 1266 01:07:05,880 --> 01:07:08,650 STUDENT: Yeah, so it's essentially like Siri, 1267 01:07:08,650 --> 01:07:11,390 but you actually-- it's a product of Bank of America, 1268 01:07:11,390 --> 01:07:14,740 and you can use it to check your spending habits. 1269 01:07:14,740 --> 01:07:17,290 You can also use it to, if you need things with regards 1270 01:07:17,290 --> 01:07:20,150 to check depositing, or if you want to know something, 1271 01:07:20,150 --> 01:07:20,990 it can do that. 1272 01:07:20,990 --> 01:07:25,626 But currently, I believe it's not as refined as Siri, 1273 01:07:25,626 --> 01:07:27,626 and I still think there is a lot room over there 1274 01:07:27,626 --> 01:07:30,650 for improvement. 1275 01:07:30,650 --> 01:07:32,780 GARY GENSLER: Why do you think it is that Erica-- 1276 01:07:32,780 --> 01:07:35,510 and JP Morgan has one as well. 1277 01:07:35,510 --> 01:07:38,450 I can't remember what her name is. 1278 01:07:38,450 --> 01:07:44,960 They all do seem to be mostly female voices, 1279 01:07:44,960 --> 01:07:46,060 if I'm not mistaken. 1280 01:07:46,060 --> 01:07:50,980 But why is it that Erica and the like, 1281 01:07:50,980 --> 01:07:55,040 a virtual assistant, as they're called in finance, 1282 01:07:55,040 --> 01:07:59,550 aren't as developed as the Siris and Alexas, do you think? 1283 01:07:59,550 --> 01:08:01,885 STUDENT: Sorry, can you repeat the question? 1284 01:08:01,885 --> 01:08:03,260 GARY GENSLER: Anybody can answer. 1285 01:08:03,260 --> 01:08:07,430 Why is it that the finance virtual assistants like Erica 1286 01:08:07,430 --> 01:08:13,370 are not yet as fully developed as the Home and other ones 1287 01:08:13,370 --> 01:08:16,050 like Siri and Alexa? 1288 01:08:16,050 --> 01:08:17,240 STUDENT: I believe some of-- 1289 01:08:17,240 --> 01:08:19,157 I think it's got something to do with the fact 1290 01:08:19,157 --> 01:08:21,810 that the number of users for financial assistance 1291 01:08:21,810 --> 01:08:25,370 is way lower as compared to Alexa, Siri, or Alexa 1292 01:08:25,370 --> 01:08:26,510 or whatever. 1293 01:08:26,510 --> 01:08:30,920 So I believe that is something which may explain it, 1294 01:08:30,920 --> 01:08:35,508 the disparity between these two kinds of assistants. 1295 01:08:35,508 --> 01:08:36,300 GARY GENSLER: Yeah. 1296 01:08:36,300 --> 01:08:39,160 Any other thoughts on that, one why-- 1297 01:08:39,160 --> 01:08:43,250 ROMAIN: Nikhil has a different take, and that we have Laira. 1298 01:08:43,250 --> 01:08:45,899 STUDENT: Probably along similar lines, banks so far, 1299 01:08:45,899 --> 01:08:49,609 the interactions probably have been in person or over phones, 1300 01:08:49,609 --> 01:08:52,550 and they weren't used to processing data and processing 1301 01:08:52,550 --> 01:08:53,310 requests. 1302 01:08:53,310 --> 01:08:55,520 I think they have a smaller data set 1303 01:08:55,520 --> 01:08:59,960 to go through and understand what problem customer 1304 01:08:59,960 --> 01:09:01,000 questions are. 1305 01:09:01,000 --> 01:09:04,040 And that's probably a limiting factor versus, say, Siri, 1306 01:09:04,040 --> 01:09:07,819 they have much more data on everyday users. 1307 01:09:07,819 --> 01:09:10,583 I think that's probably the biggest differentiator. 