WEBVTT 00:00:00.000 --> 00:00:03.040 $45 billion. I want you to just sit with that 00:00:03.040 --> 00:00:05.059 number for a second. $45 billion. It's just, 00:00:05.120 --> 00:00:07.860 it's a staggering figure. It is. It's a number 00:00:07.860 --> 00:00:09.839 so big it stops feeling like money and, you know, 00:00:09.880 --> 00:00:12.039 it starts feeling like an abstraction. But it 00:00:12.039 --> 00:00:14.640 wasn't abstract. It was real value. Very real. 00:00:14.800 --> 00:00:16.519 It was college funds. It was retirement accounts. 00:00:16.780 --> 00:00:19.260 It was the treasury of entire companies. And 00:00:19.260 --> 00:00:22.699 in May of 2022, it didn't just dip. It didn't 00:00:22.699 --> 00:00:25.760 have a bad week. It evaporated. It's arguably 00:00:25.760 --> 00:00:29.440 one of the most violent destructions of wealth 00:00:29.440 --> 00:00:32.000 in the history of financial markets. And I don't 00:00:32.000 --> 00:00:34.140 say that to be hyperbolic. Yeah, I don't think 00:00:34.140 --> 00:00:36.439 you are. If you look at the speed of it, we're 00:00:36.439 --> 00:00:39.979 talking about a top 10 global crypto asset effectively 00:00:39.979 --> 00:00:42.659 going to zero. Not in a month, not in a year, 00:00:42.740 --> 00:00:45.380 but in the span of about 72 hours. And the craziest 00:00:45.380 --> 00:00:47.600 part isn't even the speed. It's what this thing 00:00:47.600 --> 00:00:49.520 was supposed to be. I mean, we aren't talking 00:00:49.520 --> 00:00:52.100 about some speculative meme coin with a dog on 00:00:52.100 --> 00:00:54.159 it that everyone knew was a gamble. Right. We 00:00:54.159 --> 00:00:56.840 are talking about stable coins. The one thing 00:00:56.840 --> 00:00:59.920 in this entire wild west of crypto that was promised, 00:01:00.179 --> 00:01:03.640 literally engineered, to stay boring, to stay 00:01:03.640 --> 00:01:06.219 at exactly one dollar. That's the promise, isn't 00:01:06.219 --> 00:01:09.260 it? The promise was stability. The reality was 00:01:09.260 --> 00:01:12.040 a death spiral. And that's the tragedy of it. 00:01:12.099 --> 00:01:15.659 It preyed on people's search for safety. We are, 00:01:15.700 --> 00:01:17.299 of course, talking about the collapse of Terra 00:01:17.299 --> 00:01:22.379 and its token. And looking back at it now, from 00:01:22.379 --> 00:01:25.659 our vantage point here in 2026, the dust has 00:01:25.659 --> 00:01:27.659 finally settled. We have the court verdicts, 00:01:27.659 --> 00:01:29.560 we have the forensic accounting, and we have, 00:01:29.620 --> 00:01:31.439 I think, the full picture of what happened. We 00:01:31.439 --> 00:01:34.620 do. But even knowing how it ends, the story is 00:01:34.620 --> 00:01:36.920 absolutely wild. It's got a founder on the run 00:01:36.920 --> 00:01:39.540 with fake passports. It's got secret bailouts 00:01:39.540 --> 00:01:42.000 from high -frequency traders. And it has a mechanism 00:01:42.000 --> 00:01:44.629 that... when you really look at it, makes you 00:01:44.629 --> 00:01:46.329 wonder how anyone thought it would work in the 00:01:46.329 --> 00:01:48.689 first place. It's a case study in hubris. It's 00:01:48.689 --> 00:01:50.650 a story about what happens when you try to reinvent 00:01:50.650 --> 00:01:52.769 the wheel, specifically the concept of money 00:01:52.769 --> 00:01:56.689 itself. But you ignore the physics of the market. 00:01:56.750 --> 00:01:59.310 You think you're smarter than centuries of financial 00:01:59.310 --> 00:02:01.890 history. So here is the mission for this deep 00:02:01.890 --> 00:02:05.069 dive. We are going to peel back. All the layers 00:02:05.069 --> 00:02:07.510 of the Terra ecosystem. We're going to move past 00:02:07.510 --> 00:02:10.430 the headlines of crypto crash and actually understand 00:02:10.430 --> 00:02:12.550 the mechanics. I want to know exactly how this 00:02:12.550 --> 00:02:15.250 algorithmic stablecoin worked, why it failed, 00:02:15.370 --> 00:02:18.509 and why smart people, and I mean really smart 00:02:18.509 --> 00:02:21.370 institutional investors, poured hundreds of millions 00:02:21.370 --> 00:02:23.090 of dollars into it. And we have to talk about 00:02:23.090 --> 00:02:25.370 the human element. You can't separate the tech 00:02:25.370 --> 00:02:29.669 from the people. Do Kuan and Daniel Shin, the 00:02:29.669 --> 00:02:32.909 architects. Absolutely. The charismatic trash 00:02:32.909 --> 00:02:35.530 -talking leader and the quiet, connected operator. 00:02:35.789 --> 00:02:37.550 We're going to track them from the boardrooms 00:02:37.550 --> 00:02:40.550 of Seoul to a private jet on a runaway in Montenegro. 00:02:41.169 --> 00:02:43.050 But before we get to the handcuffs and the prison 00:02:43.050 --> 00:02:45.449 sentences, we have to start with a dream. Because 00:02:45.449 --> 00:02:48.650 in 2018, this didn't look like a scam. To a lot 00:02:48.650 --> 00:02:50.810 of people, it looked like the future. It really 00:02:50.810 --> 00:02:52.889 did. You have to remember the context of 2018. 00:02:53.710 --> 00:02:56.270 Crypto was still trying to find its killer app. 00:02:56.490 --> 00:02:59.169 Bitcoin was great as digital gold, but it was 00:02:59.169 --> 00:03:01.750 too slow and way too volatile for, you know, 00:03:01.770 --> 00:03:03.770 buying a coffee. Right. You don't want your latte 00:03:03.770 --> 00:03:06.990 to cost $5 one day and $7 the next. Exactly. 