1 00:00:01,020 --> 00:00:03,390 The following content is provided under a Creative 2 00:00:03,390 --> 00:00:04,780 Commons license. 3 00:00:04,780 --> 00:00:06,990 Your support will help MIT OpenCourseWare 4 00:00:06,990 --> 00:00:11,080 continue to offer high quality educational resources for free. 5 00:00:11,080 --> 00:00:13,650 To make a donation or to view additional materials 6 00:00:13,650 --> 00:00:17,580 from hundreds of MIT courses, visit MIT OpenCourseWare 7 00:00:17,580 --> 00:00:18,486 at ocw.mit.edu. 8 00:00:21,742 --> 00:00:23,700 GARY GENSLER: So we're going to come back again 9 00:00:23,700 --> 00:00:25,710 today to blockchain economics. 10 00:00:25,710 --> 00:00:29,250 A lot of what we're going to talk about today while again 11 00:00:29,250 --> 00:00:33,240 anchored in the readings will also 12 00:00:33,240 --> 00:00:38,220 be relevant as we now turn to after Sip 13 00:00:38,220 --> 00:00:41,640 week to act three where we're going through use cases 14 00:00:41,640 --> 00:00:45,130 but also hopefully relevant as you're starting to think, OK, 15 00:00:45,130 --> 00:00:48,130 what about this final project and so forth 16 00:00:48,130 --> 00:00:51,350 and what do we need to do? 17 00:00:51,350 --> 00:00:54,570 Five or six of the groups have come in either 18 00:00:54,570 --> 00:00:58,050 as an individual person or teams of three or four have come in 19 00:00:58,050 --> 00:01:00,330 and bounced ideas off of me. 20 00:01:00,330 --> 00:01:03,186 So I want to thank all of the groups that have come in, 21 00:01:03,186 --> 00:01:04,769 because you've helped me also organize 22 00:01:04,769 --> 00:01:07,070 some of my talking points for today 23 00:01:07,070 --> 00:01:09,330 as to what we're going through. 24 00:01:09,330 --> 00:01:12,900 And those of you who haven't come in, 25 00:01:12,900 --> 00:01:15,520 feel free to set things up if you find it helpful. 26 00:01:15,520 --> 00:01:18,390 It's not mandatory. 27 00:01:18,390 --> 00:01:20,370 But you can also thank the six groups 28 00:01:20,370 --> 00:01:22,290 that have come in, because now I'm 29 00:01:22,290 --> 00:01:28,410 trying to anticipate your questions as well 30 00:01:28,410 --> 00:01:30,850 along the way. 31 00:01:30,850 --> 00:01:35,970 So, the overview of course our reading and study questions 32 00:01:35,970 --> 00:01:37,770 will come through. 33 00:01:37,770 --> 00:01:40,290 There was the letter to Jamie Dimon. 34 00:01:40,290 --> 00:01:45,000 So we're going to dive into that one a little bit. 35 00:01:45,000 --> 00:01:47,340 I found it interesting when I read it about a year ago. 36 00:01:47,340 --> 00:01:51,370 And then I thought, why not throw it into to the mix here? 37 00:01:51,370 --> 00:01:55,200 The McKinsey report, which, again, I 38 00:01:55,200 --> 00:01:56,920 found this to be true for years. 39 00:01:56,920 --> 00:01:58,620 It's not just this class that you 40 00:01:58,620 --> 00:02:01,830 can find some consultant that's put a generalized paper out, 41 00:02:01,830 --> 00:02:04,652 it does sort of just skim the tops, 42 00:02:04,652 --> 00:02:06,360 but you're getting a sense of how they're 43 00:02:06,360 --> 00:02:08,970 trying to gin up business. 44 00:02:08,970 --> 00:02:11,962 Their business model is to write some of these reports. 45 00:02:11,962 --> 00:02:13,920 And some of you are going to go into consulting 46 00:02:13,920 --> 00:02:16,990 and maybe even write these reports at some point in time. 47 00:02:16,990 --> 00:02:19,470 But it's often a good way to see what I'll call it 48 00:02:19,470 --> 00:02:22,890 just a survey, a topical survey in how 49 00:02:22,890 --> 00:02:24,600 folks think about things. 50 00:02:24,600 --> 00:02:26,760 Quickly a little bit on potential use cases. 51 00:02:26,760 --> 00:02:29,970 In the heart of today's discussion 52 00:02:29,970 --> 00:02:31,710 and kind of the heart for thinking 53 00:02:31,710 --> 00:02:35,940 about the final projects is how to really assess 54 00:02:35,940 --> 00:02:39,660 the costs and benefits of any potential use case. 55 00:02:39,660 --> 00:02:44,100 So I'll skip through the study questions. 56 00:02:44,100 --> 00:02:46,830 But these are the key. 57 00:02:46,830 --> 00:02:49,200 I didn't see anybody go in the discussion board. 58 00:02:49,200 --> 00:02:49,890 I looked. 59 00:02:49,890 --> 00:02:51,690 Did you see anything there? 60 00:02:51,690 --> 00:02:52,560 No. 61 00:02:52,560 --> 00:02:53,430 we did set it up. 62 00:02:56,250 --> 00:02:58,740 But what are the potential benefits, 63 00:02:58,740 --> 00:03:02,910 and how do you assess the cost of trusts? 64 00:03:02,910 --> 00:03:07,950 And that is going to be true in every single project you 65 00:03:07,950 --> 00:03:08,460 look at. 66 00:03:08,460 --> 00:03:11,300 If you take anything from this class, 67 00:03:11,300 --> 00:03:15,240 it's this core critical reasoning of like-- 68 00:03:15,240 --> 00:03:17,490 and hopefully after the Roubini paper 69 00:03:17,490 --> 00:03:21,540 last Tuesday, we're going to come out of the doldrums 70 00:03:21,540 --> 00:03:23,880 and we're going to pull back out of the minimalist side 71 00:03:23,880 --> 00:03:25,842 and get more in the middle. 72 00:03:25,842 --> 00:03:28,050 I might not get Alin lean out of the minimalist side, 73 00:03:28,050 --> 00:03:28,750 I don't know. 74 00:03:28,750 --> 00:03:30,870 But it's interesting. 75 00:03:30,870 --> 00:03:33,480 Some groups come in and I find the three or four people that 76 00:03:33,480 --> 00:03:34,920 are sitting there I ask. 77 00:03:34,920 --> 00:03:36,810 In and even amongst the group, there's 78 00:03:36,810 --> 00:03:39,550 somebody who's a maximalist and somebody who's a minimalist. 79 00:03:39,550 --> 00:03:43,770 So I think we're being successful that we're not 80 00:03:43,770 --> 00:03:45,765 going to come out all in the same place. 81 00:03:48,700 --> 00:03:51,780 So, of course, since we're going to talk about the reading, 82 00:03:51,780 --> 00:03:55,170 let's jumped through, a letter to Jamie Dimon. 83 00:03:55,170 --> 00:03:59,920 Anybody want to tell me what you took from that letter? 84 00:03:59,920 --> 00:04:01,810 AUDIENCE: I really loved this article. 85 00:04:01,810 --> 00:04:04,290 It is my favorite reading so far. 86 00:04:04,290 --> 00:04:06,105 GARY GENSLER: Oh my gosh, 12 classes. 87 00:04:06,105 --> 00:04:06,605 All right. 88 00:04:09,330 --> 00:04:11,250 AUDIENCE: So I though that he really great 89 00:04:11,250 --> 00:04:14,940 job of stripping some of the hyperbole, both the maximalist 90 00:04:14,940 --> 00:04:17,890 and the minimalist hyperbole surrounding this out 91 00:04:17,890 --> 00:04:21,000 of it to lean on really what it is 92 00:04:21,000 --> 00:04:23,590 and a very good explanation of how it works. 93 00:04:23,590 --> 00:04:26,610 And then also, the note of caution 94 00:04:26,610 --> 00:04:29,347 to see if these are the dimensions around which-- 95 00:04:29,347 --> 00:04:30,930 you should wait til you put a word in, 96 00:04:30,930 --> 00:04:33,780 but these are the reasons why there's 97 00:04:33,780 --> 00:04:35,007 potentially big implications. 98 00:04:35,007 --> 00:04:35,840 GARY GENSLER: Right. 99 00:04:35,840 --> 00:04:36,923 And he wants to add to it. 100 00:04:36,923 --> 00:04:39,910 I'm going to hold-- there was a hand before Alin? 101 00:04:39,910 --> 00:04:40,980 AUDIENCE: Yeah, no. 102 00:04:40,980 --> 00:04:42,897 I actually was going to say something similar. 103 00:04:42,897 --> 00:04:48,240 But I thought he highlighted a lot the trade-off between trust 104 00:04:48,240 --> 00:04:53,820 and the decentralized properties of blockchain 105 00:04:53,820 --> 00:04:57,180 versus the cost you must be willing to pay. 106 00:04:57,180 --> 00:05:00,170 GARY GENSLER: And remind me your first name. 107 00:05:00,170 --> 00:05:00,980 AUDIENCE: Leonardo. 108 00:05:00,980 --> 00:05:01,938 GARY GENSLER: Leonardo. 109 00:05:01,938 --> 00:05:02,460 All right. 110 00:05:02,460 --> 00:05:03,960 AUDIENCE: I agree with both of them. 111 00:05:03,960 --> 00:05:08,015 Also, very good writing and style. 112 00:05:08,015 --> 00:05:10,140 One thing that I didn't like is that at some point, 113 00:05:10,140 --> 00:05:12,330 he raised the question, well, who 114 00:05:12,330 --> 00:05:13,500 needs censorship resistance? 115 00:05:13,500 --> 00:05:15,417 Because it seems like this is the only benefit 116 00:05:15,417 --> 00:05:17,140 of these decentralized applications. 117 00:05:17,140 --> 00:05:18,973 And he didn't answer that question actually. 118 00:05:18,973 --> 00:05:20,807 He just gave two vague broad answers, 119 00:05:20,807 --> 00:05:22,140 but didn't answer that question. 120 00:05:22,140 --> 00:05:23,940 So I think that's a really good question. 121 00:05:23,940 --> 00:05:25,320 Who needs this stuff, anyway? 122 00:05:25,320 --> 00:05:27,370 GARY GENSLER: All right. 123 00:05:27,370 --> 00:05:28,440 Censorship resistance. 124 00:05:28,440 --> 00:05:29,250 We're going to come back to it. 125 00:05:29,250 --> 00:05:30,010 Stephanie. 126 00:05:30,010 --> 00:05:30,270 AUDIENCE: Yeah. 127 00:05:30,270 --> 00:05:32,490 So actually, I thought censorship resistance was also 128 00:05:32,490 --> 00:05:34,532 one of the more interesting parts of this letter. 129 00:05:34,532 --> 00:05:37,560 Because to me, when you read through what they highlighted 130 00:05:37,560 --> 00:05:39,790 that Jamie Dimon had said, one of the things 131 00:05:39,790 --> 00:05:44,950 was people who use Bitcoin or Blockchain are criminals. 132 00:05:44,950 --> 00:05:48,990 And I didn't find that his focus on censorship resistance 133 00:05:48,990 --> 00:05:53,090 really rebutted Jamie Dimon's argument. 134 00:05:53,090 --> 00:05:55,660 GARY GENSLER: So-- 135 00:05:55,660 --> 00:05:56,210 Sean. 136 00:05:56,210 --> 00:05:58,890 AUDIENCE: Oh, I think that censorship resistance is really 137 00:05:58,890 --> 00:06:01,320 interested in a way that now I actually almost feel 138 00:06:01,320 --> 00:06:04,350 why we spend so much time and energy on trying to understand 139 00:06:04,350 --> 00:06:06,630 public policy or why is public policy 140 00:06:06,630 --> 00:06:12,161 such a big thing in response to decentralized apps. 141 00:06:12,161 --> 00:06:14,470 I think this is a key question. 142 00:06:14,470 --> 00:06:17,040 GARY GENSLER: So I think of-- 143 00:06:17,040 --> 00:06:19,645 AUDIENCE: I just wanted to-- well, I definitely agree. 144 00:06:19,645 --> 00:06:22,020 Because that was one of Dimon's points, that, well, yeah, 145 00:06:22,020 --> 00:06:24,285 if you're a criminal, this could be useful. 146 00:06:24,285 --> 00:06:25,660 And then he talked about how it's 147 00:06:25,660 --> 00:06:28,035 worse than a lot of aspects except censorship resistance. 148 00:06:28,035 --> 00:06:30,030 But I think the analogy he used was 149 00:06:30,030 --> 00:06:33,973 the rise in encrypted messaging that has just become-- 150 00:06:33,973 --> 00:06:35,556 it wasn't really predictable that that 151 00:06:35,556 --> 00:06:38,050 would be such a big thing, but now that's gotten 152 00:06:38,050 --> 00:06:40,980 use cases in the world. 153 00:06:40,980 --> 00:06:43,680 GARY GENSLER: So it seems like people really 154 00:06:43,680 --> 00:06:46,110 like the writing, not just the style 155 00:06:46,110 --> 00:06:49,795 but that it took on and had a balanced approach. 156 00:06:49,795 --> 00:06:51,420 But you're highlighting, wait a minute, 157 00:06:51,420 --> 00:06:54,000 he ends with censorship resistance, 158 00:06:54,000 --> 00:06:55,830 but then doesn't take it the next way. 159 00:06:55,830 --> 00:06:57,515 Like, what does that mean? 160 00:06:57,515 --> 00:06:58,890 And so I think we're going to try 161 00:06:58,890 --> 00:07:01,650 to tease that out a little bit, but I think of it 162 00:07:01,650 --> 00:07:03,910 as in two ways. 163 00:07:03,910 --> 00:07:04,980 It's the individual. 164 00:07:04,980 --> 00:07:08,310 Any one of us could be blocked from doing something. 165 00:07:08,310 --> 00:07:12,240 We could not get a service, we could not 166 00:07:12,240 --> 00:07:17,340 be allowed to take credit if somebody is allocating credit. 167 00:07:17,340 --> 00:07:18,330 Think about Uber. 168 00:07:18,330 --> 00:07:22,770 If Uber were censoring, we can't get a ride any longer. 169 00:07:22,770 --> 00:07:25,620 How many of us rate everybody a five when 170 00:07:25,620 --> 00:07:27,960 we ride our Ubers because we think, 171 00:07:27,960 --> 00:07:30,270 well, maybe they'll pick me up. 172 00:07:30,270 --> 00:07:35,280 If I rated everybody a three, maybe they'll censor me. 173 00:07:35,280 --> 00:07:38,430 I don't know, but maybe you all give more legitimate feedback. 174 00:07:38,430 --> 00:07:39,305 I just hit five. 175 00:07:42,060 --> 00:07:44,200 So it's the individual censorship, 176 00:07:44,200 --> 00:07:46,770 but I think of it as a second thing also, 177 00:07:46,770 --> 00:07:51,360 is barriers to entry, a commercial more broad market 178 00:07:51,360 --> 00:07:53,010 barrier. 179 00:07:53,010 --> 00:07:57,090 And that in some ways, the blockchain technology 180 00:07:57,090 --> 00:08:03,030 might allow to build something where incumbents already 181 00:08:03,030 --> 00:08:04,320 have barriers to entry. 182 00:08:04,320 --> 00:08:06,670 They might have used public policy to get the barriers, 183 00:08:06,670 --> 00:08:07,170 by the way. 184 00:08:07,170 --> 00:08:12,930 They might have regulatory barriers as well. 185 00:08:12,930 --> 00:08:15,660 But more specifically, he did define-- 186 00:08:15,660 --> 00:08:18,400 it was kind of interesting-- crypto assets. 187 00:08:18,400 --> 00:08:21,208 And he was one that-- 188 00:08:21,208 --> 00:08:23,250 I'm trying to remember the gentleman who wrote it 189 00:08:23,250 --> 00:08:24,540 because it was on Chain. 190 00:08:24,540 --> 00:08:25,920 It was the head of Chain. 191 00:08:25,920 --> 00:08:26,470 What's that? 192 00:08:26,470 --> 00:08:27,550 AUDIENCE: Adam Ludwin. 193 00:08:27,550 --> 00:08:30,480 GARY GENSLER: Adam Ludwin. 194 00:08:30,480 --> 00:08:34,110 But he too said, let's not use the word cryptocurrency. 195 00:08:34,110 --> 00:08:36,120 Let's use the word crypto assets. 196 00:08:36,120 --> 00:08:38,710 They enable a decentralized application. 197 00:08:38,710 --> 00:08:42,390 And that was a key thing he was trying to say to Jamie Dimon. 198 00:08:42,390 --> 00:08:44,850 It depends on how much you believe 199 00:08:44,850 --> 00:08:47,380 in decentralized applications. 