1308 01:09:10,583 --> 01:09:13,000 GARY GENSLER: Yeah, and was there another comment, Romain, 1309 01:09:13,000 --> 01:09:13,790 you said? 1310 01:09:13,790 --> 01:09:14,990 Sure. 1311 01:09:14,990 --> 01:09:16,779 ROMAIN: So I think Laira had her hand up, 1312 01:09:16,779 --> 01:09:18,330 but I think she withdrew it. 1313 01:09:18,330 --> 01:09:20,660 So perhaps we can hear from Brian. 1314 01:09:20,660 --> 01:09:21,819 GARY GENSLER: OK. 1315 01:09:21,819 --> 01:09:23,960 STUDENT: So in addition to the data, 1316 01:09:23,960 --> 01:09:26,439 I think there's also a human capital element. 1317 01:09:26,439 --> 01:09:29,410 It's possible that Apple has better human capital 1318 01:09:29,410 --> 01:09:32,840 capabilities than do these financial institutions, 1319 01:09:32,840 --> 01:09:34,510 so it's harnessing that data as well. 1320 01:09:34,510 --> 01:09:35,420 GARY GENSLER: Yeah. 1321 01:09:35,420 --> 01:09:37,359 So what we've just talked about was data, 1322 01:09:37,359 --> 01:09:40,779 human capital, experience-- 1323 01:09:40,779 --> 01:09:41,470 all true. 1324 01:09:41,470 --> 01:09:43,660 Also, Erica and the financial firms 1325 01:09:43,660 --> 01:09:48,470 only started more recently, and so forth. 1326 01:09:48,470 --> 01:09:51,970 But the voice recognition programs and then 1327 01:09:51,970 --> 01:09:58,660 taking that data that an Apple has or their competitors 1328 01:09:58,660 --> 01:10:00,660 in big tech around the globe-- 1329 01:10:00,660 --> 01:10:03,430 because it's not just here in the US-- 1330 01:10:03,430 --> 01:10:06,400 is really remarkable now. 1331 01:10:06,400 --> 01:10:08,710 Even to the extent that I don't know 1332 01:10:08,710 --> 01:10:13,030 how many people use earbuds, but if you look closely 1333 01:10:13,030 --> 01:10:17,380 at the user agreement on the airbuds that you use, it says-- 1334 01:10:17,380 --> 01:10:22,060 and if it's an Apple, if I'm mistaken, 1335 01:10:22,060 --> 01:10:24,280 the Apple lawyers will chase me. 1336 01:10:24,280 --> 01:10:26,695 But the user agreement says that they 1337 01:10:26,695 --> 01:10:32,350 can listen to that to help you to make sure 1338 01:10:32,350 --> 01:10:34,930 that there's not a drop between your earbuds 1339 01:10:34,930 --> 01:10:37,240 and your cell phone. 1340 01:10:37,240 --> 01:10:45,050 They are picking up vast amounts of data, vast amounts of data 1341 01:10:45,050 --> 01:10:47,240 from our text messaging. 1342 01:10:47,240 --> 01:10:49,340 If you look at something like Google, 1343 01:10:49,340 --> 01:10:53,060 they're picking it up from Google Chrome, Google Maps, 1344 01:10:53,060 --> 01:10:55,180 Gmail. 1345 01:10:55,180 --> 01:10:57,310 Multiple places that they can pick up, 1346 01:10:57,310 --> 01:11:00,640 and we talked about this conceptually, big tech 1347 01:11:00,640 --> 01:11:04,120 versus big finance versus startups in this triangle 1348 01:11:04,120 --> 01:11:06,910 of competitive landscape. 1349 01:11:06,910 --> 01:11:10,090 And why I wanted to sort of close on Bank of America Erica 1350 01:11:10,090 --> 01:11:12,460 and this discussion of Erica versus Siri 1351 01:11:12,460 --> 01:11:18,020 and Alexa is big tech using Google, just as an example, 1352 01:11:18,020 --> 01:11:22,270 has this remarkable network that they're layering activities. 