00:03:07.330 --> 00:03:09.830 And Ethereum was building this amazing world 00:03:09.830 --> 00:03:13.930 computer. But the transaction fees, the gas was 00:03:13.930 --> 00:03:16.590 getting really expensive. The space was desperate 00:03:16.590 --> 00:03:18.150 for something that could be used for payments, 00:03:18.310 --> 00:03:21.150 something fast, cheap and stable. Enter Terraform 00:03:21.150 --> 00:03:24.250 Labs, founded in Seoul, South Korea. Right. And 00:03:24.250 --> 00:03:26.330 as you said, it's founded by Do Kwon and Daniel 00:03:26.330 --> 00:03:28.909 Shin. And this is a key point. These weren't 00:03:28.909 --> 00:03:31.849 two kids hacking away in a basement. No. Daniel 00:03:31.849 --> 00:03:34.310 Shin was already a major tech mogul in Korea. 00:03:34.919 --> 00:03:36.979 He founded a company called Ticket Monster, which 00:03:36.979 --> 00:03:39.699 is a huge e -commerce platform there, a unicorn. 00:03:40.020 --> 00:03:42.680 He brought the business legitimacy, the connections. 00:03:43.060 --> 00:03:45.680 And Duquan. Duquan was the young visionary, the 00:03:45.680 --> 00:03:48.120 Stanford computer science grad, former engineer 00:03:48.120 --> 00:03:51.060 at Apple and Microsoft. He had the technical 00:03:51.060 --> 00:03:53.379 swagger, the belief that he could code a better 00:03:53.379 --> 00:03:55.740 form of money. And they had the backing to match. 00:03:56.270 --> 00:03:58.189 I was looking at the investor list in the notes 00:03:58.189 --> 00:04:01.210 and it is just staggering. It's a who's who of 00:04:01.210 --> 00:04:03.669 crypto venture capital. We're talking Coinbase 00:04:03.669 --> 00:04:06.050 Ventures, Galaxy Digital, Lightspeed Venture 00:04:06.050 --> 00:04:08.210 Partners, Arrington Capital. These are the blue 00:04:08.210 --> 00:04:11.590 bloods. They raised $200 million across various 00:04:11.590 --> 00:04:13.830 rounds. And that's crucial to understand. When 00:04:13.830 --> 00:04:16.069 you see names like that on the cap table, it 00:04:16.069 --> 00:04:18.370 turns off your skepticism alarm. Oh, absolutely. 00:04:18.910 --> 00:04:21.550 You assume that these multi -billion dollar funds 00:04:21.550 --> 00:04:23.990 have done the due diligence. You assume they've 00:04:23.990 --> 00:04:25.990 hired PhDs to look at the map and they've all 00:04:25.990 --> 00:04:29.290 said, yes, this works. It's social proof on an 00:04:29.290 --> 00:04:31.949 institutional scale. So what was the pitch? What 00:04:31.949 --> 00:04:34.269 was this revolutionary idea they were building 00:04:34.269 --> 00:04:37.370 that got everyone so excited? At its core, they 00:04:37.370 --> 00:04:40.149 were building a decentralized economy. And the 00:04:40.149 --> 00:04:43.689 foundation of any economy is its currency. The 00:04:43.689 --> 00:04:46.029 core idea was that for crypto to be useful for 00:04:46.029 --> 00:04:48.449 daily life, you needed a currency that didn't 00:04:48.449 --> 00:04:51.209 fluctuate wildly. If I pay you for a service 00:04:51.209 --> 00:04:53.870 in Bitcoin today and tomorrow Bitcoin drops 20 00:04:53.870 --> 00:04:55.949 percent, I've essentially overpaid or you've 00:04:55.949 --> 00:04:58.850 lost a huge chunk of value. Commerce needs stability. 00:04:59.170 --> 00:05:01.629 OK, so they needed a stablecoin. Now, usually 00:05:01.629 --> 00:05:03.149 when I think of a stablecoin, I think of something 00:05:03.149 --> 00:05:07.129 like USDC or back in the day Tether. Right. And 00:05:07.129 --> 00:05:09.290 the model for those is simple. I like to call 00:05:09.290 --> 00:05:11.920 it the vending machine model. You put a real 00:05:11.920 --> 00:05:14.360 U .S. dollar into the machine, which is a bank 00:05:14.360 --> 00:05:17.519 account audited by a firm, and the machine spits 00:05:17.519 --> 00:05:21.139 out one digital dollar token. One for one. One 00:05:21.139 --> 00:05:24.100 for one. If you want your real dollar back, you 00:05:24.100 --> 00:05:26.540 put the digital dollar back in the machine. It's 00:05:26.540 --> 00:05:28.660 backed, or at least it claims to be, by real 00:05:28.660 --> 00:05:31.160 cash and U .S. treasuries. But Terra didn't want 00:05:31.160 --> 00:05:33.600 to do that. No. To them, that's centralized. 00:05:34.519 --> 00:05:37.339 That requires trusting a company, Circle or Tether, 00:05:37.439 --> 00:05:39.879 to actually hold the dollars. It requires trusting 00:05:39.879 --> 00:05:42.240 the traditional banking system. Terra wanted 00:05:42.240 --> 00:05:44.980 to be purely decentralized. They wanted money 00:05:44.980 --> 00:05:47.399 made from thin air, essentially, but backed by 00:05:47.399 --> 00:05:50.079 math and economic incentives rather than a vault 00:05:50.079 --> 00:05:52.339 full of cash. Okay, this is where we need to 00:05:52.339 --> 00:05:53.899 get into the weeds. Yeah. Because this is the 00:05:53.899 --> 00:05:56.540 heart of the whole thing. The algorithmic stablecoin. 