200 00:08:47,380 --> 00:08:51,030 If you don't see a value in decentralized applications, 201 00:08:51,030 --> 00:08:54,420 Ludwin, said, all right, I get it, 202 00:08:54,420 --> 00:08:56,940 I'll agree with you, Mr. Dimon. 203 00:08:56,940 --> 00:09:01,830 But there is a benefit for decentralized applications 204 00:09:01,830 --> 00:09:04,170 and a mechanism to allocate resources 205 00:09:04,170 --> 00:09:06,910 to a specific organization. 206 00:09:06,910 --> 00:09:12,960 So some incentive way, some way to allocate resources. 207 00:09:12,960 --> 00:09:13,720 I would say this. 208 00:09:13,720 --> 00:09:16,530 We've studied money a lot this semester, 209 00:09:16,530 --> 00:09:22,180 the medium of exchange, unit of a cat, store of value. 210 00:09:22,180 --> 00:09:24,840 But it's also possible that these crypto assets have 211 00:09:24,840 --> 00:09:28,830 some other benefit, that they're not really truly competing 212 00:09:28,830 --> 00:09:33,330 with the US dollar or competing with the Euro, 213 00:09:33,330 --> 00:09:35,760 but maybe they're competing with-- who's my friend who's 214 00:09:35,760 --> 00:09:36,970 the gamer over here? 215 00:09:36,970 --> 00:09:39,002 Skins, yeah. 216 00:09:39,002 --> 00:09:39,960 What's your name again? 217 00:09:39,960 --> 00:09:40,590 AUDIENCE: Mike. 218 00:09:40,590 --> 00:09:41,382 GARY GENSLER: Mike. 219 00:09:41,382 --> 00:09:44,130 I'm going to be always looking to my right for the-- 220 00:09:44,130 --> 00:09:48,840 but it could be an incentive and be competing 221 00:09:48,840 --> 00:09:51,570 with skins, in a sense. 222 00:09:51,570 --> 00:09:55,590 And so I took that, that, Mr. Dimon, 223 00:09:55,590 --> 00:09:58,930 don't think of it only versus the Euro or the Dollar 224 00:09:58,930 --> 00:10:00,635 because maybe you're right. 225 00:10:00,635 --> 00:10:02,010 But it could have something to do 226 00:10:02,010 --> 00:10:05,760 with decentralized applications. 227 00:10:05,760 --> 00:10:06,592 Kelly. 228 00:10:06,592 --> 00:10:10,060 AUDIENCE: To add to that, Bitcoin is capitalism still. 229 00:10:10,060 --> 00:10:11,182 He should love it. 230 00:10:11,182 --> 00:10:12,640 GARY GENSLER: That you should still 231 00:10:12,640 --> 00:10:16,030 love Bitcoin and capitalism, and that the banks 232 00:10:16,030 --> 00:10:17,440 aren't going away. 233 00:10:17,440 --> 00:10:19,010 They will probably still have banks. 234 00:10:19,010 --> 00:10:21,260 AUDIENCE: It is incredible that he didn't use the word 235 00:10:21,260 --> 00:10:25,240 blockchain till the very end to asterisk and say, 236 00:10:25,240 --> 00:10:27,020 you'll notice I don't use blockchain, 237 00:10:27,020 --> 00:10:30,987 and I thought kind of a genius way to separate [INAUDIBLE].. 238 00:10:30,987 --> 00:10:31,820 GARY GENSLER: Right. 239 00:10:31,820 --> 00:10:34,750 I'm not as genius as him because the whole course 240 00:10:34,750 --> 00:10:36,666 is called Blockchain and Money. 241 00:10:36,666 --> 00:10:38,610 [LAUGHTER] 242 00:10:38,610 --> 00:10:41,530 Oh, well. 243 00:10:41,530 --> 00:10:44,002 So what are decentralized applications, at least 244 00:10:44,002 --> 00:10:44,710 in this write-up? 245 00:10:48,140 --> 00:10:51,550 Anybody? 246 00:10:51,550 --> 00:10:52,100 Kelly? 247 00:10:52,100 --> 00:10:54,828 AUDIENCE: Basically, it just says it allows you [INAUDIBLE] 248 00:10:54,828 --> 00:10:57,170 but without the trust part. 249 00:10:57,170 --> 00:11:01,146 GARY GENSLER: So you don't need a centralized trusted party. 250 00:11:01,146 --> 00:11:05,080 That's what in the 1990s so many people tried and failed with, 251 00:11:05,080 --> 00:11:05,580 in a sense. 252 00:11:05,580 --> 00:11:10,020 And the internet became this way, in a decentralized way, 253 00:11:10,020 --> 00:11:12,030 to distribute packets of information, 254 00:11:12,030 --> 00:11:14,190 even though there is centralization on the internet 255 00:11:14,190 --> 00:11:18,450 as well, as we discussed Tuesday and so forth. 256 00:11:18,450 --> 00:11:21,740 But this decentralized application-- so 257 00:11:21,740 --> 00:11:25,690 in this context, he was using it broader 258 00:11:25,690 --> 00:11:27,830 than just something built on top of Ethereum 259 00:11:27,830 --> 00:11:29,030 through smart contracts. 260 00:11:29,030 --> 00:11:32,630 Because he even said Bitcoin was a decentralized application. 261 00:11:32,630 --> 00:11:34,640 So he was using it in the broader context, 262 00:11:34,640 --> 00:11:37,430 not in what some people would call dapps. 263 00:11:37,430 --> 00:11:42,170 But a new model for creating financing operating software. 264 00:11:42,170 --> 00:11:45,320 And this is what Christian Catalini wrote about as well, 265 00:11:45,320 --> 00:11:51,290 that it might be that there's a new way to finance a software 266 00:11:51,290 --> 00:11:52,850 development. 267 00:11:52,850 --> 00:11:55,250 Maybe raise some money for the file sharing 268 00:11:55,250 --> 00:11:58,230 before you have the file sharing. 269 00:11:58,230 --> 00:12:01,200 Might be an incentive system as well. 270 00:12:03,930 --> 00:12:06,690 And then he talked about two structural trade-offs 271 00:12:06,690 --> 00:12:09,310 from the design. 272 00:12:09,310 --> 00:12:14,870 Now, this blockchain design has a lot of complexity. 273 00:12:14,870 --> 00:12:18,640 It has a lot of additional cost, whether it's mining costs 274 00:12:18,640 --> 00:12:23,240 for proof of work or some other cost 275 00:12:23,240 --> 00:12:26,100 to basically secure the data. 276 00:12:26,100 --> 00:12:28,370 And so there has to be a trade-off. 277 00:12:28,370 --> 00:12:31,790 And what can pull you out of the minimalist end of this 278 00:12:31,790 --> 00:12:33,823 is whether there's some benefits as well. 279 00:12:33,823 --> 00:12:35,240 And we're going to chat about that 280 00:12:35,240 --> 00:12:38,600 in five or 10 minutes again. 281 00:12:38,600 --> 00:12:40,380 And then the censorship resistance. 282 00:12:40,380 --> 00:12:42,890 These are my words, not his. 283 00:12:42,890 --> 00:12:46,430 I really do think it's the individual censorship, 284 00:12:46,430 --> 00:12:49,100 but also, what I'll broadly call the market. 285 00:12:49,100 --> 00:12:52,610 Are there barriers to entry in the current system that 286 00:12:52,610 --> 00:12:57,290 are, in essence, censoring economic activity broadly, 287 00:12:57,290 --> 00:12:59,520 not individually? 288 00:12:59,520 --> 00:13:02,450 And I think, though, that was not how he defined 289 00:13:02,450 --> 00:13:03,620 it-- and I agree with Alin. 290 00:13:03,620 --> 00:13:05,828 It would have been great if he had written a few more 291 00:13:05,828 --> 00:13:07,940 paragraphs or another page. 292 00:13:07,940 --> 00:13:10,160 But I think it's not just the individual, 293 00:13:10,160 --> 00:13:15,610 it's some barriers to entry for whole market structures. 294 00:13:15,610 --> 00:13:18,580 And often, centralized institutions, 295 00:13:18,580 --> 00:13:22,750 because they have such a networking effect, 296 00:13:22,750 --> 00:13:27,580 it's hard to break in when somebody is basically 297 00:13:27,580 --> 00:13:33,010 the hub of a hub and spokes network for almost anything. 298 00:13:33,010 --> 00:13:36,490 How do you topple Facebook at this point in time? 299 00:13:36,490 --> 00:13:43,950 Because it can censor other market activity, as an example. 300 00:13:43,950 --> 00:13:45,570 So then we had the McKinsey report. 301 00:13:45,570 --> 00:13:51,390 Anything that people took from yet one more survey paper 302 00:13:51,390 --> 00:13:52,080 that you've had? 303 00:13:54,740 --> 00:13:56,630 Akira? 304 00:13:56,630 --> 00:14:00,980 It's just because I can see your name on the board there. 305 00:14:00,980 --> 00:14:06,340 AUDIENCE: From [INAUDIBLE],, I found the use case [INAUDIBLE] 306 00:14:06,340 --> 00:14:08,840 use case to be [INAUDIBLE]. 307 00:14:08,840 --> 00:14:11,870 And one thing I found interesting 308 00:14:11,870 --> 00:14:14,860 is the switching costs. 309 00:14:14,860 --> 00:14:16,980 Because the incumbent financial institutions 310 00:14:16,980 --> 00:14:24,726 have to think about not only the benefit of the cost reduction, 311 00:14:24,726 --> 00:14:27,882 but also, we have to think about [INAUDIBLE] costs, existing 312 00:14:27,882 --> 00:14:32,375 infrastructure, a lot of the affect on the people 313 00:14:32,375 --> 00:14:33,250 watching [INAUDIBLE]. 314 00:14:33,250 --> 00:14:35,590 GARY GENSLER: So Akira's raised a very important thing 315 00:14:35,590 --> 00:14:36,700 about switching costs. 316 00:14:36,700 --> 00:14:39,580 And we wrote about this, even though I'm not detailing it 317 00:14:39,580 --> 00:14:44,350 again, in the joint paper that Simon, Johnson, and Nihon, 318 00:14:44,350 --> 00:14:47,650 and Michael, and Joan, and I wrote. 319 00:14:47,650 --> 00:14:49,070 Now, this the last time-- 320 00:14:49,070 --> 00:14:52,415 it's the third time you had to come back and supposedly read 321 00:14:52,415 --> 00:14:54,040 a little bit more of the Geneva Report. 322 00:14:54,040 --> 00:14:57,270 But we talked there about the switching cost as well. 323 00:14:57,270 --> 00:14:59,320 And by the way, to Alin's question 324 00:14:59,320 --> 00:15:01,930 from Tuesday about interoperability, 325 00:15:01,930 --> 00:15:06,730 a lot of switching costs relate to interoperability. 326 00:15:06,730 --> 00:15:08,860 If you're an institution and you're 327 00:15:08,860 --> 00:15:12,400 thinking about putting in place some blockchain solution, 328 00:15:12,400 --> 00:15:17,440 permissioned or permissionless, but how does it communicate 329 00:15:17,440 --> 00:15:18,790 with the existing network? 330 00:15:18,790 --> 00:15:20,710 If the Australian stock exchange is 331 00:15:20,710 --> 00:15:24,460 putting in a new back end clearing and settlement system 332 00:15:24,460 --> 00:15:25,930 on a blockchain-- 333 00:15:25,930 --> 00:15:28,900 and they are and they're using digital asset holding-- 334 00:15:28,900 --> 00:15:33,100 they have to think about, well, all of our customers 335 00:15:33,100 --> 00:15:37,340 are currently communicating with a legacy database system. 336 00:15:37,340 --> 00:15:40,180 And now we're creating a blockchain system. 337 00:15:40,180 --> 00:15:42,400 How does that communicate? 338 00:15:42,400 --> 00:15:46,120 How does the old system literally move information over 339 00:15:46,120 --> 00:15:47,660 into the new system? 340 00:15:47,660 --> 00:15:53,830 And all of those from a business perspective 341 00:15:53,830 --> 00:15:58,420 are a question of how do you operate or interoperate 342 00:15:58,420 --> 00:16:03,410 with the legacy system and all the network. 343 00:16:03,410 --> 00:16:05,470 Because any blockchain solution that you're 344 00:16:05,470 --> 00:16:08,660 going to come up with is likely to be 345 00:16:08,660 --> 00:16:12,710 replacing some other economic activity, and likely, 346 00:16:12,710 --> 00:16:15,380 you're not going to try to replace the entire network. 347 00:16:15,380 --> 00:16:17,695 You're going to be surgical. 348 00:16:17,695 --> 00:16:18,695 You're going to decide-- 349 00:16:21,247 --> 00:16:22,830 I'm not trying to give anything, Eric, 350 00:16:22,830 --> 00:16:26,150 but can I say what you all are talking about it 351 00:16:26,150 --> 00:16:29,300 in your final project? 352 00:16:29,300 --> 00:16:31,070 You have intellectual property, so I don't 353 00:16:31,070 --> 00:16:32,180 know if you wanted to share it. 354 00:16:32,180 --> 00:16:33,440 But do you want to say what-- 355 00:16:33,440 --> 00:16:35,330 AUDIENCE: Could you elaborate a little more? 356 00:16:35,330 --> 00:16:38,450 In the point we were discussing, I was kind of lost. 357 00:16:38,450 --> 00:16:40,890 GARY GENSLER: Well, so Eric's group-- 358 00:16:40,890 --> 00:16:45,290 and Brotish, and Catalina, and Ross-- 359 00:16:45,290 --> 00:16:49,160 are talking about maybe doing a permissioned blockchain 360 00:16:49,160 --> 00:16:53,280 application for credit reports. 361 00:16:53,280 --> 00:16:55,350 Not credit reports for the commercial side, 362 00:16:55,350 --> 00:16:57,240 but consumer credit. 363 00:16:57,240 --> 00:17:00,880 Basically, an-- uh-oh, Alf. 364 00:17:00,880 --> 00:17:02,100 Are you competing with them? 365 00:17:02,100 --> 00:17:02,610 Did I just-- 366 00:17:02,610 --> 00:17:03,590 AUDIENCE: No, no. 367 00:17:03,590 --> 00:17:04,660 GARY GENSLER: OK. 368 00:17:04,660 --> 00:17:05,160 All right. 369 00:17:05,160 --> 00:17:07,950 Maybe you are, so I shouldn't say much more. 370 00:17:07,950 --> 00:17:08,950 No, it's all right. 371 00:17:08,950 --> 00:17:12,295 But I won't say anything more. 372 00:17:12,295 --> 00:17:14,520 AUDIENCE: Maybe it would be interesting to point out 373 00:17:14,520 --> 00:17:17,069 the fundamental-- 374 00:17:17,069 --> 00:17:19,440 or one of the interesting things that comes up 375 00:17:19,440 --> 00:17:21,810 from the discussion of the topic we're engaging 376 00:17:21,810 --> 00:17:27,839 is how to approach a blockchain solution 377 00:17:27,839 --> 00:17:33,255 from an entrepreneurial perspective. 378 00:17:36,720 --> 00:17:40,107 Going back to the letter to-- 379 00:17:40,107 --> 00:17:41,190 GARY GENSLER: Jamie Dimon. 380 00:17:41,190 --> 00:17:45,590 AUDIENCE: To Jamie Dimon reading, 381 00:17:45,590 --> 00:17:47,840 we're trying to get rid of intermediaries 382 00:17:47,840 --> 00:17:52,990 or getting economic rents with that differing kind of paradigm 383 00:17:52,990 --> 00:17:58,240 that justifies the costs of using a blockchain approach 384 00:17:58,240 --> 00:18:01,650 to promote [INAUDIBLE]. 385 00:18:01,650 --> 00:18:04,270 But from an entrepreneurial point of view, 386 00:18:04,270 --> 00:18:08,680 you would end up trying to get some benefit from that kind 387 00:18:08,680 --> 00:18:09,490 of implementation. 388 00:18:09,490 --> 00:18:12,190 So you don't want to replace intermediaries 389 00:18:12,190 --> 00:18:14,360 with another intermediary. 390 00:18:14,360 --> 00:18:17,020 So you would have to be really creative in terms 391 00:18:17,020 --> 00:18:21,460 of how to come up with really lowering costs and gaining 392 00:18:21,460 --> 00:18:23,360 some profits in the process. 393 00:18:23,360 --> 00:18:26,350 GARY GENSLER: So Eric's raising that if you're just 394 00:18:26,350 --> 00:18:29,470 replacing one intermediary with another intermediary, 395 00:18:29,470 --> 00:18:31,690 will you be successful as an entrepreneur? 