1353 01:11:22,270 --> 01:11:25,430 Remember, we said data network activities-- 1354 01:11:25,430 --> 01:11:29,040 that's the Bank of International Settlement way to put it, 1355 01:11:29,040 --> 01:11:30,520 and what a perfect example to show. 1356 01:11:30,520 --> 01:11:32,680 Google has Gmail. 1357 01:11:32,680 --> 01:11:34,660 It has Maps. 1358 01:11:34,660 --> 01:11:36,100 It has Google Chrome. 1359 01:11:36,100 --> 01:11:39,110 It has Android, the operating system. 1360 01:11:39,110 --> 01:11:42,070 So all of these different ways to build their network, 1361 01:11:42,070 --> 01:11:44,980 and they layer activities on top of it, 1362 01:11:44,980 --> 01:11:47,890 and then vast amounts of data come in 1363 01:11:47,890 --> 01:11:52,430 and the human capital that was mentioned at the end there. 1364 01:11:52,430 --> 01:11:56,350 And they have more experience to move it forward. 1365 01:11:56,350 --> 01:11:58,400 Apple, similarly. 1366 01:11:58,400 --> 01:12:04,290 Baidu and Alibaba in China and so forth, similarly. 1367 01:12:04,290 --> 01:12:07,800 If I were a CEO of a big incumbent, 1368 01:12:07,800 --> 01:12:12,774 yes, I would be very focused on the fintech startups, but I 1369 01:12:12,774 --> 01:12:13,710 tell you-- 1370 01:12:13,710 --> 01:12:16,230 I'd be looking at big tech in a way 1371 01:12:16,230 --> 01:12:20,010 that their advantages are really significant, very 1372 01:12:20,010 --> 01:12:22,260 significant advantages. 1373 01:12:22,260 --> 01:12:25,225 Romain, I see some hands up maybe. 1374 01:12:25,225 --> 01:12:28,310 ROMAIN: We now have Laira who has her hand up. 1375 01:12:28,310 --> 01:12:32,060 STUDENT: Yeah, I'm just curious to know what do you think? 1376 01:12:32,060 --> 01:12:37,580 So currently, we know that tech has kind of-- or AI has kind of 1377 01:12:37,580 --> 01:12:39,890 emerged to the financial and payment space 1378 01:12:39,890 --> 01:12:41,820 in the form of virtual assistants, 1379 01:12:41,820 --> 01:12:45,170 but what do you anticipate the next step would be in terms 1380 01:12:45,170 --> 01:12:48,370 of this integration of technology and AI 1381 01:12:48,370 --> 01:12:50,090 into the payment space? 1382 01:12:50,090 --> 01:12:54,610 What next after the virtual assistants? 1383 01:12:54,610 --> 01:12:57,440 GARY GENSLER: We're going to have a whole class on payments 1384 01:12:57,440 --> 01:12:59,960 specifically, but I think that what's 1385 01:12:59,960 --> 01:13:04,060 happening in the payment space is 1386 01:13:04,060 --> 01:13:07,690 we've seen specialized payment service providers. 1387 01:13:07,690 --> 01:13:09,640 Of course, we've seen a lot of the competition 1388 01:13:09,640 --> 01:13:12,400 starting with PayPal in 1998. 1389 01:13:12,400 --> 01:13:18,460 This is not a new space for disruption. 1390 01:13:18,460 --> 01:13:20,890 But what we've seen more recently 1391 01:13:20,890 --> 01:13:24,640 is, in the retail space, whether it's companies I've mentioned 1392 01:13:24,640 --> 01:13:27,490 earlier, like Toast, that got into one vertical, 1393 01:13:27,490 --> 01:13:31,010 one slice within payments, which was restaurants-- 1394 01:13:31,010 --> 01:13:38,280 they can provide a better product for that slice. 1395 01:13:38,280 --> 01:13:42,780 And then can collect back to AI enough robust data 1396 01:13:42,780 --> 01:13:44,550 within that slice-- 1397 01:13:44,550 --> 01:13:47,640 this is using Toast as an example-- 1398 01:13:47,640 --> 01:13:54,420 that they can provide better software, better hardware, 1399 01:13:54,420 --> 01:14:00,090 and also less risk loans. 