00:05:56.759 --> 00:05:59.980 The burn and mint equilibrium. Every time I try 00:05:59.980 --> 00:06:02.379 to explain this to friends, my eyes glaze over 00:06:02.379 --> 00:06:04.660 halfway through. Help us visualize this. Okay, 00:06:04.740 --> 00:06:06.540 let's strip away all the code and the jargon. 00:06:06.620 --> 00:06:09.319 Let's use an analogy. Imagine a teeter totter 00:06:09.319 --> 00:06:12.319 on a playground. a seesaw i'm with you a teeter 00:06:12.319 --> 00:06:14.420 -totter on the left seat of the teeter -totter 00:06:14.420 --> 00:06:18.699 sits the stable coin terra usd or ust its job 00:06:18.699 --> 00:06:21.660 is to stay perfectly level to hover exactly at 00:06:21.660 --> 00:06:23.720 the height of one dollar it should never go up 00:06:23.720 --> 00:06:26.959 never go down it's the boring stable side okay 00:06:26.959 --> 00:06:29.680 ust is on the left on the right seat sits the 00:06:29.680 --> 00:06:32.860 variable token lna luna is the shock absorber 00:06:32.860 --> 00:06:35.120 it's allowed to go up and down like crazy it's 00:06:35.120 --> 00:06:37.889 designed to be volatile So Ulanae takes all the 00:06:37.889 --> 00:06:40.269 volatility so the USD doesn't have to? Exactly. 00:06:40.430 --> 00:06:42.949 Now, how do they stay balanced? How do they talk 00:06:42.949 --> 00:06:45.250 to each other? They are connected by a special 00:06:45.250 --> 00:06:48.509 rule, a mechanism in the code that allows you, 00:06:48.629 --> 00:06:51.290 the trader, to always swap one for the other 00:06:51.290 --> 00:06:54.769 at a fixed value of $1. Always. Even if the price 00:06:54.769 --> 00:06:57.250 on an exchange like Binance is totally different. 00:06:57.660 --> 00:06:59.899 That is the absolute key. This protocol, the 00:06:59.899 --> 00:07:03.439 core code, always, always values one UST at exactly 00:07:03.439 --> 00:07:06.060 $1. It doesn't care what the open market says. 00:07:06.100 --> 00:07:08.339 It's a hard -coded assumption. So let's play 00:07:08.339 --> 00:07:10.160 on a scenario. Let's say there's huge demand 00:07:10.160 --> 00:07:12.459 for UST. Everyone wants it, maybe to put it into 00:07:12.459 --> 00:07:14.339 a savings protocol we'll talk about later. The 00:07:14.339 --> 00:07:17.139 price on the open market shoots up to, say, $1 00:07:17.139 --> 00:07:20.459 .01. Okay, so it's trading at a premium. UST 00:07:20.459 --> 00:07:24.470 is worth $1 .01. Right. Now you, the savvy arbitrage 00:07:24.470 --> 00:07:27.110 trader, see a free money opportunity. You can 00:07:27.110 --> 00:07:30.250 go to the protocol with $1 worth of Lunae. Let's 00:07:30.250 --> 00:07:32.970 say Lunae is worth $100 a token. You take 11 00:07:32.970 --> 00:07:34.709 hundredths of your Luna Day token, you give it 00:07:34.709 --> 00:07:36.769 to the protocol, and you say, burn this. Burn 00:07:36.769 --> 00:07:38.870 it? You mean delete it? Precisely. You send it 00:07:38.870 --> 00:07:40.750 to a wallet no one can access. It's gone forever. 00:07:41.110 --> 00:07:44.069 And in exchange for burning that $1 worth of 00:07:44.069 --> 00:07:46.949 Lunae, the protocol mints, it creates out of 00:07:46.949 --> 00:07:50.370 thin air one... brand new UST token for you. 00:07:50.490 --> 00:07:53.870 OK. So I burn my Lunanay. I get one UST. Which 00:07:53.870 --> 00:07:56.350 you can immediately go and sell on the open market 00:07:56.350 --> 00:07:58.529 for a dollar and one cent. You just made a penny 00:07:58.529 --> 00:08:01.149 of profit almost completely risk free. A penny 00:08:01.149 --> 00:08:03.269 doesn't sound like much. It isn't. But do it 00:08:03.269 --> 00:08:05.250 with 100 million dollars. Have a computer do 00:08:05.250 --> 00:08:07.750 it a thousand times a second. That is arbitrage. 00:08:08.350 --> 00:08:10.430 And now think about the effect of what you just 00:08:10.430 --> 00:08:13.230 did. You created more UST. You minted it, which 00:08:13.230 --> 00:08:15.829 increases the total supply. And what happens 00:08:15.829 --> 00:08:18.129 in economics when supply goes up? Price comes 00:08:18.129 --> 00:08:21.069 down. Price comes down. So the army of arbitragers 00:08:21.069 --> 00:08:24.029 all chasing these tiny profits naturally push 00:08:24.029 --> 00:08:26.910 the price of UST back down to one dollar. The 00:08:26.910 --> 00:08:28.850 system self -corrects. OK, that makes a weird 00:08:28.850 --> 00:08:30.750 kind of sense. High demand leads to more supply. 00:08:30.990 --> 00:08:33.649 Price stabilizes. What about the other direction? 00:08:33.789 --> 00:08:36.090 This seems like the important one. What if people 00:08:36.090 --> 00:08:38.870 panic and sell UST? and the price on the open 00:08:38.870 --> 00:08:42.590 market drops to 99 cents this is the dangerous 00:08:42.590 --> 00:08:44.509 side of the teeter -totter this is where the 00:08:44.509 --> 00:08:47.230 whole thing can break so if ust is trading on 00:08:47.230 --> 00:08:49.610 the open market at 99 cents the arbitrage opportunity 00:08:49.610 --> 00:08:52.450 is reversed okay you can go into the market buy 00:08:52.450 --> 00:08:56.049 a ust for just 99 cents then you take that ust 00:08:56.049 --> 00:08:58.710 to the protocol and say i want to swap this The 00:08:58.710 --> 00:09:00.830 protocol, remember, is hard -coded. It says, 00:09:00.889 --> 00:09:03.789 sure, I value this UST at $1. So you burn the 00:09:03.789 --> 00:09:07.250 UST, and the protocol mints you $1 worth of brand 00:09:07.250 --> 00:09:10.710 -new Lunanay. So I spent 99 cents, and I get 00:09:10.710 --> 00:09:13.470 $1 worth of Lunanay. I can immediately sell that 00:09:13.470 --> 00:09:14.970 to Lunanay on the market, and I've made a penny. 