396 00:18:31,690 --> 00:18:35,710 Yes, if you're providing a better service or a lower cost 397 00:18:35,710 --> 00:18:39,590 structure, lower economic rents. 398 00:18:39,590 --> 00:18:43,190 So I wouldn't doubt yourself. 399 00:18:43,190 --> 00:18:45,340 I would say, we might be able to do it. 400 00:18:45,340 --> 00:18:48,470 In your use case-- and apologies, 401 00:18:48,470 --> 00:18:49,670 maybe it's multiple use-- 402 00:18:49,670 --> 00:18:53,070 maybe it's two or three groups thinking about it. 403 00:18:53,070 --> 00:18:56,310 There's a lot of market power right now, and thus a lot 404 00:18:56,310 --> 00:18:59,490 of economic rents, but a lot of market power and economic rents 405 00:18:59,490 --> 00:19:01,320 and credit reporting. 406 00:19:01,320 --> 00:19:05,370 Equifax, First Union, and others earn a certain fees 407 00:19:05,370 --> 00:19:08,760 for all of those FICA scores. 408 00:19:08,760 --> 00:19:11,010 In the US, we use this. 409 00:19:11,010 --> 00:19:12,480 If you're getting a job, somebody 410 00:19:12,480 --> 00:19:14,500 runs a credit report on you. 411 00:19:14,500 --> 00:19:18,830 If you're buying an automobile and taking out an auto loan. 412 00:19:18,830 --> 00:19:20,760 It feels like each of us probably 413 00:19:20,760 --> 00:19:24,900 have a credit report pulled on us multiple times a year. 414 00:19:24,900 --> 00:19:28,140 Sometimes we don't even know it. 415 00:19:28,140 --> 00:19:30,360 But back to interoperability and switching 416 00:19:30,360 --> 00:19:33,100 costs, just to pull it all together. 417 00:19:33,100 --> 00:19:37,830 How could you convince the commercial banks to use-- 418 00:19:37,830 --> 00:19:40,020 I'll call it Eric's project, it's 419 00:19:40,020 --> 00:19:44,940 four people, but how could you convince all the banks to use 420 00:19:44,940 --> 00:19:49,380 their project and how about all the auto dealerships 421 00:19:49,380 --> 00:19:53,010 on the other side who right now are very comfortable using 422 00:19:53,010 --> 00:19:54,750 First Union? 423 00:19:54,750 --> 00:19:57,480 Everybody's probably paying a little bit extra. 424 00:19:57,480 --> 00:20:02,330 And there's a fat market power, economic rents. 425 00:20:02,330 --> 00:20:04,260 And so that's part of the switching costs 426 00:20:04,260 --> 00:20:06,750 that Akira mentioned. 427 00:20:06,750 --> 00:20:09,960 And to the interoperability, I believe 428 00:20:09,960 --> 00:20:12,840 they'll be more successful if they 429 00:20:12,840 --> 00:20:18,150 don't try to completely change how data flows, 430 00:20:18,150 --> 00:20:21,150 that at some point you need to know 431 00:20:21,150 --> 00:20:24,180 that you can't be successful by biting off 432 00:20:24,180 --> 00:20:26,530 too much all at once. 433 00:20:26,530 --> 00:20:28,997 I'm sorry, Catalina. 434 00:20:28,997 --> 00:20:31,080 AUDIENCE: I was going to say that, precisely, that 435 00:20:31,080 --> 00:20:32,590 is one of the forms of blockchain 436 00:20:32,590 --> 00:20:33,780 he raised in the paper. 437 00:20:33,780 --> 00:20:39,255 And it's not changing one intermediary for another. 438 00:20:39,255 --> 00:20:41,130 Financially, what they say is that he's going 439 00:20:41,130 --> 00:20:43,490 to be more successful for-- 440 00:20:43,490 --> 00:20:46,825 blockchain's going to be more successful in the permissioned 441 00:20:46,825 --> 00:20:49,740 than in the permissionless because the incumbents are more 442 00:20:49,740 --> 00:20:51,680 willing to work on the permissioned 443 00:20:51,680 --> 00:20:54,423 than with the permissionless. 444 00:20:54,423 --> 00:20:55,840 GARY GENSLER: In essence, Catalina 445 00:20:55,840 --> 00:21:00,210 is saying you might be more successful not trying 446 00:21:00,210 --> 00:21:03,390 to disrupt the entire community, whatever 447 00:21:03,390 --> 00:21:06,750 that community in some ecosystem, some business 448 00:21:06,750 --> 00:21:10,650 community, and do it in a permissioned 449 00:21:10,650 --> 00:21:11,700 versus permissionless. 450 00:21:11,700 --> 00:21:14,190 But we're going to play with that a little longer. 451 00:21:14,190 --> 00:21:14,960 Kelly. 452 00:21:14,960 --> 00:21:18,160 AUDIENCE: I feel like they addressed the timeline of where 453 00:21:18,160 --> 00:21:19,440 the strategic value is. 454 00:21:19,440 --> 00:21:21,420 For example, they said, currently 455 00:21:21,420 --> 00:21:25,260 the best way to reap the benefits from blockchain 456 00:21:25,260 --> 00:21:27,450 are to focus on this, this, and this. 457 00:21:27,450 --> 00:21:30,930 And then they say, however, feasibility at scale, they say, 458 00:21:30,930 --> 00:21:33,067 is probably three to five years away, 459 00:21:33,067 --> 00:21:34,400 exactly what you've been saying. 460 00:21:34,400 --> 00:21:34,900 So-- 461 00:21:34,900 --> 00:21:37,410 GARY GENSLER: Oh, my god. 462 00:21:37,410 --> 00:21:40,120 We could both be wrong. 463 00:21:40,120 --> 00:21:41,650 All right. 464 00:21:41,650 --> 00:21:42,535 Jack. 465 00:21:42,535 --> 00:21:44,410 AUDIENCE: Something else I found interesting. 466 00:21:44,410 --> 00:21:47,590 There's a lot of talk right now about robotics and supply chain 467 00:21:47,590 --> 00:21:48,770 using blockchain. 468 00:21:48,770 --> 00:21:51,472 And the McKinsey report kind of touched on that. 469 00:21:51,472 --> 00:21:52,930 No matter how secure the blockchain 470 00:21:52,930 --> 00:21:57,250 is, when you start involving actual real products 471 00:21:57,250 --> 00:22:00,160 and assets, it could be completely corrupted, 472 00:22:00,160 --> 00:22:01,920 even if the blockchain folds. 473 00:22:01,920 --> 00:22:02,753 GARY GENSLER: Right. 474 00:22:02,753 --> 00:22:04,180 And also, it's interesting-- 475 00:22:04,180 --> 00:22:07,330 I think, Jack, if I can pick further on that. 476 00:22:07,330 --> 00:22:09,220 When you have something on the blockchain, 477 00:22:09,220 --> 00:22:12,280 even if it arguably should be there, 478 00:22:12,280 --> 00:22:13,900 it's a property right and so forth, 479 00:22:13,900 --> 00:22:19,510 but it's an off-chain asset, it's not just a digital asset, 480 00:22:19,510 --> 00:22:22,180 and secured and verified on a blockchain. 481 00:22:22,180 --> 00:22:26,410 But in supply chain, it's often an off-chain asset, and often 482 00:22:26,410 --> 00:22:28,240 a physical asset. 483 00:22:28,240 --> 00:22:30,757 Then there's additional challenges. 484 00:22:30,757 --> 00:22:32,840 I think there's some really interesting use cases, 485 00:22:32,840 --> 00:22:36,610 but it's a little bit of a different place. 486 00:22:36,610 --> 00:22:37,180 James. 487 00:22:37,180 --> 00:22:38,305 And then I'm going to-- no? 488 00:22:40,970 --> 00:22:44,960 So McKinsey defines a blockchain. 489 00:22:44,960 --> 00:22:47,450 And they point out it doesn't have 490 00:22:47,450 --> 00:22:51,440 to necessarily disintermediate to generate value. 491 00:22:51,440 --> 00:22:54,890 I thought that was an interesting intuition. 492 00:22:54,890 --> 00:22:58,310 Don't always think that you have to disintermediate 493 00:22:58,310 --> 00:22:59,150 to generate value. 494 00:22:59,150 --> 00:23:00,942 They might be right or they might be wrong. 495 00:23:00,942 --> 00:23:03,170 But you might actually still end up 496 00:23:03,170 --> 00:23:05,660 with something that's kind of-- 497 00:23:05,660 --> 00:23:09,245 as-- did Catalina say it? 498 00:23:09,245 --> 00:23:10,620 I can't remember now who said it, 499 00:23:10,620 --> 00:23:13,310 but they also said they gave a timescale. 500 00:23:13,310 --> 00:23:15,620 A short term value might be predominantly-- 501 00:23:15,620 --> 00:23:20,360 Kelly did, sorry-- three to five years away. 502 00:23:20,360 --> 00:23:23,600 How to capture value. 503 00:23:23,600 --> 00:23:27,550 Be pragmatic and skeptical, but you 504 00:23:27,550 --> 00:23:30,458 need to go down to a granular level. 505 00:23:30,458 --> 00:23:32,000 And that's also what I'm going to ask 506 00:23:32,000 --> 00:23:34,460 you to do in the final paper, is really 507 00:23:34,460 --> 00:23:37,640 to challenge yourself, challenge your group 508 00:23:37,640 --> 00:23:40,400 to go down to the granular level and say, wait, 509 00:23:40,400 --> 00:23:43,040 what transactions are we actually going 510 00:23:43,040 --> 00:23:44,990 to change the flow? 511 00:23:44,990 --> 00:23:48,050 What ledger system might we shift? 512 00:23:48,050 --> 00:23:50,750 And even though it's the back office-- 513 00:23:50,750 --> 00:23:52,640 and to some people, you might say, oh, 514 00:23:52,640 --> 00:23:55,280 my god, that's kind of the boring side of business. 515 00:23:55,280 --> 00:23:57,200 But that's really what blockchain is about. 516 00:23:57,200 --> 00:24:01,920 It's a database and ledger management system also. 517 00:24:01,920 --> 00:24:04,130 And to be successful, you have to get down 518 00:24:04,130 --> 00:24:08,960 to that granular level and figure it out 519 00:24:08,960 --> 00:24:11,820 in terms of whatever that current ecosystem is 520 00:24:11,820 --> 00:24:13,715 or the regulatory system. 521 00:24:17,180 --> 00:24:22,390 They said it's particularly valuable in low trust 522 00:24:22,390 --> 00:24:24,070 environments. 523 00:24:24,070 --> 00:24:26,380 And that could be where currently we 524 00:24:26,380 --> 00:24:30,040 can't trade directly and you've created a new way to do peer 525 00:24:30,040 --> 00:24:34,150 to peer, or where there's not currently 526 00:24:34,150 --> 00:24:38,150 a central intermediary. 527 00:24:38,150 --> 00:24:39,290 You can go right at it. 528 00:24:39,290 --> 00:24:42,260 You could be like this bold group that says, 529 00:24:42,260 --> 00:24:46,940 I want to take on Transunion, I want to take on Equifax. 530 00:24:46,940 --> 00:24:51,110 That's something which is a central intermediary right now. 531 00:24:51,110 --> 00:24:54,880 But it is correct, we're not really trading directly. 532 00:24:54,880 --> 00:25:01,790 It's not like when I walk into the auto shop or to buy a car 533 00:25:01,790 --> 00:25:04,280 and I'm going to take a loan out that I can transact. 534 00:25:04,280 --> 00:25:07,470 They want to validate that I can pay back that darn loan. 535 00:25:10,200 --> 00:25:10,880 Yes? 536 00:25:10,880 --> 00:25:14,220 AUDIENCE: Just two points, particularly the fact where 537 00:25:14,220 --> 00:25:15,075 your-- 538 00:25:15,075 --> 00:25:16,810 well, you can't trade directly now, 539 00:25:16,810 --> 00:25:18,780 but there's a lot of trade happening 540 00:25:18,780 --> 00:25:21,310 where you don't have an intermediary, 541 00:25:21,310 --> 00:25:23,825 but some sort of intermediary would help. 542 00:25:23,825 --> 00:25:26,723 I couldn't help but think of ag markets in very-- 543 00:25:26,723 --> 00:25:28,765 GARY GENSLER: You're saying ag, like agriculture? 544 00:25:28,765 --> 00:25:32,780 AUDIENCE: Agriculture markets in very remote parts of the world, 545 00:25:32,780 --> 00:25:34,740 where actually being networked. 546 00:25:34,740 --> 00:25:37,320 And precisely, there's no trust. 547 00:25:37,320 --> 00:25:39,210 That kind of environment exists. 548 00:25:39,210 --> 00:25:44,190 So a good use case for that, extreme remote parts 549 00:25:44,190 --> 00:25:47,640 of the world where internet is there, but not necessarily 550 00:25:47,640 --> 00:25:50,087 financial institutions or regulatory environments. 551 00:25:50,087 --> 00:25:50,920 GARY GENSLER: Right. 552 00:25:50,920 --> 00:25:53,310 And there are many parts, particularly in the third world 553 00:25:53,310 --> 00:25:54,870 that are still unbanked. 554 00:25:54,870 --> 00:25:59,040 I've said this statistic before, but the World Bank report 555 00:25:59,040 --> 00:26:05,310 on banking in 2017, half of sub-Saharan African 556 00:26:05,310 --> 00:26:08,650 was still unbanked. 557 00:26:08,650 --> 00:26:12,280 But half of that half has mobile phones. 558 00:26:12,280 --> 00:26:17,400 So it's just-- those gaps might change, but then it will move. 559 00:26:17,400 --> 00:26:19,400 Yes, they might have a banking account, 560 00:26:19,400 --> 00:26:21,280 but it would be not something they can really 561 00:26:21,280 --> 00:26:24,970 do with agriculture. 562 00:26:24,970 --> 00:26:26,765 So two opportunities-- and I'm going 563 00:26:26,765 --> 00:26:28,690 to come to the question-- is like, 564 00:26:28,690 --> 00:26:31,180 do you go head-on against a central intermediary 565 00:26:31,180 --> 00:26:33,220 and use this technology to do it better? 566 00:26:33,220 --> 00:26:34,658 [PHONE RINGING] 567 00:26:34,658 --> 00:26:35,200 Look at that. 568 00:26:35,200 --> 00:26:36,310 I should turn this off. 569 00:26:39,640 --> 00:26:42,400 Head-on against a central intermediary? 570 00:26:42,400 --> 00:26:47,020 Or some market structure that has no really core 571 00:26:47,020 --> 00:26:52,420 intermediary, but you can build a better trustless 572 00:26:52,420 --> 00:26:54,700 peer to peer network? 573 00:26:54,700 --> 00:26:57,160 AUDIENCE: Yeah, [INAUDIBLE] that's 574 00:26:57,160 --> 00:27:00,390 when it comes to private permission. 575 00:27:00,390 --> 00:27:04,290 Architectural blockchain, it's highly likely 576 00:27:04,290 --> 00:27:05,630 to be more scalable. 577 00:27:05,630 --> 00:27:07,480 Why is that? 578 00:27:07,480 --> 00:27:10,510 GARY GENSLER: Highly likely to be more scalable, yes. 579 00:27:10,510 --> 00:27:16,600 So permissioned blockchains, they've made a trade-off. 580 00:27:16,600 --> 00:27:19,570 So Nakamoto's paper comes out. 581 00:27:19,570 --> 00:27:21,970 For a few years, people don't really notice it. 582 00:27:21,970 --> 00:27:26,830 But then it starts to get some lift 583 00:27:26,830 --> 00:27:31,030 and folks are looking at it in 2013 to 2015. 584 00:27:31,030 --> 00:27:34,120 And the financial industry starts to think about this 585 00:27:34,120 --> 00:27:37,928 and says, this actually is pretty interesting. 586 00:27:37,928 --> 00:27:39,470 We're not sure what it's going to do, 587 00:27:39,470 --> 00:27:40,960 but we're pretty interested. 588 00:27:40,960 --> 00:27:43,870 And they get attracted to it because it might be a way 589 00:27:43,870 --> 00:27:47,050 to move data around, it might be able to literally lower 590 00:27:47,050 --> 00:27:49,720 the costs of the back office. 591 00:27:49,720 --> 00:27:54,220 Many banks keeping the same set of records-- 592 00:27:54,220 --> 00:27:55,910 I have an equity trading business, 593 00:27:55,910 --> 00:27:57,430 you have an equity trading business. 