1400 01:14:00,090 --> 01:14:03,030 Basically, as Toast started to provide lending 1401 01:14:03,030 --> 01:14:08,430 to restaurants within that space, built upon the payment. 1402 01:14:08,430 --> 01:14:13,740 So it's the marriage between the user experience providing 1403 01:14:13,740 --> 01:14:14,550 the users-- 1404 01:14:14,550 --> 01:14:17,170 in that case, the restaurants in the payment space-- 1405 01:14:17,170 --> 01:14:20,940 but providing the users in that space something 1406 01:14:20,940 --> 01:14:23,780 that the generalized platform-- 1407 01:14:23,780 --> 01:14:26,340 a bank payment system is generalized. 1408 01:14:26,340 --> 01:14:29,460 It's multi sector. 1409 01:14:29,460 --> 01:14:31,635 It's a general product, and Toast 1410 01:14:31,635 --> 01:14:33,510 was able to say, no, we can provide something 1411 01:14:33,510 --> 01:14:35,233 that just restaurants-- 1412 01:14:35,233 --> 01:14:36,900 there might be something a little unique 1413 01:14:36,900 --> 01:14:38,760 about the restaurant business that we can 1414 01:14:38,760 --> 01:14:40,740 provide software and hardware. 1415 01:14:40,740 --> 01:14:42,750 In their case, it was tablets. 1416 01:14:42,750 --> 01:14:45,330 They were providing tablets for the servers 1417 01:14:45,330 --> 01:14:48,000 to walk around and take the orders. 1418 01:14:48,000 --> 01:14:51,540 They could integrate the menu right into the payment app. 1419 01:14:51,540 --> 01:14:54,630 So there was something a little bit unique about that. 1420 01:14:54,630 --> 01:14:57,450 But then based on that, they get a bunch of data, 1421 01:14:57,450 --> 01:15:01,020 and that data helps them with, I would say, 1422 01:15:01,020 --> 01:15:04,770 underwriting decisions based on-- 1423 01:15:04,770 --> 01:15:07,040 it doesn't have to be machine learning. 1424 01:15:07,040 --> 01:15:11,010 But it's enhanced data analytics because of machine learning. 1425 01:15:11,010 --> 01:15:12,020 So I hope that helps. 1426 01:15:12,020 --> 01:15:15,670 I think on the conversational agents 1427 01:15:15,670 --> 01:15:19,610 and the virtual assistants, what we're seeing in the payment 1428 01:15:19,610 --> 01:15:21,230 space, because that was your question, 1429 01:15:21,230 --> 01:15:26,960 is we're moving from card authorized payments 1430 01:15:26,960 --> 01:15:30,260 to mobile app QR codes. 1431 01:15:30,260 --> 01:15:35,540 Then the QR codes is not based upon virtual assistants, 1432 01:15:35,540 --> 01:15:37,820 but it's an interesting question whether we'll 1433 01:15:37,820 --> 01:15:42,830 get to some voice authenticated payments. 1434 01:15:42,830 --> 01:15:48,080 There are a lot of uses of voice authentication already. 1435 01:15:48,080 --> 01:15:50,930 Vanguard and many other asset managers, 1436 01:15:50,930 --> 01:15:52,760 where you can have your brokerage accounts, 1437 01:15:52,760 --> 01:15:55,220 you can call in and get voice authenticated 1438 01:15:55,220 --> 01:15:57,640 before you can do a trade. 1439 01:15:57,640 --> 01:16:00,280 And that voice authentication is just 1440 01:16:00,280 --> 01:16:04,900 like other forms of authentication, 1441 01:16:04,900 --> 01:16:09,840 but it's not perfect as of this moment.