00:09:15.489 --> 00:09:18.269 Exactly. And in doing so, you burned the UST. 00:09:18.330 --> 00:09:20.330 You took it out of circulation. You decreased 00:09:20.330 --> 00:09:22.429 the supply. Lower supply means the price should 00:09:22.429 --> 00:09:25.429 go up, in theory, back to $1. So the whole system 00:09:25.429 --> 00:09:28.049 relies on these greedy traders acting in their 00:09:28.049 --> 00:09:30.090 own self -interest. They aren't trying to save 00:09:30.090 --> 00:09:32.409 the protocol. They're just trying to make a quick 00:09:32.409 --> 00:09:35.470 buck. And in the process, they keep the peg stable. 00:09:35.870 --> 00:09:38.759 In theory, yes. When you draw it on a whiteboard, 00:09:38.899 --> 00:09:41.759 it's an elegant piece of game theory. It's beautiful 00:09:41.759 --> 00:09:45.220 even. But there is a huge, massive hidden assumption 00:09:45.220 --> 00:09:48.360 here, a fatal flaw that everyone chose to ignore. 00:09:48.539 --> 00:09:51.240 Which is? The system assumes that L &A always 00:09:51.240 --> 00:09:54.139 has value. And not just some value, but enough 00:09:54.139 --> 00:09:58.360 market capitalization to absorb all the UST that 00:09:58.360 --> 00:10:00.159 might want to exit. Because you're paying for 00:10:00.159 --> 00:10:03.210 the UST with newly minted L &A. Right. Remember, 00:10:03.289 --> 00:10:05.190 when you are saving the peg in that downward 00:10:05.190 --> 00:10:07.850 scenario, you are minting LNA. You're creating 00:10:07.850 --> 00:10:10.809 more of it. If you mint more LNA, the supply 00:10:10.809 --> 00:10:12.730 of LNA goes up. Which usually means the price 00:10:12.730 --> 00:10:15.789 of LNA goes down. Supply and demand. Right. So 00:10:15.789 --> 00:10:18.370 you are essentially printing your own company's 00:10:18.370 --> 00:10:21.330 stock to buy back your own company's debt. If 00:10:21.330 --> 00:10:24.129 LMA is valuable, if it has a market cap of $40 00:10:24.129 --> 00:10:26.809 billion like it did at its peak, it can easily 00:10:26.809 --> 00:10:28.909 absorb a few hundred million dollars of shock. 00:10:29.149 --> 00:10:32.629 The price might dip a little, but it holds. But 00:10:32.629 --> 00:10:35.029 what if LMA's price starts crashing at the same 00:10:35.029 --> 00:10:37.389 time? What if people lose faith in the shock 00:10:37.389 --> 00:10:40.450 absorber itself? Then you're in deep, deep trouble. 00:10:40.929 --> 00:10:43.929 Because to get that same $1 of value to bail 00:10:43.929 --> 00:10:47.789 out one U .S. You have to get more and more and 00:10:47.789 --> 00:10:51.149 more LNA as its price falls. Precisely. It's 00:10:51.149 --> 00:10:53.549 a reflexive loop. And we will see exactly how 00:10:53.549 --> 00:10:56.009 that math plays out in the crash section. But 00:10:56.009 --> 00:10:58.269 for a few years, nobody looked at the downside. 00:10:58.450 --> 00:11:00.610 They only saw the upside. Because the ecosystem 00:11:00.610 --> 00:11:03.169 wasn't just tokens. It was exploding with applications. 00:11:03.350 --> 00:11:05.470 They were building that whole city on the blockchain, 00:11:05.610 --> 00:11:07.230 you mentioned. They were. They were really good 00:11:07.230 --> 00:11:09.210 at building an ecosystem to create demand for 00:11:09.210 --> 00:11:11.629 UST. They had the mirror protocol, which was 00:11:11.629 --> 00:11:14.049 kind of brilliant and also, you know, wildly 00:11:14.049 --> 00:11:15.889 illegal. What did it do? It allowed people to 00:11:15.889 --> 00:11:18.710 trade synthetic stocks. You could deposit UST 00:11:18.710 --> 00:11:20.909 and in return, you'd get a token that tracked 00:11:20.909 --> 00:11:24.070 the price of, say, Tesla stock or Apple stock. 00:11:24.570 --> 00:11:26.330 So you could get exposure to the U .S. stock 00:11:26.330 --> 00:11:28.250 market without ever leaving the crypto world. 00:11:28.389 --> 00:11:31.269 Exactly. The SEC hated it, obviously, because 00:11:31.269 --> 00:11:33.950 it bypassed all securities laws. But for users 00:11:33.950 --> 00:11:36.570 in countries without easy access to U .S. markets, 00:11:36.769 --> 00:11:39.649 it was a killer app. And then there was Chai. 00:11:39.789 --> 00:11:42.090 This was the thing that always gave Terra so 00:11:42.090 --> 00:11:44.070 much credibility. I remember reading all these 00:11:44.070 --> 00:11:46.690 articles saying Terra is the only crypto actually 00:11:46.690 --> 00:11:49.490 being used for real payments by real people. 00:11:49.669 --> 00:11:52.720 That was the narrative. Daniel Shin's app, Chai, 00:11:52.840 --> 00:11:55.039 was a popular payment service in South Korea, 00:11:55.179 --> 00:11:57.419 and they claimed to be processing millions of 00:11:57.419 --> 00:11:59.340 transactions on the back end using the Terra 00:11:59.340 --> 00:12:03.240 blockchain. The story was, look, grandma in Seoul 00:12:03.240 --> 00:12:05.480 is buying groceries with crypto and she doesn't 00:12:05.480 --> 00:12:07.860 even know it. It provided this shield of real 00:12:07.860 --> 00:12:10.240 world utility. It made it seem less like a financial 00:12:10.240 --> 00:12:12.840 game and more like real technology adoption. 