594 00:27:57,430 --> 00:28:01,880 We both have to confirm and correct those records. 595 00:28:01,880 --> 00:28:06,050 But we keep full databases. 596 00:28:06,050 --> 00:28:10,410 What they found attractive was real. 597 00:28:10,410 --> 00:28:12,320 But then remember, Bitcoin can only 598 00:28:12,320 --> 00:28:15,560 do seven transactions a second. 599 00:28:15,560 --> 00:28:22,610 So all of those challenges of scalability, privacy, security 600 00:28:22,610 --> 00:28:24,350 led them to think about, was there 601 00:28:24,350 --> 00:28:27,380 a way that we could literally toss 602 00:28:27,380 --> 00:28:31,490 some of the technical things over the side? 603 00:28:31,490 --> 00:28:35,100 And most of them thought, we don't need a native token. 604 00:28:35,100 --> 00:28:38,260 We don't need an incentive structure. 605 00:28:38,260 --> 00:28:43,490 And if we limit the number of nodes in the network, 606 00:28:43,490 --> 00:28:45,650 we can increase scalability and we 607 00:28:45,650 --> 00:28:47,740 can get rid of proof of work. 608 00:28:47,740 --> 00:28:53,940 So my narrative might not matter, 609 00:28:53,940 --> 00:28:55,440 but it's important to understand it. 610 00:28:55,440 --> 00:28:58,800 The permissionless thing comes along, and then they say, 611 00:28:58,800 --> 00:28:59,850 it's not scalable. 612 00:28:59,850 --> 00:29:04,140 What can we toss overboard and make it scalable? 613 00:29:04,140 --> 00:29:05,910 And what they tossed overboard-- 614 00:29:05,910 --> 00:29:08,410 they didn't even feel they needed the native token, 615 00:29:08,410 --> 00:29:12,270 and they tossed overboard proof of work, consensus, 616 00:29:12,270 --> 00:29:15,070 and had a club deal. 617 00:29:15,070 --> 00:29:17,920 In the Australian stock exchange example, 618 00:29:17,920 --> 00:29:19,990 they might only have two or three nodes. 619 00:29:19,990 --> 00:29:23,090 But in some of these, there's 15 or 20. 620 00:29:23,090 --> 00:29:25,640 So why is it more scalable? 621 00:29:25,640 --> 00:29:30,290 Fewer nodes and much more efficient consensus mechanisms. 622 00:29:30,290 --> 00:29:32,630 But the trade-off is you better damn well 623 00:29:32,630 --> 00:29:36,300 trust now those 15 or 20 nodes. 624 00:29:36,300 --> 00:29:39,400 And when we talk about 51% attacks, 625 00:29:39,400 --> 00:29:42,270 you could revise a whole permissioned blockchain 626 00:29:42,270 --> 00:29:46,480 if those 15 parties wanted to. 627 00:29:46,480 --> 00:29:49,340 Now, some people say, well, you could just do a hash function. 628 00:29:49,340 --> 00:29:51,040 Remember that whole hashing. 629 00:29:51,040 --> 00:29:54,280 You could do a hash function of a whole permissioned blockchain 630 00:29:54,280 --> 00:29:56,650 once a day, once a week, once an hour 631 00:29:56,650 --> 00:29:58,400 and store it somewhere else. 632 00:29:58,400 --> 00:30:02,620 In fact, you could store the hash of all of that on Bitcoin. 633 00:30:02,620 --> 00:30:04,570 You could use a ledger like Bitcoin 634 00:30:04,570 --> 00:30:07,360 and just store it there. 635 00:30:07,360 --> 00:30:10,385 So there's some feedback loops. 636 00:30:10,385 --> 00:30:11,010 Does that help? 637 00:30:11,010 --> 00:30:12,800 Give you a sense? 638 00:30:12,800 --> 00:30:14,920 AUDIENCE: I was also just going to build on that. 639 00:30:14,920 --> 00:30:17,790 So the power of a permissioned blockchain 640 00:30:17,790 --> 00:30:22,450 is that, unlike Ethereum or Bitcoin, it's not public. 641 00:30:22,450 --> 00:30:24,055 So there is no public ledger. 642 00:30:24,055 --> 00:30:26,530 You can make it public, but it's permissioned, not 643 00:30:26,530 --> 00:30:27,400 permissionless. 644 00:30:27,400 --> 00:30:31,490 Because you need to have trusted entities in this network. 645 00:30:31,490 --> 00:30:35,670 So this can also be, you could say, 646 00:30:35,670 --> 00:30:39,670 a feature, not a bug when it comes to scaling. 647 00:30:39,670 --> 00:30:43,780 Because essentially, you control the scale. 648 00:30:43,780 --> 00:30:46,570 So you control these nodes. 649 00:30:46,570 --> 00:30:49,060 And then like mentioned, these nodes 650 00:30:49,060 --> 00:30:52,120 are only relevant because they are 651 00:30:52,120 --> 00:30:55,600 trusted pieces of this puzzle. 652 00:30:55,600 --> 00:30:56,450 Yeah. 653 00:30:56,450 --> 00:30:57,760 AUDIENCE: Right. 654 00:30:57,760 --> 00:31:00,680 I think it's easier to get scale, 655 00:31:00,680 --> 00:31:04,780 but there is not a large enough market 656 00:31:04,780 --> 00:31:08,080 to actually make it scale or make 657 00:31:08,080 --> 00:31:10,990 it reach to the critical mass. 658 00:31:10,990 --> 00:31:15,220 GARY GENSLER: So are you worried that there's not 659 00:31:15,220 --> 00:31:18,660 enough to get to the critical mass? 660 00:31:18,660 --> 00:31:20,510 Now, I'm not sure-- 661 00:31:20,510 --> 00:31:27,620 so a permissionless blockchain and a permissioned blockchain 662 00:31:27,620 --> 00:31:30,380 both can be open to public users. 663 00:31:30,380 --> 00:31:36,350 You can have millions of people use a permissioned blockchain, 664 00:31:36,350 --> 00:31:41,290 but only 15 have the right to amend the records. 665 00:31:41,290 --> 00:31:43,490 So one might think of who has a right 666 00:31:43,490 --> 00:31:46,760 to read it, who has a right to write on it. 667 00:31:46,760 --> 00:31:48,950 Read, write. 668 00:31:48,950 --> 00:31:54,560 But also, absent even that, it might provide a service, 669 00:31:54,560 --> 00:31:57,920 like file sharing or some service. 670 00:31:57,920 --> 00:31:59,870 There might even be a third list of who 671 00:31:59,870 --> 00:32:01,670 has a right to use the service. 672 00:32:01,670 --> 00:32:04,070 Now, usually to use a service, you actually 673 00:32:04,070 --> 00:32:05,180 have to read the data. 674 00:32:05,180 --> 00:32:10,057 But you could partition and say, how much can be read as well. 675 00:32:10,057 --> 00:32:11,640 Yes, and remind me of your first name. 676 00:32:11,640 --> 00:32:11,950 I'm sorry. 677 00:32:11,950 --> 00:32:12,970 AUDIENCE: [INAUDIBLE] 678 00:32:12,970 --> 00:32:13,190 GARY GENSLER: [INAUDIBLE] 679 00:32:13,190 --> 00:32:13,815 AUDIENCE: Yeah. 680 00:32:17,040 --> 00:32:18,570 I really appreciate your perspective 681 00:32:18,570 --> 00:32:22,213 from how you envision [INAUDIBLE] 682 00:32:22,213 --> 00:32:23,130 permissioned projects. 683 00:32:23,130 --> 00:32:24,630 GARY GENSLER: The pathway, you said? 684 00:32:24,630 --> 00:32:27,850 AUDIENCE: [INAUDIBLE] in terms of its relationship with 685 00:32:27,850 --> 00:32:29,640 [INAUDIBLE]. 686 00:32:29,640 --> 00:32:35,040 Do you think it would pave the way 687 00:32:35,040 --> 00:32:38,350 towards the true decentralization eventually? 688 00:32:38,350 --> 00:32:45,680 Or you were just [INAUDIBLE] Or you would just 689 00:32:45,680 --> 00:32:50,240 become irrelevant just as the intranet compared to internet. 690 00:32:50,240 --> 00:32:53,660 GARY GENSLER: So [INAUDIBLE] question is unanswerable, 691 00:32:53,660 --> 00:32:57,860 but I'll try to give my opinion in 2018, and then we'll see. 692 00:32:57,860 --> 00:33:03,120 And it's being recorded, so we'll see how I'll do. 693 00:33:03,120 --> 00:33:04,910 But the question really is, what's 694 00:33:04,910 --> 00:33:07,760 the relationship for permissioned and permissionless 695 00:33:07,760 --> 00:33:09,680 blockchains? 696 00:33:09,680 --> 00:33:12,080 And what is it going to be longer term? 697 00:33:12,080 --> 00:33:16,790 I think today in 2018, because of the scalability issues 698 00:33:16,790 --> 00:33:19,040 and some of the privacy and security 699 00:33:19,040 --> 00:33:24,440 issues, the dominant place that the investment will 700 00:33:24,440 --> 00:33:27,350 be is on the permissioned side, especially 701 00:33:27,350 --> 00:33:30,710 with what's called enterprise applications, the banks, 702 00:33:30,710 --> 00:33:32,600 the large institutions. 703 00:33:32,600 --> 00:33:37,580 There's not enough scalability because 1,000 users a day 704 00:33:37,580 --> 00:33:40,940 or 1,500 users a day on the big apps 705 00:33:40,940 --> 00:33:43,100 right now on the Ethereum network, that's just not 706 00:33:43,100 --> 00:33:50,390 enough beyond gaming sites and some small hobbyist type 707 00:33:50,390 --> 00:33:53,990 things, like CryptoKitties. 708 00:33:53,990 --> 00:33:56,390 It's fun, it's interesting, but it's not 709 00:33:56,390 --> 00:34:01,111 truly scalable to 7 billion people living around the globe. 710 00:34:03,880 --> 00:34:07,270 But I wouldn't-- and I've said this to some IBM folks. 711 00:34:07,270 --> 00:34:13,000 And they have 1,000 people at IBM in the hyper ledger team. 712 00:34:13,000 --> 00:34:15,400 Anybody who is a blockchain minimalist-- 713 00:34:15,400 --> 00:34:17,002 you haven't heard this number? 714 00:34:17,002 --> 00:34:18,556 AUDIENCE: I've heard 2,500. 715 00:34:18,556 --> 00:34:19,389 GARY GENSLER: 2,500? 716 00:34:19,389 --> 00:34:19,889 All right. 717 00:34:19,889 --> 00:34:22,000 So it's more than I thought. 718 00:34:22,000 --> 00:34:23,260 So 1,000 to 2,500. 719 00:34:23,260 --> 00:34:26,260 I knew it was more than 1,000. 720 00:34:26,260 --> 00:34:27,969 IBM is throwing a lot of money there. 721 00:34:27,969 --> 00:34:32,820 Now, they might be doing that to protect their risk. 722 00:34:32,820 --> 00:34:36,150 They might be investing even if all those 1,000 people 723 00:34:36,150 --> 00:34:38,850 ultimately produce nothing. 724 00:34:38,850 --> 00:34:43,050 But I think that five years from now or seven years from now, 725 00:34:43,050 --> 00:34:45,670 the gap will close. 726 00:34:45,670 --> 00:34:48,219 I think that permissionless systems will always 727 00:34:48,219 --> 00:34:52,270 have a lot of challenge about governance, collective action. 728 00:34:52,270 --> 00:34:55,300 How you have collective action is not just 729 00:34:55,300 --> 00:34:59,800 a matter of technical and coding and smart scientist 730 00:34:59,800 --> 00:35:02,995 at MIT and elsewhere. 731 00:35:02,995 --> 00:35:04,870 I think that's going to still be a challenge, 732 00:35:04,870 --> 00:35:07,080 and there will be trade-offs. 733 00:35:07,080 --> 00:35:09,240 So it might be that they coexist, 734 00:35:09,240 --> 00:35:11,820 but I think the gap will close. 735 00:35:11,820 --> 00:35:15,150 And you said, well, will it be like the intranet went away 736 00:35:15,150 --> 00:35:16,410 and the internet was there. 737 00:35:16,410 --> 00:35:18,870 There are some people that believe that. 738 00:35:18,870 --> 00:35:20,820 And that might be-- 739 00:35:20,820 --> 00:35:21,930 I'm probably not there. 740 00:35:21,930 --> 00:35:24,000 I'm probably not all the way there. 741 00:35:24,000 --> 00:35:26,790 But I think when the gap closes, some 742 00:35:26,790 --> 00:35:29,940 of these permissioned systems will not be as necessary. 743 00:35:29,940 --> 00:35:31,800 Zero knowledge proofs, other ways 744 00:35:31,800 --> 00:35:35,190 that come along that make it more usable. 745 00:35:35,190 --> 00:35:36,352 Yes, Zann. 746 00:35:36,352 --> 00:35:37,560 AUDIENCE: Totally understood. 747 00:35:37,560 --> 00:35:41,370 I'm a little bit skeptical on the permissioned blockchains. 748 00:35:41,370 --> 00:35:42,900 One of the-- 749 00:35:42,900 --> 00:35:44,280 GARY GENSLER: Are you skeptical-- 750 00:35:44,280 --> 00:35:45,660 if I can interrupt you-- 751 00:35:45,660 --> 00:35:48,150 and thus it moves you over to traditional databases? 752 00:35:48,150 --> 00:35:51,270 Or skeptical and you're moved over to permissionless open? 753 00:35:51,270 --> 00:35:53,420 So I don't know, which way is it pushing you? 754 00:35:53,420 --> 00:35:54,920 AUDIENCE: Depending on the use case. 755 00:35:54,920 --> 00:35:57,510 For enterprises, I think traditional databases 756 00:35:57,510 --> 00:35:59,230 for the enterprise use case. 757 00:35:59,230 --> 00:36:02,790 Reading here, this line killed me. 758 00:36:02,790 --> 00:36:05,970 It said, "Accenture developed immutable blockchain 759 00:36:05,970 --> 00:36:07,890 in which the content of individual blocks 760 00:36:07,890 --> 00:36:10,440 can be modified." 761 00:36:10,440 --> 00:36:14,040 It's a ridiculous statement considering the definition 762 00:36:14,040 --> 00:36:16,680 of blockchain is anti-- 763 00:36:16,680 --> 00:36:18,310 this is the antithesis of basically-- 764 00:36:18,310 --> 00:36:19,143 GARY GENSLER: Right. 765 00:36:19,143 --> 00:36:22,110 So Accenture created an immutable blockchain that 766 00:36:22,110 --> 00:36:27,300 isn't immutable, is what you're saying, it can be amended. 767 00:36:27,300 --> 00:36:31,290 So it's a backdoor I can change. 768 00:36:31,290 --> 00:36:35,700 And yet, the economics of it is that a lot of times, people 769 00:36:35,700 --> 00:36:37,830 do want a backdoor. 770 00:36:37,830 --> 00:36:41,580 And the challenges-- and these are the weaknesses as well, 771 00:36:41,580 --> 00:36:43,800 as Nouriel Roubini says. 772 00:36:43,800 --> 00:36:46,140 Why would you create a system where if you forget 773 00:36:46,140 --> 00:36:48,840 your private key, lose your private key, 774 00:36:48,840 --> 00:36:52,110 or the cave collapses-- that you're hiding the private key 775 00:36:52,110 --> 00:36:54,960 in a cave and maybe it floods-- 776 00:36:54,960 --> 00:36:56,340 that you've lost your assets? 777 00:36:56,340 --> 00:36:57,548 Wouldn't you want a backdoor? 778 00:37:02,780 --> 00:37:06,120 But I respect-- you would say this isn't even a blockchain. 779 00:37:06,120 --> 00:37:08,670 I'm trying to teach this class in a little bit 780 00:37:08,670 --> 00:37:12,690 neutral between permissionless and permissioned blockchains. 781 00:37:12,690 --> 00:37:14,640 But you need something to get out 782 00:37:14,640 --> 00:37:16,723 of the traditional databases. 783 00:37:16,723 --> 00:37:19,140 And I've chosen, at least for purposes of this class, that 784 00:37:19,140 --> 00:37:21,120 means you've got to have an append-only log, 785 00:37:21,120 --> 00:37:23,430 you have to have time stamping. 786 00:37:23,430 --> 00:37:27,690 So the basis of multiple people writing to the ledger, 787 00:37:27,690 --> 00:37:30,480 even if it's only three or five. 788 00:37:30,480 --> 00:37:35,105 But these are not well-defined vocabulary terms. 