00:12:13.159 --> 00:12:15.620 It was the perfect story. But as we later find 00:12:15.620 --> 00:12:17.779 out during the investigations by Korean prosecutors, 00:12:18.220 --> 00:12:22.700 much of that claimed volume was. let's say, questionable. 00:12:23.460 --> 00:12:25.840 The connection to the blockchain wasn't as deep 00:12:25.840 --> 00:12:28.320 or as real as they claimed. We'll get to the 00:12:28.320 --> 00:12:30.860 fraud allegations. But before we do, we have 00:12:30.860 --> 00:12:33.259 to talk about the hook, the thing that really 00:12:33.259 --> 00:12:36.159 poured gasoline on this fire. Because a clever 00:12:36.159 --> 00:12:38.580 algorithm is nice, but it doesn't get you to 00:12:38.580 --> 00:12:41.460 a $40 billion market cap on its own. You need 00:12:41.460 --> 00:12:44.059 something more primal. You need greed. You need 00:12:44.059 --> 00:12:46.399 the anchor protocol. Anchor was the black hole 00:12:46.399 --> 00:12:48.279 that sucked in all the liquidity in the crypto 00:12:48.279 --> 00:12:51.679 market. It was, on the surface, a lending and 00:12:51.679 --> 00:12:55.139 borrowing protocol launched in March 2021, but 00:12:55.139 --> 00:12:56.879 really it only had one feature that mattered. 00:12:57.019 --> 00:12:59.429 The savings account. And I use air quotes because, 00:12:59.509 --> 00:13:03.149 well, tell us the number. 19 .45 % APY. Let's 00:13:03.149 --> 00:13:05.669 call it 20%. 20 % interest. And this wasn't a 00:13:05.669 --> 00:13:07.610 fluctuating rate, not initially. It was marketed 00:13:07.610 --> 00:13:10.830 as a stable, predictable 20%. Now, let's contextualize 00:13:10.830 --> 00:13:13.230 this for a second. In 2021, the Federal Reserve 00:13:13.230 --> 00:13:15.769 interest rate was new zero. I remember my savings 00:13:15.769 --> 00:13:17.870 account at Wells Fargo was paying, I think, 0 00:13:17.870 --> 00:13:21.279 .01%. Right. You were actively losing money to 00:13:21.279 --> 00:13:23.840 inflation by keeping it in a bank. And here comes 00:13:23.840 --> 00:13:26.559 Duke Kwan and Terra saying, give me your dollars, 00:13:26.820 --> 00:13:29.759 convert them to our stablecoin UST, park them 00:13:29.759 --> 00:13:32.139 in this protocol called Anchor, and we will pay 00:13:32.139 --> 00:13:35.080 you 20 % a year. It was irresistible. It was 00:13:35.080 --> 00:13:38.080 the best marketing tool ever invented. And it 00:13:38.080 --> 00:13:40.700 wasn't just for crypto bros. We are talking about 00:13:40.700 --> 00:13:43.779 regular people. We are talking about fintech 00:13:43.779 --> 00:13:46.019 startups that built their entire business model 00:13:46.019 --> 00:13:48.159 on the back end of Anchor. They take customer 00:13:48.159 --> 00:13:50.559 to p***. deposits, promised them 8 % yield, put 00:13:50.559 --> 00:13:53.399 it all in Anchor to get 20%, and pocket the difference. 00:13:53.759 --> 00:13:55.600 So people were using this who didn't even know 00:13:55.600 --> 00:13:57.179 what a blockchain was. They just thought they 00:13:57.179 --> 00:13:59.620 found a high -yield savings account. A huge number. 00:13:59.879 --> 00:14:02.759 At its peak, Anchor had something like $17 billion 00:14:02.759 --> 00:14:06.039 in deposits. Most of the entire supply of UST 00:14:06.039 --> 00:14:08.460 was locked up in Anchor. But hold on. I mean, 00:14:08.460 --> 00:14:10.299 it's a fundamental question. How do you generate 00:14:10.299 --> 00:14:13.500 20 % risk -free? If I walk into a bank and ask 00:14:13.500 --> 00:14:17.100 for a loan, they charge me maybe 5 % or 6 % interest. 00:14:17.610 --> 00:14:20.070 If Anchor's paying depositors 20%, they must 00:14:20.070 --> 00:14:23.570 be charging borrowers. What? 25%, 30%. Who's 00:14:23.570 --> 00:14:26.210 borrowing money at 30 %? Nobody. That was the 00:14:26.210 --> 00:14:29.190 problem. The borrowing demand on Anchor was nowhere 00:14:29.190 --> 00:14:32.330 near high enough to pay the depositors. The protocol 00:14:32.330 --> 00:14:35.289 was earning maybe 5%, 6 % from lending and staking 00:14:35.289 --> 00:14:38.649 rewards, but it was paying out 20%. The math 00:14:38.649 --> 00:14:41.169 just didn't add up. So where was the extra 14, 00:14:41.389 --> 00:14:44.450 15 % coming from? It was subsidized. It was coming 00:14:44.450 --> 00:14:47.389 directly from Terraform Labs and later the Luna 00:14:47.389 --> 00:14:49.450 Foundation Guard. They created something called 00:14:49.450 --> 00:14:52.309 the Yield Reserve, and they just dumped hundreds 00:14:52.309 --> 00:14:54.429 of millions of their own dollars into it. So 00:14:54.429 --> 00:14:57.049 it was a marketing budget. It was purely a customer 00:14:57.049 --> 00:14:58.990 acquisition cost. They were burning their own 00:14:58.990 --> 00:15:01.470 cash to pay out that 20 % just to get people 00:15:01.470 --> 00:15:04.669 to buy and hold UST. The goal was to get UST 00:15:04.669 --> 00:15:07.549 so big, so integrated into the crypto economy 00:15:07.549 --> 00:15:09.710 that it would become too big to fail. So they 00:15:09.710 --> 00:15:11.590 were literally paying people to use their money. 00:15:11.750 --> 00:15:14.149 Yes. And critics saw this. There were voices, 00:15:14.289 --> 00:15:16.529 loud voices on Twitter and in crypto media saying 00:15:16.529 --> 00:15:19.309 this is a Ponzi structure. They laid it out perfectly. 00:15:19.549 --> 00:15:21.929 If the only way you can pay the yield to existing 00:15:21.929 --> 00:15:24.509 investors is by bringing in new money or by running 00:15:24.509 --> 00:15:26.929 your own reserves, that is the definition of 00:15:26.929 --> 00:15:29.549 an unsustainable system. Eventually, the reserves 00:15:29.549 --> 00:15:32.740 run dry. And how did Duquan respond to these 00:15:32.740 --> 00:15:34.320 critics? Because this is where his personality 00:15:34.320 --> 00:15:37.159 becomes a major part of the story. Duquan was 00:15:37.159 --> 00:15:40.340 aggressive. That's the polite word. He was a 00:15:40.340 --> 00:15:43.179 prolific tweeter. His handle, ironically, was 00:15:43.179 --> 00:15:46.480 at Spablequan. And whenever anyone with a serious 00:15:46.480 --> 00:15:48.419 following questioned the mechanics, whenever 00:15:48.419 --> 00:15:50.220 anyone asked, hey, what happens when the yield 00:15:50.220 --> 00:15:53.519 reserve runs out? He would mock them. He would 00:15:53.519 --> 00:15:55.919 belittle them. He had that famous line. I don't 00:15:55.919 --> 00:15:58.019 debate the poor on Twitter. I don't debate the 00:15:58.019 --> 00:16:01.320 poor. Just imagine the CEO of a major financial 00:16:01.320 --> 00:16:04.320 institution saying that. Jamie Dimon or something. 00:16:04.399 --> 00:16:07.220 It's unthinkable. It is. But in the bubble of 00:16:07.220 --> 00:16:10.360 crypto in 2021, it worked. It created a cult 00:16:10.360 --> 00:16:13.200 of personality. If you held ONA, you were a lunatic. 00:16:13.299 --> 00:16:15.399 That was the actual name for the community. And 00:16:15.399 --> 00:16:18.919 they viewed Duquan as this genius god king who 00:16:18.919 --> 00:16:21.059 had figured out how to solve money. Any criticism 00:16:21.059 --> 00:16:24.230 was just FUD fear. uncertainty and doubt. It 00:16:24.230 --> 00:16:26.090 was dismissed as noise from people who were just 00:16:26.090 --> 00:16:28.889 salty. They missed the boat. And the hubris wasn't 00:16:28.889 --> 00:16:30.610 just on Twitter. It spilled out into the real 00:16:30.610 --> 00:16:32.690 world. I mean, the Washington Nationals deal. 00:16:32.990 --> 00:16:35.830 Oh, that was peak bubble behavior. This was February 00:16:35.830 --> 00:16:38.509 2022. We are now just three months before everything 00:16:38.509 --> 00:16:41.830 collapsed. Terraform Labs signs a massive sponsorship 00:16:41.830 --> 00:16:44.190 deal with the MLB team, the Washington Nationals. 00:16:44.269 --> 00:16:48.789 For how much? $38 .5 million. Wow. A five -year 00:16:48.789 --> 00:16:51.570 deal. They rebranded the premium lounge right 00:16:51.570 --> 00:16:54.259 behind home plate as the Terra Club. There were 00:16:54.259 --> 00:16:58.080 ads for a UST in a baseball stadium. They were 00:16:58.080 --> 00:17:01.019 trying to market an algorithmic stable coin to 00:17:01.019 --> 00:17:03.960 mainstream sports fans. And the way that deal 00:17:03.960 --> 00:17:06.240 happened is wild, too. Kaywon proposed it to 00:17:06.240 --> 00:17:08.720 the community for a governance vote, right? Yes, 00:17:08.799 --> 00:17:11.619 that was the veneer of decentralization. But 00:17:11.619 --> 00:17:13.400 he didn't even say which team it was. The proposal 00:17:13.400 --> 00:17:15.900 just said something like a sponsorship opportunity 00:17:15.900 --> 00:17:18.319 with a sports franchise in one of the four major 00:17:18.319 --> 00:17:20.539 American leagues. He basically said, trust me, 00:17:20.599 --> 00:17:23.000 it's good marketing. Give me the money. And the 00:17:23.000 --> 00:17:26.839 community, the lunatics, blindly voted yes. They 00:17:26.839 --> 00:17:29.559 authorized spending $38 million of the community 00:17:29.559 --> 00:17:32.359 treasury on a mystery box. That is the definition 00:17:32.359 --> 00:17:35.119 of too much money, too little diligence. But 00:17:35.119 --> 00:17:37.819 even as they were buying baseball lounges, behind 00:17:37.819 --> 00:17:39.759 the scenes, the cracks were starting to show. 00:17:39.960 --> 00:17:42.619 They knew the algorithm wasn't bulletproof. They 00:17:42.619 --> 00:17:45.630 did. In January 2022, you start to see a subtle 00:17:45.630 --> 00:17:48.250 pivot in their strategy. They realized that if 00:17:48.250 --> 00:17:51.289 a massive panic happened, a true bank run, the 00:17:51.289 --> 00:17:53.490 mint and burn mechanism might not be fast enough 00:17:53.490 --> 00:17:56.089 or strong enough. Lunae's price might crash too 00:17:56.089 --> 00:17:59.549 hard, too fast. So they created a new entity, 00:17:59.869 --> 00:18:03.529 the Luna Foundation Guard. Or LFG. A nonprofit 00:18:03.529 --> 00:18:06.170 based in Singapore. Correct. And the stated goal 00:18:06.170 --> 00:18:08.869 of the LFG was to build a war chest, a reserve 00:18:08.869 --> 00:18:11.710 of assets that were not Lunanay or UST. They 00:18:11.710 --> 00:18:13.650 wanted hard collateral, exogenous collateral. 