789 00:37:35,105 --> 00:37:35,605 Alin. 790 00:37:35,605 --> 00:37:37,605 AUDIENCE: This is why we should stop using them. 791 00:37:37,605 --> 00:37:39,810 And this is a perfect opportunity 792 00:37:39,810 --> 00:37:41,930 to actually talk about append-only ledgers 793 00:37:41,930 --> 00:37:43,430 because that's actually meaningful-- 794 00:37:43,430 --> 00:37:45,555 a meaningful notion that's been studied in computer 795 00:37:45,555 --> 00:37:46,830 science for decades. 796 00:37:46,830 --> 00:37:48,630 Or consensus algorithms, which is 797 00:37:48,630 --> 00:37:50,640 another meaningful notion studied in computer 798 00:37:50,640 --> 00:37:51,870 science for decades. 799 00:37:51,870 --> 00:37:56,550 And consensus algorithms on append-only ledgers, sometimes 800 00:37:56,550 --> 00:37:58,633 referred to colloquially as blockchain, 801 00:37:58,633 --> 00:38:01,050 and then we would know what are we actually talking about. 802 00:38:01,050 --> 00:38:02,203 If we-- 803 00:38:02,203 --> 00:38:03,600 [LAUGHTER] 804 00:38:03,600 --> 00:38:05,790 GARY GENSLER: Well, I will try to help. 805 00:38:05,790 --> 00:38:08,370 But we're only 80 people, and there is a lot 806 00:38:08,370 --> 00:38:09,960 of people using these terms. 807 00:38:09,960 --> 00:38:12,150 And when you go outside of this class or even 808 00:38:12,150 --> 00:38:15,900 when you read an open letter to Jamie Dimon-- 809 00:38:15,900 --> 00:38:23,580 of the somewhere between 25 and 50 readings you had so far 810 00:38:23,580 --> 00:38:26,760 this semester, they've used the word blockchain 811 00:38:26,760 --> 00:38:27,720 in multiple ways. 812 00:38:27,720 --> 00:38:29,040 I'm agreeing with you. 813 00:38:29,040 --> 00:38:31,500 You're just asking me not to ever use the word blockchain 814 00:38:31,500 --> 00:38:32,000 again? 815 00:38:32,000 --> 00:38:32,710 AUDIENCE: Yes. 816 00:38:32,710 --> 00:38:34,677 GARY GENSLER: Oh, my god. 817 00:38:34,677 --> 00:38:36,510 And he's sitting in the center of the class. 818 00:38:36,510 --> 00:38:39,210 Let's keep going. 819 00:38:39,210 --> 00:38:41,280 So we're going to study these. 820 00:38:41,280 --> 00:38:43,750 This is just this list that we've talked about. 821 00:38:43,750 --> 00:38:47,580 But in Act Three, we're going to come to it after SIP week 822 00:38:47,580 --> 00:38:51,150 and start to go through venture capital payment systems 823 00:38:51,150 --> 00:38:52,670 and so forth. 824 00:38:52,670 --> 00:38:54,420 It might be some of what you're picking up 825 00:38:54,420 --> 00:38:57,630 for the final projects. 826 00:38:57,630 --> 00:39:02,677 But here-- and I stress the word, potential use cases. 827 00:39:02,677 --> 00:39:03,510 Will these all work? 828 00:39:03,510 --> 00:39:05,640 Will they be places? 829 00:39:05,640 --> 00:39:10,210 In the non-financial space, supply chain management 830 00:39:10,210 --> 00:39:12,480 I put on this list, but it's kind of financial 831 00:39:12,480 --> 00:39:14,500 and I wanted to include it. 832 00:39:14,500 --> 00:39:18,510 Digital identity is both within finance and outside of finance. 833 00:39:18,510 --> 00:39:21,210 Property and asset registries-- you can call it finance, 834 00:39:21,210 --> 00:39:24,810 but I'm not going to dig into that a lot 835 00:39:24,810 --> 00:39:26,370 in the rest of the semester. 836 00:39:26,370 --> 00:39:30,690 Medical records, internet of things, and so forth. 837 00:39:30,690 --> 00:39:32,250 This isn't going to be exhaustive. 838 00:39:32,250 --> 00:39:38,520 If you come on Tuesday, October 30, to the Bitcoin blockchain 839 00:39:38,520 --> 00:39:40,600 dinner that Simon Johnson sponsors, 840 00:39:40,600 --> 00:39:43,080 we'll have somebody doing election and voting. 841 00:39:43,080 --> 00:39:47,520 So there's a lot of other interesting applications. 842 00:39:47,520 --> 00:39:50,130 I tend to be a little doubtful about the voting 843 00:39:50,130 --> 00:39:53,640 because I think official sector and governments want something 844 00:39:53,640 --> 00:39:56,520 so centralized in that. 845 00:39:56,520 --> 00:40:01,470 But maybe it will be more decentralized. 846 00:40:01,470 --> 00:40:04,350 Maybe it's append-only without a consensus. 847 00:40:08,100 --> 00:40:10,980 McKinsey, they broke it down into six buckets. 848 00:40:10,980 --> 00:40:15,230 They had the record keeping and the transaction-based. 849 00:40:15,230 --> 00:40:17,820 And a lot of these are the similar buckets. 850 00:40:20,340 --> 00:40:23,060 It's just to give a spur to all of it. 851 00:40:23,060 --> 00:40:29,080 So accessing costs and benefits. 852 00:40:29,080 --> 00:40:31,960 So we've talked about what are the benefits. 853 00:40:31,960 --> 00:40:34,140 These are the questions that I would be saying. 854 00:40:34,140 --> 00:40:36,110 And some of this was in the Geneva Report, 855 00:40:36,110 --> 00:40:38,340 but rather than trying to tease it out 856 00:40:38,340 --> 00:40:46,920 of all of the discussion, here I would lay out four big buckets 857 00:40:46,920 --> 00:40:51,690 as to how I think about this and how I might get your feedback. 858 00:40:51,690 --> 00:40:53,700 If you see other ways to look at this, 859 00:40:53,700 --> 00:40:56,990 I want to learn from you all as well. 860 00:40:56,990 --> 00:40:59,280 And when I say four, this is the first bucket. 861 00:40:59,280 --> 00:41:03,910 Basically, what are the benefits of using this technology? 862 00:41:03,910 --> 00:41:08,400 Whether it's this idea about consumer credit or some of you 863 00:41:08,400 --> 00:41:11,880 have had ideas about government procurement 864 00:41:11,880 --> 00:41:18,300 or payments, wherever you are, what is the true benefit? 865 00:41:18,300 --> 00:41:20,730 You've got to find a pain point. 866 00:41:20,730 --> 00:41:23,530 Every entrepreneur always has to find a pain point, something 867 00:41:23,530 --> 00:41:27,180 that you can be solving it for a group of stakeholders. 868 00:41:27,180 --> 00:41:32,200 And who are the stakeholders you're solving something for? 869 00:41:32,200 --> 00:41:36,820 Or if you're a company, is it an internal pain point? 870 00:41:36,820 --> 00:41:39,160 Pain point is some constraint to making 871 00:41:39,160 --> 00:41:41,950 more profits, more revenues. 872 00:41:41,950 --> 00:41:44,500 Or it's a friction, it's a cost and you 873 00:41:44,500 --> 00:41:46,780 want to get rid of a cost. 874 00:41:46,780 --> 00:41:50,500 But what is the pain point for providing a good service 875 00:41:50,500 --> 00:41:53,120 or making more money? 876 00:41:53,120 --> 00:41:56,250 How can you capture the value? 877 00:41:56,250 --> 00:41:58,560 As an entrepreneur, if you do something 878 00:41:58,560 --> 00:42:01,230 for eleemosynary purposes, meaning 879 00:42:01,230 --> 00:42:03,360 you don't care about making money, 880 00:42:03,360 --> 00:42:05,070 I applaud that, that's terrific. 881 00:42:05,070 --> 00:42:08,490 You can even hand that in for your final project. 882 00:42:08,490 --> 00:42:12,420 But I'm assuming that you're going to be thinking like me, 883 00:42:12,420 --> 00:42:14,550 how do you capture value? 884 00:42:14,550 --> 00:42:18,150 Not only do something, but also capture that value 885 00:42:18,150 --> 00:42:22,110 and get a little of somebody else's economic rents 886 00:42:22,110 --> 00:42:24,330 if possible. 887 00:42:24,330 --> 00:42:26,518 Thirdly, what are the competition doing? 888 00:42:26,518 --> 00:42:27,810 What are the competitors doing? 889 00:42:27,810 --> 00:42:30,060 Even if they're using traditional databases, 890 00:42:30,060 --> 00:42:36,580 how are they addressing the same pain point in anything 891 00:42:36,580 --> 00:42:37,760 you're doing? 892 00:42:37,760 --> 00:42:40,030 And why is blockchain technology-- 893 00:42:40,030 --> 00:42:41,960 or, as Alin would wish me to say, 894 00:42:41,960 --> 00:42:46,510 why are append-only longs and consensus protocols-- 895 00:42:46,510 --> 00:42:47,860 what is the answer? 896 00:42:47,860 --> 00:42:51,640 What is it about append-only logs with their time 897 00:42:51,640 --> 00:42:55,480 stamping, with their final settlement, 898 00:42:55,480 --> 00:42:59,470 the finality of that settlement, what about that? 899 00:42:59,470 --> 00:43:04,330 Or even a full consensus protocol and a native currency. 900 00:43:04,330 --> 00:43:07,150 So I'd add a third thing because permissionless would also 901 00:43:07,150 --> 00:43:10,480 have an incentive system in a native currency. 902 00:43:10,480 --> 00:43:15,330 So what is it about append-only logs, consensus protocol, 903 00:43:15,330 --> 00:43:16,430 and a native currency? 904 00:43:16,430 --> 00:43:17,260 How am I doing? 905 00:43:17,260 --> 00:43:18,177 AUDIENCE: I love this. 906 00:43:18,177 --> 00:43:19,510 GARY GENSLER: You love it, yeah. 907 00:43:19,510 --> 00:43:21,094 AUDIENCE: Talk like this all the time. 908 00:43:21,094 --> 00:43:22,042 Don't forget it. 909 00:43:22,042 --> 00:43:24,840 GARY GENSLER: Yeah. 910 00:43:24,840 --> 00:43:26,028 Oh, my god. 911 00:43:26,028 --> 00:43:27,940 [LAUGHTER] 912 00:43:28,900 --> 00:43:31,020 Oh, I need this. 913 00:43:31,020 --> 00:43:32,920 But that's the core. 914 00:43:32,920 --> 00:43:36,790 And I talk about it being the final project, 915 00:43:36,790 --> 00:43:40,330 but this is going on right now. $28 billion dollars 916 00:43:40,330 --> 00:43:43,120 has been raised in initial coin offerings 917 00:43:43,120 --> 00:43:45,520 and a few extra billion, 3 or 4 other billion 918 00:43:45,520 --> 00:43:48,550 has been raised by direct venture capital in this world, 919 00:43:48,550 --> 00:43:52,090 but call it $30 billion. 920 00:43:52,090 --> 00:43:56,020 It's not fully mature, but it's maturing, 921 00:43:56,020 --> 00:43:58,840 and a lot of venture capitalists are saying, wait a minute, 922 00:43:58,840 --> 00:44:02,200 wait a minute, we're past the hype stage. 923 00:44:02,200 --> 00:44:04,330 What pain point are you trying to address? 924 00:44:04,330 --> 00:44:06,010 How are you creating value? 925 00:44:06,010 --> 00:44:08,360 What are the competitions doing? 926 00:44:08,360 --> 00:44:13,330 And why do you need append-only logs, consensus protocols, 927 00:44:13,330 --> 00:44:15,797 and possibly native tokens? 928 00:44:15,797 --> 00:44:18,130 AUDIENCE: I just wanted to follow-up on that point about 929 00:44:18,130 --> 00:44:18,755 the Accenture-- 930 00:44:18,755 --> 00:44:20,463 GARY GENSLER: You're not going to tell me 931 00:44:20,463 --> 00:44:21,640 how to define something? 932 00:44:21,640 --> 00:44:24,245 All right, Kelly. 933 00:44:24,245 --> 00:44:27,612 AUDIENCE: The Accenture thing he was talking about from the-- 934 00:44:27,612 --> 00:44:29,810 GARY GENSLER: It's immutable but amendable. 935 00:44:29,810 --> 00:44:30,550 AUDIENCE: Right. 936 00:44:30,550 --> 00:44:32,380 So we ask ourselves the question-- 937 00:44:32,380 --> 00:44:33,602 that seems a little bit-- 938 00:44:37,750 --> 00:44:39,020 it defeats the purpose, right? 939 00:44:39,020 --> 00:44:44,090 So I then ask myself, what value is being created by doing that? 940 00:44:44,090 --> 00:44:50,840 If it's not necessarily for their customers, 941 00:44:50,840 --> 00:44:53,863 where is the internal value lying? 942 00:44:53,863 --> 00:44:56,030 GARY GENSLER: I don't know what the application was. 943 00:44:56,030 --> 00:45:07,240 But I think that in every application that we find, 944 00:45:07,240 --> 00:45:11,580 not just in blockchain but in other applications, 945 00:45:11,580 --> 00:45:15,900 you have to ask absolute final settlement. 946 00:45:15,900 --> 00:45:20,690 If you can never amend it, if it is truly immutable, 947 00:45:20,690 --> 00:45:23,810 what cost does that also make? 948 00:45:23,810 --> 00:45:25,880 I assume that in this application 949 00:45:25,880 --> 00:45:27,650 that was too great a cost for them 950 00:45:27,650 --> 00:45:29,960 and they wanted a backdoor, a way 951 00:45:29,960 --> 00:45:34,160 to amend that which was supposed to be immutable. 952 00:45:34,160 --> 00:45:36,020 Again, I don't know that application. 953 00:45:36,020 --> 00:45:37,430 But that is one of the-- 954 00:45:37,430 --> 00:45:39,410 I think it's a feature of blockchain. 955 00:45:39,410 --> 00:45:41,390 But it's both a feature and, to some, 956 00:45:41,390 --> 00:45:45,160 a bug that you cannot amend something. 957 00:45:45,160 --> 00:45:46,190 AUDIENCE: Yeah. 958 00:45:46,190 --> 00:45:49,210 It says that one of the benefits was 959 00:45:49,210 --> 00:45:51,160 that it allowed different entities involved 960 00:45:51,160 --> 00:45:52,630 to draw on the same record. 961 00:45:52,630 --> 00:45:55,480 So yeah, of course, I can see where that's valuable. 962 00:45:55,480 --> 00:45:57,708 But then it also begs the question, 963 00:45:57,708 --> 00:45:59,500 if you want to change it, then why not just 964 00:45:59,500 --> 00:46:01,170 use a traditional database? 965 00:46:01,170 --> 00:46:05,450 GARY GENSLER: Vitalik Buterin in 2016 966 00:46:05,450 --> 00:46:11,950 helped when an organization called DAO, the DAO-- 967 00:46:11,950 --> 00:46:15,430 is it Decentralized Autonomous Organization? 968 00:46:15,430 --> 00:46:17,570 Thank you for those who-- 969 00:46:17,570 --> 00:46:20,780 it was one of the largest at the time initial coin offerings. 970 00:46:20,780 --> 00:46:24,490 It raised about $160 US million. 971 00:46:24,490 --> 00:46:29,330 And very quickly, right in the smart contract-- 972 00:46:29,330 --> 00:46:31,240 it wasn't in the base Ethereum layer, 973 00:46:31,240 --> 00:46:33,970 but I believe it was in the smart contract on top, 974 00:46:33,970 --> 00:46:37,900 somebody saw what has later been called a bug. 975 00:46:37,900 --> 00:46:39,820 I don't know if it's truly a bug. 976 00:46:39,820 --> 00:46:42,730 It was programmed in, in a sense, 977 00:46:42,730 --> 00:46:45,790 but there was a way that they could see how to get in there. 978 00:46:45,790 --> 00:46:48,630 And they took about a third of the tokens, 979 00:46:48,630 --> 00:46:51,550 so the equivalent of about $50 million. 980 00:46:51,550 --> 00:46:53,110 But the way the programming worked, 981 00:46:53,110 --> 00:46:57,340 there was about two weeks before it actually was truly final. 982 00:46:57,340 --> 00:47:01,780 It was something in the code also, and that wasn't a bug. 983 00:47:01,780 --> 00:47:06,130 And the whole Ethereum community was debating it. 984 00:47:06,130 --> 00:47:10,130 And Vitalik Buterin said, no, we can't let this happen. 985 00:47:10,130 --> 00:47:12,880 So in a sense, they big footed it. 986 00:47:12,880 --> 00:47:15,130 They almost did what Essentia. 