00:18:13.970 --> 00:18:16.210 So they started buying Bitcoin. Aggressively. 00:18:16.309 --> 00:18:18.869 Very aggressively. Caron went on podcasts and 00:18:18.869 --> 00:18:20.569 said his goal was to become the largest single 00:18:20.569 --> 00:18:22.809 holder of Bitcoin in the world, second only to 00:18:22.809 --> 00:18:26.069 Satoshi Nakamoto. By early May 2022, just before 00:18:26.069 --> 00:18:28.410 the crash, they had accumulated over 80 ,000 00:18:28.410 --> 00:18:30.630 Bitcoin. Which was worth how much at the time? 00:18:30.769 --> 00:18:35.720 About $2. And the plan was, if the peg breaks, 00:18:35.920 --> 00:18:39.220 if our algorithm fails, we'll sell all this Bitcoin 00:18:39.220 --> 00:18:42.259 to buy UST off the market and prop up the price. 00:18:42.799 --> 00:18:44.859 Exactly. It was a backstop. It was an emergency 00:18:44.859 --> 00:18:47.240 fund. But think about the irony of that. They 00:18:47.240 --> 00:18:50.579 spent years, years saying our algorithm is superior. 00:18:50.700 --> 00:18:52.519 We don't need collateral like those other guys. 00:18:53.369 --> 00:18:55.450 And then right at the end, they panicked and 00:18:55.450 --> 00:18:58.269 bought $2 .5 billion worth of collateral. It's 00:18:58.269 --> 00:18:59.950 like a tightrope walker who spends the whole 00:18:59.950 --> 00:19:01.910 show bragging about how he doesn't need a net. 00:19:02.109 --> 00:19:04.430 And then five minutes before the grand finale, 00:19:04.789 --> 00:19:07.789 he quietly asks his assistant to go and put a 00:19:07.789 --> 00:19:09.650 mattress down below. A mattress that turned out 00:19:09.650 --> 00:19:11.750 to be the size of a postage stamp compared to 00:19:11.750 --> 00:19:13.450 the height of the fall. So let's move to May 00:19:13.450 --> 00:19:17.349 2022, the collapse. Because this wasn't a slow 00:19:17.349 --> 00:19:19.869 decline. This was a free fall. It was a cliff. 00:19:20.049 --> 00:19:22.109 And we can pretty much trace the trigger event. 00:19:22.640 --> 00:19:24.960 It started over the weekend of May 7th and 8th. 00:19:25.000 --> 00:19:27.500 The broader crypto market was already jittery. 00:19:27.500 --> 00:19:29.400 The Fed was raising interest rates. Things were 00:19:29.400 --> 00:19:32.259 looking shaky. But then a series of massive sell 00:19:32.259 --> 00:19:35.740 orders hit the main UST liquidity pools on an 00:19:35.740 --> 00:19:38.480 exchange called Curve. Curve is a decentralized 00:19:38.480 --> 00:19:41.640 exchange where people mostly swap stablecoins 00:19:41.640 --> 00:19:44.119 with each other. Exactly. It's the main plumbing 00:19:44.119 --> 00:19:47.609 for stablecoin trading in DeFi. Someone, and 00:19:47.609 --> 00:19:50.410 we still debate exactly who, though later evidence 00:19:50.410 --> 00:19:52.990 points towards some large trading firms unwinding 00:19:52.990 --> 00:19:55.809 their positions, started dumping hundreds of 00:19:55.809 --> 00:19:58.869 millions of dollars of UST onto the market in 00:19:58.869 --> 00:20:01.930 a very short period. So suddenly there's way 00:20:01.930 --> 00:20:05.839 more UST being sold. than being bought and the 00:20:05.839 --> 00:20:08.359 peg slipped it wasn't a catastrophe at first 00:20:08.359 --> 00:20:11.279 it went to 99 cents then 98 cents now remember 00:20:11.279 --> 00:20:13.200 the arbitrage mechanism we talked about right 00:20:13.200 --> 00:20:15.359 the traders should step in buy the cheap ust 00:20:15.359 --> 00:20:18.680 at 98 cents Burn it with the protocol for a dollar's 00:20:18.680 --> 00:20:21.119 worth of LNA and make a profit. And they tried. 00:20:21.339 --> 00:20:23.579 For a little while, they did. But the selling 00:20:23.579 --> 00:20:25.940 pressure was just too high. It was a tidal wave. 00:20:26.259 --> 00:20:28.579 There was too much UST flooding the exit all 00:20:28.579 --> 00:20:31.319 at once. The arbitrageurs couldn't keep up. And 00:20:31.319 --> 00:20:33.319 this is where the death spiral mechanics took 00:20:33.319 --> 00:20:35.200 over. Walk us through the math of the spiral. 00:20:35.279 --> 00:20:38.539 Why did it get so bad so fast? Okay. So, to save 00:20:38.539 --> 00:20:42.269 the peg, the protocol has to mint LNA. If I come 00:20:42.269 --> 00:20:45.230 with one UST that's worth, say, 90 cents on the 00:20:45.230 --> 00:20:48.170 market, the protocol still owes me $1 worth of 00:20:48.170 --> 00:20:51.529 Lunay. When Lunay is trading at $80 a token, 00:20:51.690 --> 00:20:54.130 which it was right before the crash, the protocol 00:20:54.130 --> 00:20:57.230 only needs to mint a tiny fraction of a Lunay 00:20:57.230 --> 00:20:59.849 token 180th to pay me my dollar. Right. It's 00:20:59.849 --> 00:21:01.809 a small amount of new supply. The dilution is 00:21:01.809 --> 00:21:04.259 minimal. But as the panic set in and the peg 00:21:04.259 --> 00:21:07.200 broke more severely, people started selling Lunanay, 00:21:07.200 --> 00:21:09.099 too. They saw the system was failing. They got 00:21:09.099 --> 00:21:10.920 scared. They dumped their Lunanay. So the price 00:21:10.920 --> 00:21:15.279 of Lunanay dropped from $80 to, say, $40. Okay, 00:21:15.319 --> 