987 00:47:15,130 --> 00:47:16,900 Now, he might say, no, actually, it 988 00:47:16,900 --> 00:47:22,000 was right at the 11th hour and 59th minute 989 00:47:22,000 --> 00:47:25,340 because there was this 10-day or two-week period. 990 00:47:25,340 --> 00:47:28,570 But they basically, Vitalik Buterin even did kind of what 991 00:47:28,570 --> 00:47:30,492 Essentia did at that moment. 992 00:47:30,492 --> 00:47:32,034 AUDIENCE: They forked it then, right? 993 00:47:34,587 --> 00:47:35,920 GARY GENSLER: I don't even know. 994 00:47:35,920 --> 00:47:38,160 Was it a technical fork because of how they did it? 995 00:47:38,160 --> 00:47:40,232 AUDIENCE: [INAUDIBLE] 996 00:47:40,232 --> 00:47:41,690 GARY GENSLER: Right at that moment. 997 00:47:41,690 --> 00:47:44,372 AUDIENCE: Ethereum classic still has the DAO-- 998 00:47:44,372 --> 00:47:46,510 GARY GENSLER: The stolen $50 million 999 00:47:46,510 --> 00:47:48,200 or whatever, the number of tokens. 1000 00:47:48,200 --> 00:47:51,590 So that led to the fork. 1001 00:47:51,590 --> 00:47:56,280 The majority of the community went with Vitalik 1002 00:47:56,280 --> 00:48:00,000 rather than what the fork happened. 1003 00:48:00,000 --> 00:48:06,320 So how immutable is immutable if there's a broad consensus? 1004 00:48:06,320 --> 00:48:09,440 Now that's the backdoor, in a sense. 1005 00:48:09,440 --> 00:48:11,690 Second-- was there another hand? 1006 00:48:11,690 --> 00:48:13,200 I'm sorry, Brodish. 1007 00:48:13,200 --> 00:48:16,220 AUDIENCE: I think the [INAUDIBLE] 1008 00:48:16,220 --> 00:48:19,340 GARY GENSLER: To address Kelly's and Zann's. 1009 00:48:19,340 --> 00:48:21,117 AUDIENCE: And the word blockchain. 1010 00:48:21,117 --> 00:48:22,700 GARY GENSLER: And the word blockchain. 1011 00:48:22,700 --> 00:48:23,660 All right. 1012 00:48:23,660 --> 00:48:26,300 AUDIENCE: So I think when people are talking about blockchain 1013 00:48:26,300 --> 00:48:28,460 as a term, they're essentially referring 1014 00:48:28,460 --> 00:48:32,060 to a set of potential benefits of this technology. 1015 00:48:32,060 --> 00:48:34,530 And some of them can be relevant for a particular use case. 1016 00:48:34,530 --> 00:48:36,405 Some of them might [INAUDIBLE],, some of them 1017 00:48:36,405 --> 00:48:40,260 might actually [INAUDIBLE] for that particular use case. 1018 00:48:40,260 --> 00:48:42,530 So the way it can be thought of is 1019 00:48:42,530 --> 00:48:45,670 that we have an umbrella term which talks about referring 1020 00:48:45,670 --> 00:48:46,940 potential benefits. 1021 00:48:46,940 --> 00:48:48,740 We take what is [? referred ?] from that. 1022 00:48:48,740 --> 00:48:51,430 It could be append-only, it could be not append-only, 1023 00:48:51,430 --> 00:48:53,510 it could be consensus, it could be permission, 1024 00:48:53,510 --> 00:48:54,740 permissionless, whatever. 1025 00:48:54,740 --> 00:48:57,690 And then we cannot calibrate with them and create a solution 1026 00:48:57,690 --> 00:49:00,950 which is applicable for the particular use case that I want 1027 00:49:00,950 --> 00:49:03,110 to look at. 1028 00:49:03,110 --> 00:49:05,810 That is why even if it is not immutable, 1029 00:49:05,810 --> 00:49:07,940 it is still a potential use of blockchain, 1030 00:49:07,940 --> 00:49:09,530 from my understanding. 1031 00:49:09,530 --> 00:49:12,730 Because we are bringing up some of the benefits of the umbrella 1032 00:49:12,730 --> 00:49:13,730 term. 1033 00:49:13,730 --> 00:49:19,430 GARY GENSLER: I think you're correct, 1034 00:49:19,430 --> 00:49:23,420 but I'm not going to go all the way to where you are. 1035 00:49:23,420 --> 00:49:26,420 I think you're correct that, broadly speaking, 1036 00:49:26,420 --> 00:49:31,280 many people are using this term to promote 1037 00:49:31,280 --> 00:49:34,130 their own innovation, the innovation of their colleagues, 1038 00:49:34,130 --> 00:49:36,050 and so forth. 1039 00:49:36,050 --> 00:49:39,120 And yet, some want a backdoor, like Essentia, 1040 00:49:39,120 --> 00:49:40,680 and make it less immutable. 1041 00:49:40,680 --> 00:49:43,220 And so all of a sudden, if it has a big backdoor 1042 00:49:43,220 --> 00:49:47,990 and you can amend, is it really an append-only log any longer? 1043 00:49:47,990 --> 00:49:52,850 What I do want to say is, in this class, I am going to say, 1044 00:49:52,850 --> 00:49:55,770 anything you're working on for your final paper, 1045 00:49:55,770 --> 00:50:00,430 it has to be at least a permissioned or permissionless 1046 00:50:00,430 --> 00:50:01,120 system. 1047 00:50:01,120 --> 00:50:03,010 Thus it has to have a-- 1048 00:50:03,010 --> 00:50:04,114 what's that? 1049 00:50:04,114 --> 00:50:04,990 AUDIENCE: Not mine. 1050 00:50:04,990 --> 00:50:05,530 GARY GENSLER: Not yours? 1051 00:50:05,530 --> 00:50:06,680 But you're going to get there. 1052 00:50:06,680 --> 00:50:08,230 You're going to get there, Brodish. 1053 00:50:08,230 --> 00:50:09,950 You're going to get there. 1054 00:50:09,950 --> 00:50:16,720 Since I told you earlier in the day, get to yes. 1055 00:50:16,720 --> 00:50:22,150 But if it's a traditional database, 1056 00:50:22,150 --> 00:50:26,020 I'm saying stretch your minds, try to figure out something. 1057 00:50:26,020 --> 00:50:27,910 Even if it's got a little bit of risk 1058 00:50:27,910 --> 00:50:32,230 into it as to how you get to using append-only logs, 1059 00:50:32,230 --> 00:50:36,040 and even in a permissioned closed loose private consensus, 1060 00:50:36,040 --> 00:50:37,660 but some consensus. 1061 00:50:37,660 --> 00:50:39,720 Think about whether you use a token that 1062 00:50:39,720 --> 00:50:40,720 gets the permissionless. 1063 00:50:40,720 --> 00:50:41,220 Uh-oh. 1064 00:50:41,220 --> 00:50:43,310 Eric's going to plead his case. 1065 00:50:43,310 --> 00:50:48,497 AUDIENCE: I'm just going to try to completely reconcile this. 1066 00:50:48,497 --> 00:50:50,330 GARY GENSLER: And then I need to keep going. 1067 00:50:50,330 --> 00:50:51,070 AUDIENCE: Yeah. 1068 00:50:51,070 --> 00:50:53,697 Those two apparently divergent points-- 1069 00:50:53,697 --> 00:50:55,280 GARY GENSLER: Right, Brotish and Alin. 1070 00:50:55,280 --> 00:50:55,930 AUDIENCE: Yeah. 1071 00:50:55,930 --> 00:50:58,630 AUDIENCE: I actually agree with Brotish. 1072 00:50:58,630 --> 00:50:59,860 AUDIENCE: The point is that-- 1073 00:50:59,860 --> 00:51:02,085 GARY GENSLER: That he might agree with him. 1074 00:51:02,085 --> 00:51:03,460 AUDIENCE: I'm still in the middle 1075 00:51:03,460 --> 00:51:07,780 because from a technical standpoint, it's true. 1076 00:51:07,780 --> 00:51:11,770 You have append-only log in a consensus mechanism 1077 00:51:11,770 --> 00:51:16,300 that make up the original use case that 1078 00:51:16,300 --> 00:51:18,693 brought the term blockchain. 1079 00:51:18,693 --> 00:51:20,610 GARY GENSLER: Plus a native token in that one. 1080 00:51:20,610 --> 00:51:21,756 AUDIENCE: Yeah, of course. 1081 00:51:21,756 --> 00:51:25,330 [INAUDIBLE] But to the point, though, what 1082 00:51:25,330 --> 00:51:29,410 happens if you want to change or amend the log? 1083 00:51:29,410 --> 00:51:31,000 Then is that really a blockchain? 1084 00:51:31,000 --> 00:51:35,860 That's exactly the reason why the term [INAUDIBLE] 1085 00:51:35,860 --> 00:51:38,560 later technologies came up. 1086 00:51:38,560 --> 00:51:43,570 Because to a certain purist extent, 1087 00:51:43,570 --> 00:51:47,980 it's not an append-only log because you're actually 1088 00:51:47,980 --> 00:51:52,830 updating it from a purist definition. 1089 00:51:52,830 --> 00:51:54,090 That makes sense, of course. 1090 00:51:54,090 --> 00:51:58,870 But then the nature of a distributed database 1091 00:51:58,870 --> 00:52:03,640 that's not located in an intranet 1092 00:52:03,640 --> 00:52:09,000 that governs the updates and whatever changes you make it, 1093 00:52:09,000 --> 00:52:14,185 it's the reason that make up these ledger technologies that 1094 00:52:14,185 --> 00:52:16,750 are blockchain inspired, if you may. 1095 00:52:16,750 --> 00:52:23,560 So it's OK, you can say if you amend the log, 1096 00:52:23,560 --> 00:52:24,830 then it's not blockchain. 1097 00:52:24,830 --> 00:52:28,420 Right, but it's DLT, and the DLT is blockchain inspired. 1098 00:52:28,420 --> 00:52:32,078 So we can keep everything in the neighborhood. 1099 00:52:32,078 --> 00:52:34,120 GARY GENSLER: We're going to keep this discussion 1100 00:52:34,120 --> 00:52:35,170 and debate going. 1101 00:52:35,170 --> 00:52:37,660 And we're probably not all that far off. 1102 00:52:37,660 --> 00:52:39,370 It's not about vocabulary, it's also 1103 00:52:39,370 --> 00:52:44,050 about I'm trying to spur you all to get to yes, even if you're 1104 00:52:44,050 --> 00:52:45,550 a minimalist. 1105 00:52:45,550 --> 00:52:49,360 Think about some project that uses the essence of this, 1106 00:52:49,360 --> 00:52:51,790 and not just inspired. 1107 00:52:51,790 --> 00:52:56,275 So then what are the specifics of this append-only log 1108 00:52:56,275 --> 00:53:01,930 and consensus use case or blockchain use case? 1109 00:53:01,930 --> 00:53:04,970 So what cost of verification or networking 1110 00:53:04,970 --> 00:53:07,060 are you actually reducing? 1111 00:53:07,060 --> 00:53:10,522 Back to Catalini's work about verification cost 1112 00:53:10,522 --> 00:53:12,730 and networking costs that we talked about on Tuesday. 1113 00:53:12,730 --> 00:53:16,210 What is that cost you're actually reducing? 1114 00:53:16,210 --> 00:53:18,640 Which transactions need to be recorded? 1115 00:53:18,640 --> 00:53:20,800 What accounts or transactions, you might say, 1116 00:53:20,800 --> 00:53:24,070 need to be recorded on this log? 1117 00:53:24,070 --> 00:53:26,260 What stakeholders need to be able to write? 1118 00:53:26,260 --> 00:53:30,760 If there's only one stakeholder that needs to update something, 1119 00:53:30,760 --> 00:53:32,860 I'm not really entirely sure that you'd 1120 00:53:32,860 --> 00:53:36,280 be in this distributed decentralized place. 1121 00:53:36,280 --> 00:53:41,610 So I list these questions that weren't directly 1122 00:53:41,610 --> 00:53:43,620 in the readings, but I list these questions 1123 00:53:43,620 --> 00:53:48,030 to say this is a way to start to get your mind around where 1124 00:53:48,030 --> 00:53:52,230 this technology, however we label it, 1125 00:53:52,230 --> 00:53:54,210 this technology can help. 1126 00:53:54,210 --> 00:53:57,630 In essence, does it lower cost of verification and networking? 1127 00:53:57,630 --> 00:53:58,878 And if so, which ones? 1128 00:53:58,878 --> 00:54:00,420 Which ones are you trying to address? 1129 00:54:00,420 --> 00:54:03,360 Which ones are you trying to dig and basically create 1130 00:54:03,360 --> 00:54:06,280 value for your startup? 1131 00:54:06,280 --> 00:54:09,640 What transactions or accounts are you trying to record? 1132 00:54:09,640 --> 00:54:12,010 Transactions, like changing property rights 1133 00:54:12,010 --> 00:54:15,340 or moving around birth records or land records 1134 00:54:15,340 --> 00:54:16,810 or supply chain. 1135 00:54:16,810 --> 00:54:18,640 But where's that? 1136 00:54:18,640 --> 00:54:22,150 Which stakeholders actually need to write to this distributed 1137 00:54:22,150 --> 00:54:23,337 ledger? 1138 00:54:23,337 --> 00:54:24,670 Because if it's only one party-- 1139 00:54:24,670 --> 00:54:29,230 I'm still trying to get my head around why the Australian Stock 1140 00:54:29,230 --> 00:54:31,556 Exchange really needs it, but OK. 1141 00:54:31,556 --> 00:54:35,920 What is it that multiple parties need 1142 00:54:35,920 --> 00:54:41,710 to write to this shared collective state of what 1143 00:54:41,710 --> 00:54:43,150 is there? 1144 00:54:43,150 --> 00:54:46,210 And then the last one is also terribly important, 1145 00:54:46,210 --> 00:54:49,180 is, what's the customer user interface? 1146 00:54:49,180 --> 00:54:53,710 And a lot of challenges are basically like, 1147 00:54:53,710 --> 00:54:55,240 currently we're not in a place where 1148 00:54:55,240 --> 00:54:58,090 a lot of blockchain applications are distributed, 1149 00:54:58,090 --> 00:55:00,800 applications have great customer user interfaces. 1150 00:55:00,800 --> 00:55:05,950 But that as well. 1151 00:55:05,950 --> 00:55:08,050 Then what are the cost of the transitions? 1152 00:55:08,050 --> 00:55:10,512 Akira mentioned this from the Geneva Report 1153 00:55:10,512 --> 00:55:11,470 and the other readings. 1154 00:55:11,470 --> 00:55:13,900 But what are the costs and challenges? 1155 00:55:13,900 --> 00:55:16,840 Well, of course, we still have a bunch of scalability 1156 00:55:16,840 --> 00:55:17,950 and performance issues. 1157 00:55:17,950 --> 00:55:21,040 We have the privacy issues and coordination. 1158 00:55:21,040 --> 00:55:23,620 I'm not trying to hold us all to that, by the way, 1159 00:55:23,620 --> 00:55:26,560 for your final projects. 1160 00:55:26,560 --> 00:55:29,380 If we're not going to get to scalability for five or seven 1161 00:55:29,380 --> 00:55:32,120 years, I'm not going to say, well, 1162 00:55:32,120 --> 00:55:34,600 then you all have to write about permissioned systems. 1163 00:55:34,600 --> 00:55:35,290 No. 1164 00:55:35,290 --> 00:55:41,080 I'm OK if you say, well, this is a pilot 1165 00:55:41,080 --> 00:55:42,610 and somebody else is going to solve 1166 00:55:42,610 --> 00:55:47,020 through layer two or zk-SNARKs or something, 1167 00:55:47,020 --> 00:55:49,730 these other things. 1168 00:55:49,730 --> 00:55:52,390 But I'd like you to at least identify 1169 00:55:52,390 --> 00:55:54,620 what are the issues of scalability, 1170 00:55:54,620 --> 00:55:57,910 performance, privacy, security, those issues that might 1171 00:55:57,910 --> 00:56:00,650 hold your implementation out. 1172 00:56:00,650 --> 00:56:02,470 Can a permissioned blockchain adequately 1173 00:56:02,470 --> 00:56:06,000 addresses whatever the use case is? 1174 00:56:06,000 --> 00:56:09,910 And how can you actually jump start broad adoption? 1175 00:56:09,910 --> 00:56:12,820 That's your network challenges. 1176 00:56:12,820 --> 00:56:15,340 How, when you start Uber, can you get a lot of people 1177 00:56:15,340 --> 00:56:16,750 to use your Uber app? 1178 00:56:16,750 --> 00:56:21,280 But in this case, if I can use that consumer credit-- 1179 00:56:21,280 --> 00:56:25,270 I thought the consumer credit idea was really a neat one. 1180 00:56:25,270 --> 00:56:28,060 Maybe it could go right at the heart of a lot of market power. 1181 00:56:28,060 --> 00:56:30,525 But now, how do you get all the auto dealers, 1182 00:56:30,525 --> 00:56:31,900 how do you get all the employers, 1183 00:56:31,900 --> 00:56:35,410 how do you get the banks to actually say, 1184 00:56:35,410 --> 00:56:40,990 I want to get rid of First Union and I want the Blockchain 1185 00:56:40,990 --> 00:56:45,170 and Money final project team's-- 1186 00:56:45,170 --> 00:56:48,920 and by the way, just replacing one central monopolist 1187 00:56:48,920 --> 00:56:51,770 with another central monopolist, a lot of commercial banks 1188 00:56:51,770 --> 00:56:54,367 aren't terribly excited about that. 1189 00:56:54,367 --> 00:56:56,450 Maybe you're going to have to give some ownership. 1190 00:56:56,450 --> 00:56:58,700 Maybe you'll have to give 50% of your ownership 1191 00:56:58,700 --> 00:57:02,940 to the 20 banks that are now a part of this. 1192 00:57:02,940 --> 00:57:04,940 There are other ways to build incentive systems. 1193 00:57:04,940 --> 00:57:07,700 Maybe you put a native token in there 1194 00:57:07,700 --> 00:57:11,150 and you give them the native token. 1195 00:57:11,150 --> 00:57:13,520 So there may be other incentive ways 1196 00:57:13,520 --> 00:57:16,730 to beat the current monopolist. 1197 00:57:20,000 --> 00:57:22,250 But having been around commercial banks 1198 00:57:22,250 --> 00:57:29,042 for a long time, they really are looking for ways 1199 00:57:29,042 --> 00:57:32,400 to replace their aggregators and their back office, 1200 00:57:32,400 --> 00:57:37,250 whether it's clearing, settlement, exchanges, credit 1201 00:57:37,250 --> 00:57:38,580 agencies. 1202 00:57:38,580 --> 00:57:41,000 But they are deeply understanding-- 1203 00:57:41,000 --> 00:57:44,630 because they are pretty ambitious and very 1204 00:57:44,630 --> 00:57:46,970 good at making money as well, they deeply 1205 00:57:46,970 --> 00:57:50,330 understand that every idea that's pitched to them 1206 00:57:50,330 --> 00:57:53,660 is somebody who currently is small 1207 00:57:53,660 --> 00:57:57,920 but wants to get big and gain market power. 1208 00:57:57,920 --> 00:58:00,350 So they're also always thinking about how 1209 00:58:00,350 --> 00:58:04,610 to ensure that there's a check to slow down 1210 00:58:04,610 --> 00:58:08,040 your startup from gaining market power 5, 10, 1211 00:58:08,040 --> 00:58:09,080 and 15 years from now. 1212 00:58:09,080 --> 00:58:11,810 They don't want you to be the next Clearinghouse, 1213 00:58:11,810 --> 00:58:15,740 like the Chicago Mercantile Exchange or Intercontinental 1214 00:58:15,740 --> 00:58:16,250 Exchange. 1215 00:58:16,250 --> 00:58:20,960 They don't want you to be the next central node of market 1216 00:58:20,960 --> 00:58:22,070 power. 1217 00:58:22,070 --> 00:58:23,840 So you might share some of that with them, 1218 00:58:23,840 --> 00:58:27,370 or you might find an incentive system. 1219 00:58:27,370 --> 00:58:29,450 And then what are the net? 1220 00:58:29,450 --> 00:58:34,130 So key questions for companies designing blockchains. 1221 00:58:34,130 --> 00:58:38,090 This is something from MIT's own Sloan Management Review. 1222 00:58:38,090 --> 00:58:39,950 It was not a required reading, but I 1223 00:58:39,950 --> 00:58:42,290 thought it was a nice little way to do it. 1224 00:58:42,290 --> 00:58:43,442 What are you trying to do? 1225 00:58:43,442 --> 00:58:44,900 Are you trying to record something? 1226 00:58:44,900 --> 00:58:45,567 Track something? 1227 00:58:45,567 --> 00:58:46,430 Verify something? 1228 00:58:46,430 --> 00:58:47,390 Aggregate something? 1229 00:58:47,390 --> 00:58:51,050 This was in the Fall MIT Review. 1230 00:58:51,050 --> 00:58:53,070 What value do you want to capture? 1231 00:58:53,070 --> 00:58:55,070 Literally, here's a list of the things 1232 00:58:55,070 --> 00:58:56,970 you might want to capture. 1233 00:58:56,970 --> 00:58:58,310 Is it about contracts? 1234 00:58:58,310 --> 00:58:59,870 Is it about permissioned? 1235 00:58:59,870 --> 00:59:04,850 What is the value we're trying to capture and for whom? 1236 00:59:04,850 --> 00:59:08,100 Obviously, is it for suppliers or customers and so forth? 1237 00:59:08,100 --> 00:59:11,750 So it's another slice by thoughtful people, MIT. 1238 00:59:14,750 --> 00:59:18,410 That's a link to the reading if you do want to read it. 1239 00:59:18,410 --> 00:59:22,540 It was a well-written five page article. 1240 00:59:22,540 --> 00:59:26,180 I even held back on one reading. 1241 00:59:26,180 --> 00:59:30,973 So the benefits of blockchain-- we reviewed this last Tuesday, 1242 00:59:30,973 --> 00:59:32,640 so I'm not going to go through it again. 1243 00:59:32,640 --> 00:59:34,970 But again, as you're thinking about this 1244 00:59:34,970 --> 00:59:37,090 and thinking about which use cases and so forth, 1245 00:59:37,090 --> 00:59:39,250 think of these costs of verification. 1246 00:59:39,250 --> 00:59:42,440 Which of these are you really trying to hit? 1247 00:59:42,440 --> 00:59:43,610 Or is it networking? 1248 00:59:43,610 --> 00:59:46,190 And I changed the networking a little bit, 1249 00:59:46,190 --> 00:59:51,580 but a token incentive system, which could be like the skins, 1250 00:59:51,580 --> 00:59:53,770 rewards, and affinity, identity could help 1251 00:59:53,770 --> 00:59:56,680 you start it up or operate it. 1252 00:59:56,680 --> 01:00:00,250 Or it could be like CryptoKitties, 1253 01:00:00,250 --> 01:00:02,710 where it's like a collectible. 1254 01:00:02,710 --> 01:00:04,600 So that's another way that you have 1255 01:00:04,600 --> 01:00:06,250 token economics floating in. 1256 01:00:09,140 --> 01:00:11,210 So this you've seen. 1257 01:00:11,210 --> 01:00:14,860 I've flipped it around, but basically, 1258 01:00:14,860 --> 01:00:17,600 do you use a traditional database? 1259 01:00:17,600 --> 01:00:19,790 Do you think about a private blockchain? 1260 01:00:19,790 --> 01:00:22,280 Do you think about a public blockchain? 1261 01:00:22,280 --> 01:00:24,980 If you're over here, I really ask you, Brodish, you've 1262 01:00:24,980 --> 01:00:26,490 got to get to yes. 1263 01:00:26,490 --> 01:00:29,933 You've got to get to here. 1264 01:00:29,933 --> 01:00:31,600 You wouldn't want me to just be teaching 1265 01:00:31,600 --> 01:00:34,390 a traditional database, how do you build something in Sloan, 1266 01:00:34,390 --> 01:00:35,110 right? 1267 01:00:35,110 --> 01:00:36,910 This is blockchained money. 1268 01:00:36,910 --> 01:00:41,230 Or as Alin wants me to relabel it, append-only logs, consensus 1269 01:00:41,230 --> 01:00:42,971 protocols, and money. 1270 01:00:42,971 --> 01:00:43,893 [LAUGHTER] 1271 01:00:43,893 --> 01:00:45,280 AUDIENCE: And a native currency. 1272 01:00:45,280 --> 01:00:46,240 GARY GENSLER: And a native currency. 1273 01:00:46,240 --> 01:00:48,010 But how many of you would sign up? 1274 01:00:48,010 --> 01:00:49,615 Would you recommend that I list it 1275 01:00:49,615 --> 01:00:52,982 in the course catalog that way for next September? 1276 01:00:52,982 --> 01:00:54,220 AUDIENCE: No. 1277 01:00:54,220 --> 01:00:55,540 GARY GENSLER: Really? 1278 01:00:55,540 --> 01:00:57,310 No, not really. 1279 01:00:57,310 --> 01:00:58,000 All right. 1280 01:00:58,000 --> 01:01:03,640 But if you want to really juice, get to here. 1281 01:01:03,640 --> 01:01:06,280 Put an append-only log in there and some even 1282 01:01:06,280 --> 01:01:09,610 private permissioned system. 1283 01:01:09,610 --> 01:01:11,800 Or be bold. 1284 01:01:11,800 --> 01:01:13,720 Try to get all the way to there. 1285 01:01:13,720 --> 01:01:17,150 Don't stay there, please. 1286 01:01:17,150 --> 01:01:17,650 All right. 1287 01:01:17,650 --> 01:01:18,460 I've made my point. 1288 01:01:18,460 --> 01:01:19,917 James. 1289 01:01:19,917 --> 01:01:20,500 Where are you? 1290 01:01:20,500 --> 01:01:22,110 Are you there or there? 1291 01:01:22,110 --> 01:01:24,491 AUDIENCE: I think I'm in the middle, private blockchain. 1292 01:01:24,491 --> 01:01:25,574 GARY GENSLER: Ah, come on. 1293 01:01:25,574 --> 01:01:26,970 AUDIENCE: [INAUDIBLE] 1294 01:01:26,970 --> 01:01:27,970 GARY GENSLER: All right. 1295 01:01:27,970 --> 01:01:30,810 Good. 1296 01:01:30,810 --> 01:01:35,030 Yeah, but they're going to get at least to there. 1297 01:01:35,030 --> 01:01:38,222 They're not staying there, they want to get a good grade. 1298 01:01:38,222 --> 01:01:40,530 [LAUGHTER] 1299 01:01:40,530 --> 01:01:41,260 No, no. 1300 01:01:41,260 --> 01:01:42,670 Really, this is going to be fun. 1301 01:01:42,670 --> 01:01:44,020 Listen, I get it. 1302 01:01:44,020 --> 01:01:46,273 It's not just like any Sloan class, where you 1303 01:01:46,273 --> 01:01:47,440 have a final and everything. 1304 01:01:47,440 --> 01:01:49,000 I've given you a hard test. 1305 01:01:49,000 --> 01:01:52,850 There's $30 billion chasing around to try to find uses. 1306 01:01:52,850 --> 01:01:56,680 There's thousands-- 2,500 at IBM or whatever-- there's thousands 1307 01:01:56,680 --> 01:01:57,880 of people trying to do this. 1308 01:01:57,880 --> 01:02:02,500 And there's no really true market-wide applications 1309 01:02:02,500 --> 01:02:04,330 right now. 1310 01:02:04,330 --> 01:02:09,100 But that's the nature of what we're doing here together. 1311 01:02:09,100 --> 01:02:11,080 And I'm not asking that somebody actually 1312 01:02:11,080 --> 01:02:13,720 be able to go raise money and actually start it. 1313 01:02:13,720 --> 01:02:17,350 You don't have to code it, except for you, Alin. 1314 01:02:17,350 --> 01:02:19,670 But you don't have to actually code this thing. 1315 01:02:19,670 --> 01:02:24,340 But I'm really looking for your business savvy, 1316 01:02:24,340 --> 01:02:28,150 and it's more about critical reasoning skills. 1317 01:02:28,150 --> 01:02:32,530 And these four or five pages, pull it off of Canvas 1318 01:02:32,530 --> 01:02:34,960 and think about it, think about how it fits 1319 01:02:34,960 --> 01:02:38,500 in with your projects or not. 1320 01:02:38,500 --> 01:02:40,440 Yes? 1321 01:02:40,440 --> 01:02:43,570 AUDIENCE: Given that-- well, the question in previous slides 1322 01:02:43,570 --> 01:02:44,590 are helpful. 1323 01:02:44,590 --> 01:02:48,820 But I still find it's hard to decide whether we should really 1324 01:02:48,820 --> 01:02:52,900 use blockchain database [INAUDIBLE] 1325 01:02:52,900 --> 01:02:54,310 traditional database. 1326 01:02:54,310 --> 01:02:58,820 So is there a non-technical model or framework that we can 1327 01:02:58,820 --> 01:03:03,040 [INAUDIBLE] between those two? 1328 01:03:03,040 --> 01:03:06,100 GARY GENSLER: I think of it-- 1329 01:03:06,100 --> 01:03:10,720 I apologize if to go back to the slides, so I'll try not to. 1330 01:03:10,720 --> 01:03:15,470 But I think about it as, are you moving 1331 01:03:15,470 --> 01:03:18,760 around something of value? 1332 01:03:18,760 --> 01:03:20,590 Do you need multiple people to be 1333 01:03:20,590 --> 01:03:25,020 able to help in that network? 1334 01:03:25,020 --> 01:03:26,160 Because it's peer to peer. 1335 01:03:26,160 --> 01:03:28,350 If it were the 80 or 90 of us in this room, 1336 01:03:28,350 --> 01:03:30,540 you might be just there right now, 1337 01:03:30,540 --> 01:03:32,850 that there's a benefit of that peer to peer 1338 01:03:32,850 --> 01:03:35,100 movement of something of value. 1339 01:03:35,100 --> 01:03:40,500 Or frankly, even back to this, if there's 1340 01:03:40,500 --> 01:03:41,760 a lot of economic rents. 1341 01:03:41,760 --> 01:03:44,910 If there's market power in this, if there's 1342 01:03:44,910 --> 01:03:50,650 big market power and somebody is just collecting lots, 1343 01:03:50,650 --> 01:03:53,710 that might be your opportunity because this is the technology 1344 01:03:53,710 --> 01:03:55,780 that can get you there. 1345 01:03:55,780 --> 01:03:58,990 And you might say, well, I don't absolutely 1346 01:03:58,990 --> 01:04:02,950 have to have an append-only consensus native currency 1347 01:04:02,950 --> 01:04:07,680 type of technology, but it's how I get to peer to peer. 1348 01:04:07,680 --> 01:04:09,200 So it might be that that's the way 1349 01:04:09,200 --> 01:04:11,660 that I'm going to disrupt the party 1350 01:04:11,660 --> 01:04:13,640 with the big economic rents. 1351 01:04:13,640 --> 01:04:16,580 Like the letter to Jamie Dimon, if there's 1352 01:04:16,580 --> 01:04:18,628 a bunch of things about censorship risk. 1353 01:04:18,628 --> 01:04:21,170 And again, I don't know what use cases you're thinking about. 1354 01:04:21,170 --> 01:04:23,770 But I think if you don't have one 1355 01:04:23,770 --> 01:04:27,910 of these six, the verification costs, 1356 01:04:27,910 --> 01:04:32,290 it's unlikely you need an append-only log and consensus 1357 01:04:32,290 --> 01:04:34,030 in the whole mechanism. 1358 01:04:34,030 --> 01:04:37,630 But I think if you do have something here, hopefully not 1359 01:04:37,630 --> 01:04:40,760 just economic rents but market power, 1360 01:04:40,760 --> 01:04:43,930 this might be the way you get underneath it, by having 1361 01:04:43,930 --> 01:04:47,410 a decentralized peer to peer approach. 1362 01:04:47,410 --> 01:04:49,090 So if that helps. 1363 01:04:49,090 --> 01:04:53,500 And it could be something that has no central authority right 1364 01:04:53,500 --> 01:04:56,140 now, no central intermediary, and it's just how 1365 01:04:56,140 --> 01:04:57,910 to jump start something. 1366 01:04:57,910 --> 01:04:59,440 I think in that circumstance, it's 1367 01:04:59,440 --> 01:05:02,320 more likely you need a native token. 1368 01:05:02,320 --> 01:05:06,370 I think if you're trying to jump start a network-- 1369 01:05:06,370 --> 01:05:08,970 and there's a lot of collective action reasons 1370 01:05:08,970 --> 01:05:10,470 that will still face you. 1371 01:05:10,470 --> 01:05:12,630 Some of our colleagues, Andy Lippman 1372 01:05:12,630 --> 01:05:15,780 and others over at the Media Lab, spent a lot of time 1373 01:05:15,780 --> 01:05:18,420 a year, year and a half ago on a medical records approach 1374 01:05:18,420 --> 01:05:22,017 with Beth Israel Hospital locally here. 1375 01:05:22,017 --> 01:05:23,850 And if Andy were here, he'd say, but I still 1376 01:05:23,850 --> 01:05:25,830 have the collective action issues. 1377 01:05:25,830 --> 01:05:27,330 How do I get a bunch of hospitals 1378 01:05:27,330 --> 01:05:29,250 and a bunch of doctors and how do I 1379 01:05:29,250 --> 01:05:33,000 get all that collective action to work together 1380 01:05:33,000 --> 01:05:36,780 to use one database system, if you wish? 1381 01:05:36,780 --> 01:05:39,090 So I think to get over that collective action issue, 1382 01:05:39,090 --> 01:05:41,340 you start getting into, well, this native token 1383 01:05:41,340 --> 01:05:43,596 might be an interesting way to do that. 1384 01:05:43,596 --> 01:05:44,820 AUDIENCE: Right. 1385 01:05:44,820 --> 01:05:48,720 It seemed that Uber satisfied the things you just mentioned. 1386 01:05:48,720 --> 01:05:49,720 GARY GENSLER: They did. 1387 01:05:49,720 --> 01:05:49,990 They did. 1388 01:05:49,990 --> 01:05:51,300 But would they have done it differently 1389 01:05:51,300 --> 01:05:52,950 if they started in a different era? 1390 01:05:56,250 --> 01:05:58,550 I don't know. 1391 01:05:58,550 --> 01:06:01,370 It's remarkable that they took on and just completely 1392 01:06:01,370 --> 01:06:05,830 transformed the ride share business, the taxi 1393 01:06:05,830 --> 01:06:08,890 business, if you wish. 1394 01:06:08,890 --> 01:06:11,278 But I'm pretty confident of this, 1395 01:06:11,278 --> 01:06:12,820 there will be something that we don't 1396 01:06:12,820 --> 01:06:17,530 envision right now that really transforms the next thing. 1397 01:06:17,530 --> 01:06:19,682 As my brother said when he said, it 1398 01:06:19,682 --> 01:06:21,140 might be five and 10 years from now 1399 01:06:21,140 --> 01:06:22,930 and you won't even expect it. 1400 01:06:22,930 --> 01:06:26,570 Who would have thought the internet in the 1990s 1401 01:06:26,570 --> 01:06:30,720 was going to go straight and transform the taxi business? 1402 01:06:30,720 --> 01:06:32,490 Nobody in the New York Taxi business 1403 01:06:32,490 --> 01:06:36,020 would have envisioned that, which was one of the things 1404 01:06:36,020 --> 01:06:39,850 that he mentioned. 1405 01:06:39,850 --> 01:06:43,360 So why use blockchain versus traditional database? 1406 01:06:43,360 --> 01:06:44,560 We did that. 1407 01:06:44,560 --> 01:06:46,600 I know, you got it got, Brodish? 1408 01:06:46,600 --> 01:06:47,730 Right there? 1409 01:06:47,730 --> 01:06:48,230 OK. 1410 01:06:48,230 --> 01:06:50,900 No, I'm having some fun. 1411 01:06:50,900 --> 01:06:54,850 So the trade-offs we've talked about, so these are costs. 1412 01:06:54,850 --> 01:06:56,740 This is the cost of centralization 1413 01:06:56,740 --> 01:06:58,750 versus decentralization. 1414 01:06:58,750 --> 01:07:00,340 We've looked at this slide before. 1415 01:07:03,978 --> 01:07:05,520 I think this was an earlier question. 1416 01:07:05,520 --> 01:07:07,590 Over time, I think what's going to happen 1417 01:07:07,590 --> 01:07:11,610 is these graphs might have different slopes. 1418 01:07:11,610 --> 01:07:15,180 So it's just like, OK, what if the slopes change? 1419 01:07:15,180 --> 01:07:18,600 I think that the cost of decentralization 1420 01:07:18,600 --> 01:07:21,270 are going to come down. 1421 01:07:21,270 --> 01:07:27,990 But that's more my forecasting prognostication. 1422 01:07:27,990 --> 01:07:30,720 And then you'll have more applications 1423 01:07:30,720 --> 01:07:33,660 that feel more comfortable over using something 1424 01:07:33,660 --> 01:07:35,340 around this space. 1425 01:07:35,340 --> 01:07:38,060 Now, I could be wrong and it could go the other way. 1426 01:07:38,060 --> 01:07:39,200 Just saying. 1427 01:07:39,200 --> 01:07:41,550 And this would be more of the minimalist side and says, 1428 01:07:41,550 --> 01:07:44,487 well, you can address some of those centralization costs 1429 01:07:44,487 --> 01:07:45,820 and it comes down the other way. 1430 01:07:45,820 --> 01:07:48,000 So it's just a visual way to think 1431 01:07:48,000 --> 01:07:51,468 about what will happen over time with innovation. 1432 01:07:51,468 --> 01:07:53,010 But I think innovation will pull down 1433 01:07:53,010 --> 01:08:02,300 the slope towards that from where we are today because 1434 01:08:02,300 --> 01:08:04,670 of the newer technology, this decentralized 1435 01:08:04,670 --> 01:08:08,070 how can we do this. 1436 01:08:08,070 --> 01:08:11,850 And I think there will be more opportunities around it. 1437 01:08:11,850 --> 01:08:12,350 All right. 1438 01:08:12,350 --> 01:08:16,130 So on the 30th, we're going to be talking 1439 01:08:16,130 --> 01:08:20,620 about basically our first area. 1440 01:08:20,620 --> 01:08:22,092 We're going to talk about payments. 1441 01:08:22,092 --> 01:08:23,800 We're going to talk a lot about payments. 1442 01:08:23,800 --> 01:08:26,380 I'm going to take two days to talk about payments. 1443 01:08:26,380 --> 01:08:29,710 So we're going to lay out a lot about what's 1444 01:08:29,710 --> 01:08:33,700 going on in finance outside of append-only logs and consensus 1445 01:08:33,700 --> 01:08:36,840 protocols and blockchain. 1446 01:08:36,840 --> 01:08:40,029 But we're going to have fun because what we all 1447 01:08:40,029 --> 01:08:40,810 live together. 1448 01:08:40,810 --> 01:08:42,852 And some of you actually live in these countries. 1449 01:08:42,852 --> 01:08:47,290 But whether it's Alipay, [INAUDIBLE],, and M-Pesa, so 1450 01:08:47,290 --> 01:08:49,810 all these initiatives. 1451 01:08:49,810 --> 01:08:55,120 And then how could maybe blockchain fit into it? 1452 01:08:55,120 --> 01:08:59,020 In terms of the use cases, these are all those use cases 1453 01:08:59,020 --> 01:09:05,290 that we're going to be doing from October 30 to December 6. 1454 01:09:05,290 --> 01:09:08,560 But this gives you a sense, what I call Act Three. 1455 01:09:08,560 --> 01:09:10,359 And we're going to really be trying 1456 01:09:10,359 --> 01:09:14,300 to challenge each other about-- 1457 01:09:14,300 --> 01:09:17,420 some weeks, I'm going to give you a heads up, 1458 01:09:17,420 --> 01:09:19,040 it's going to be a little thin. 1459 01:09:19,040 --> 01:09:20,630 I'm going to be looking for who's 1460 01:09:20,630 --> 01:09:22,008 got really good use cases. 1461 01:09:22,008 --> 01:09:24,050 If you have something to bring into the classroom 1462 01:09:24,050 --> 01:09:26,510 that you're saying, hey, this reading 1463 01:09:26,510 --> 01:09:29,120 list I put together in late August-- 1464 01:09:29,120 --> 01:09:31,250 and by November, it might be that there's 1465 01:09:31,250 --> 01:09:34,729 somebody that's actually moved something further. 1466 01:09:34,729 --> 01:09:35,899 So don't be bashful. 1467 01:09:35,899 --> 01:09:38,060 Shoot it into me in time and I'll distribute it 1468 01:09:38,060 --> 01:09:39,740 to the whole class. 1469 01:09:39,740 --> 01:09:44,479 But in a couple of places, digital ID, for instance, 1470 01:09:44,479 --> 01:09:46,970 I thought it was thin. 1471 01:09:46,970 --> 01:09:50,600 There's the Estonia case, but it's not really-- 1472 01:09:50,600 --> 01:09:52,310 not sure it's a blockchain. 1473 01:09:56,630 --> 01:09:58,280 I'm admitting in advance, some of these 1474 01:09:58,280 --> 01:10:02,810 are a little thin but still good. 1475 01:10:02,810 --> 01:10:05,330 And the readings for the 30th-- 1476 01:10:05,330 --> 01:10:07,040 the Federal Reserve Payment Study really 1477 01:10:07,040 --> 01:10:08,510 shows a bunch of information. 1478 01:10:08,510 --> 01:10:11,893 And you have to read the words, just look at the few charts. 1479 01:10:11,893 --> 01:10:13,310 But it's going to give you trends. 1480 01:10:13,310 --> 01:10:17,180 You're going to really see some interesting trends in there. 1481 01:10:17,180 --> 01:10:20,450 What the best mobile apps are, and the global payment report, 1482 01:10:20,450 --> 01:10:22,640 and things like that. 1483 01:10:22,640 --> 01:10:26,450 Any questions on any of that? 1484 01:10:26,450 --> 01:10:28,220 And then what did we talk about? 1485 01:10:28,220 --> 01:10:30,710 Blockchain technology can address costs 1486 01:10:30,710 --> 01:10:33,980 through verification and networking. 1487 01:10:33,980 --> 01:10:38,730 The potential use cases-- the devil is in the detail. 1488 01:10:38,730 --> 01:10:41,510 It's true in most things in life, 1489 01:10:41,510 --> 01:10:44,930 but it really is these natty little details 1490 01:10:44,930 --> 01:10:46,760 about what transactions, who am I going 1491 01:10:46,760 --> 01:10:50,900 to try to disintermediate, where am I going to create value, 1492 01:10:50,900 --> 01:10:55,100 why use a permissioned or permissionless 1493 01:10:55,100 --> 01:10:58,060 blockchain rather than a traditional database, 1494 01:10:58,060 --> 01:11:02,750 and really get some value. 1495 01:11:02,750 --> 01:11:06,380 I think you really have to address this question, 1496 01:11:06,380 --> 01:11:08,600 but I think there's potential there. 1497 01:11:08,600 --> 01:11:11,660 Or why a public blockchain versus a private? 1498 01:11:11,660 --> 01:11:15,110 Is a token going to help you jump start something? 1499 01:11:15,110 --> 01:11:17,290 And remember, there's two different types of tokens. 1500 01:11:17,290 --> 01:11:20,070 There's the broad currency type token. 1501 01:11:20,070 --> 01:11:22,900 That might be Bitcoin, and they might have won the whole thing, 1502 01:11:22,900 --> 01:11:25,873 or it'll be somebody else, or there might be two or three. 1503 01:11:25,873 --> 01:11:28,165 I personally don't think there'll be hundreds of those. 1504 01:11:30,670 --> 01:11:32,790 But then there might be, in essence, 1505 01:11:32,790 --> 01:11:37,000 use case-specific tokens. 1506 01:11:37,000 --> 01:11:39,550 Right now we have 1,600, and most will fail. 1507 01:11:44,450 --> 01:11:45,540 So that's it. 1508 01:11:45,540 --> 01:11:49,274 Any other queries or anything? 1509 01:11:49,274 --> 01:11:50,650 Yeah? 1510 01:11:50,650 --> 01:11:56,930 AUDIENCE: Once all the bitcoins are mined in 2040 or 2030-- 1511 01:11:56,930 --> 01:11:59,310 [INTERPOSING VOICES] 1512 01:11:59,310 --> 01:12:00,880 AUDIENCE: 2040. 1513 01:12:00,880 --> 01:12:05,280 GARY GENSLER: But by 2040, it's a very slow inflation curve. 1514 01:12:05,280 --> 01:12:06,890 It's probably a slow inflation curve 1515 01:12:06,890 --> 01:12:08,810 in the middle of the 2030s. 1516 01:12:08,810 --> 01:12:10,430 AUDIENCE: The incentives, if there 1517 01:12:10,430 --> 01:12:14,240 will continue to be incentives for node operators 1518 01:12:14,240 --> 01:12:17,540 to continue operating benevolently, if there's 1519 01:12:17,540 --> 01:12:18,530 no further incentives-- 1520 01:12:18,530 --> 01:12:24,020 GARY GENSLER: So the question is, what happens in 20-- 1521 01:12:26,840 --> 01:12:29,300 when there's very few mining rewards? 1522 01:12:29,300 --> 01:12:30,590 And the answer is-- 1523 01:12:30,590 --> 01:12:31,730 I'm going to have some fun with you here. 1524 01:12:31,730 --> 01:12:33,313 I'm going to show you one other thing. 1525 01:12:35,720 --> 01:12:40,340 The answer is, yes, because it's fees. 1526 01:12:42,990 --> 01:12:45,860 And the fees in the system-- 1527 01:12:49,878 --> 01:12:51,070 I apologize. 1528 01:12:51,070 --> 01:12:54,640 It's just one thing that I want to show you that's fun. 1529 01:12:54,640 --> 01:12:55,900 The fees in the system-- 1530 01:13:04,180 --> 01:13:08,680 so right now you can have fees, if you remember how 1531 01:13:08,680 --> 01:13:11,290 there are fees in the system. 1532 01:13:11,290 --> 01:13:15,280 And instead of having mining rewards, 1533 01:13:15,280 --> 01:13:17,890 right now there are very small fees. 1534 01:13:17,890 --> 01:13:20,500 So for every transaction you put into Bitcoin, 1535 01:13:20,500 --> 01:13:24,190 you can actually say, instead of taking at exactly 1536 01:13:24,190 --> 01:13:25,890 that the inputs equal the outputs, 1537 01:13:25,890 --> 01:13:30,910 you have inputs greater than the outputs, and the net is fees. 1538 01:13:30,910 --> 01:13:45,640 And what is the case there is that in 2030 1539 01:13:45,640 --> 01:13:48,010 or something like that, I think the fees 1540 01:13:48,010 --> 01:13:50,770 will continue to increase, I think, if this takes off. 1541 01:13:50,770 --> 01:13:53,500 The question is, does it take off? 1542 01:13:53,500 --> 01:13:56,470 So I just had some fun because this day is an important day 1543 01:13:56,470 --> 01:14:00,070 for me since you met him. 1544 01:14:00,070 --> 01:14:04,616 So whom is whom? 1545 01:14:04,616 --> 01:14:07,352 [LAUGHTER] 1546 01:14:08,720 --> 01:14:11,390 I sent this picture to Rob this morning 1547 01:14:11,390 --> 01:14:14,237 because today is our birthday. 1548 01:14:14,237 --> 01:14:17,111 [CHEERING, APPLAUSE] 1549 01:14:18,798 --> 01:14:20,590 AUDIENCE: (SINGING) Happy birthday to you-- 1550 01:14:20,590 --> 01:14:21,850 GARY GENSLER: No, no, no, no, no, no, no. 1551 01:14:21,850 --> 01:14:22,784 That's all right. 1552 01:14:22,784 --> 01:14:25,840 AUDIENCE: (SINGING) Happy birthday to you. 1553 01:14:25,840 --> 01:14:27,500 GARY GENSLER: No, no, no, no. 1554 01:14:27,500 --> 01:14:31,000 AUDIENCE: (SINGING) Happy birthday, dear Gary. 1555 01:14:31,000 --> 01:14:35,594 Happy birthday to you. 1556 01:14:35,594 --> 01:14:37,482 [APPLAUSE] 1557 01:14:38,335 --> 01:14:40,806 GARY GENSLER: So who's who? 1558 01:14:40,806 --> 01:14:42,590 AUDIENCE: You're in the white. 1559 01:14:42,590 --> 01:14:46,430 GARY GENSLER: White in striped shirt or not? 1560 01:14:46,430 --> 01:14:47,213 OK. 1561 01:14:47,213 --> 01:14:48,575 AUDIENCE: Solid. 1562 01:14:48,575 --> 01:14:49,565 AUDIENCE: Your stripes. 1563 01:14:49,565 --> 01:14:50,565 GARY GENSLER: All right. 1564 01:14:50,565 --> 01:14:54,040 How many striped shirters? 1565 01:14:54,040 --> 01:14:54,730 All right. 1566 01:14:54,730 --> 01:14:55,230 All right. 1567 01:14:55,230 --> 01:14:57,820 How many solid shirters? 1568 01:14:57,820 --> 01:14:58,930 Wow. 1569 01:14:58,930 --> 01:15:01,180 So I sent it to Rob this morning, 1570 01:15:01,180 --> 01:15:02,680 and I sent it to my three daughters, 1571 01:15:02,680 --> 01:15:03,972 and I sent it to my girlfriend. 1572 01:15:03,972 --> 01:15:06,130 I said, who is this? 1573 01:15:06,130 --> 01:15:07,640 It's split view there, too. 1574 01:15:07,640 --> 01:15:09,490 [LAUGHTER] 1575 01:15:09,490 --> 01:15:13,970 I went with striped shirt, but Rob said he doesn't know. 1576 01:15:13,970 --> 01:15:15,850 [LAUGHTER] 1577 01:15:17,260 --> 01:15:18,640 We were three years old there. 1578 01:15:23,380 --> 01:15:26,120 So that's just a goof around with you. 1579 01:15:26,120 --> 01:15:26,620 All right. 1580 01:15:26,620 --> 01:15:28,620 Thank you all.