00:21:17.900 so now to pay me my $1, the protocol has to mint 00:21:17.900 --> 00:21:20.920 twice as much Lunanay as before. Exactly. And 00:21:20.920 --> 00:21:23.180 that new Lunanay hits the open market, further 00:21:23.180 --> 00:21:25.119 diluting the supply, which pushes the price down 00:21:25.119 --> 00:21:27.599 even more, to $20. Now you have to mint four 00:21:27.599 --> 00:21:30.099 times as much Lunanay to cover that same $1 of 00:21:30.099 --> 00:21:33.079 UST debt. Then Lunanay hits $1. Now you are minting 00:21:33.079 --> 00:21:36.039 one full Lunanay token for every single UST being 00:21:36.039 --> 00:21:39.000 redeemed. And then Lunanay hits $0 .10. Now you 00:21:39.000 --> 00:21:43.579 are minting 10 Alunays for every UST. Do you 00:21:43.579 --> 00:21:46.319 see the exponential curve? As the price of UNA 00:21:46.319 --> 00:21:49.119 approaches zero, the amount of new UNA you have 00:21:49.119 --> 00:21:51.200 to mint to cover the debt approaches infinity. 00:21:51.599 --> 00:21:54.480 It's hyperinflation on steroids. A feedback loop 00:21:54.480 --> 00:21:56.579 from hell. It was terrifying to watch in real 00:21:56.579 --> 00:21:59.920 time. The supply of LNA went from about 350 million 00:21:59.920 --> 00:22:04.519 tokens in circulation to get this 6 .5 trillion 00:22:04.519 --> 00:22:07.259 tokens in the span of a few days. The algorithm 00:22:07.259 --> 00:22:10.039 was just mindlessly printing LNA to try and cover 00:22:10.039 --> 00:22:12.180 the debt. But the debt was toxic and the currency 00:22:12.180 --> 00:22:14.180 it was printing was becoming worthless. So what 00:22:14.180 --> 00:22:16.240 was the Luna Foundation Guard doing with that 00:22:16.240 --> 00:22:19.019 $2 .4 billion of Bitcoin, the emergency fund? 00:22:19.180 --> 00:22:21.880 They deployed it. On May 9th and 10th, they completely 00:22:21.880 --> 00:22:24.160 emptied the vaults. They sent the Bitcoin to 00:22:24.160 --> 00:22:26.339 exchange. like Binance and Gemini and started 00:22:26.339 --> 00:22:29.119 market selling it to buy UST. Did it work? Did 00:22:29.119 --> 00:22:31.920 it stabilize the price? For a few hours, it created 00:22:31.920 --> 00:22:34.880 a bounce. The peg went from a low of around 60 00:22:34.880 --> 00:22:38.079 cents back up towards 90 cents. There was a moment 00:22:38.079 --> 00:22:40.880 of hope. But the market simply absorbed all that 00:22:40.880 --> 00:22:43.500 capital and kept selling. Two and a half billion 00:22:43.500 --> 00:22:45.920 dollars just vanished. It was like trying to 00:22:45.920 --> 00:22:48.359 fill the Grand Canyon with a garden hose. And 00:22:48.359 --> 00:22:50.920 worse, by dumping all that Bitcoin so quickly, 00:22:51.480 --> 00:22:54.180 They crashed the price of Bitcoin, too. So the 00:22:54.180 --> 00:22:56.839 entire crypto market tanked along with them. 00:22:56.980 --> 00:22:59.359 I remember looking at Twitter and Reddit during 00:22:59.359 --> 00:23:02.369 those days. It was pure panic. There were suicide 00:23:02.369 --> 00:23:04.410 hotlines being pinned to the top of the terrorist 00:23:04.410 --> 00:23:06.769 subreddits by the moderators. It was horrific. 00:23:06.970 --> 00:23:08.990 You had people who had put their life savings 00:23:08.990 --> 00:23:10.869 into Anchor because they were told it was a stable 00:23:10.869 --> 00:23:12.730 savings account. They didn't think they were 00:23:12.730 --> 00:23:14.769 gambling on some altcoin. There were people in 00:23:14.769 --> 00:23:17.029 their 60s who put their retirement in. They woke 00:23:17.029 --> 00:23:20.029 up and saw their $500 ,000 portfolio was now 00:23:20.029 --> 00:23:22.869 worth $500. It was life -altering destruction. 00:23:23.269 --> 00:23:25.109 And the response from Duquan and the team? It 00:23:25.109 --> 00:23:27.809 was chaos. Duquan was tweeting things like, deploying 00:23:27.809 --> 00:23:55.059 more capital steady lads. But on May 13, So the 00:23:55.059 --> 00:24:00.000 dust settles. $45 billion is gone. UST is trading 00:24:00.000 --> 00:24:03.579 at a few cents. LNA is effectively zero. What 00:24:03.579 --> 00:24:06.259 happens next? Where do you even begin to pick 00:24:06.259 --> 00:24:08.240 up the pieces? Well, first comes the confusion 00:24:08.240 --> 00:24:11.839 and the anger. Then comes the rebranding attempt. 00:24:12.079 --> 00:24:15.220 About two weeks later, on May 25th, Terraform 00:24:15.220 --> 00:24:18.599 Labs and Dugaquan push a plan to fork the chain. 00:24:18.759 --> 00:24:20.940 Meaning they split the software into two different 00:24:20.940 --> 00:24:23.059 versions. Yes. They basically said, OK, that 00:24:23.059 --> 00:24:25.619 old chain with the algorithmic stablecoin, that's 00:24:25.619 --> 00:24:27.220 broken. It's a failure. We're going to leave 00:24:27.220 --> 00:24:29.420 it behind. We'll rename it Terra Classic and 00:24:29.420 --> 00:24:33.430 its token will be LUNC. LUNC. Got it. And we 00:24:33.430 --> 00:24:35.789 are going to launch a brand new chain, a fresh 00:24:35.789 --> 00:24:39.089 start called Terra 2 .0, and the new token on 00:24:39.089 --> 00:24:41.710 that chain would take the original Elune ticker. 00:24:42.029 --> 00:24:44.329 But crucially, the new chain didn't have the 00:24:44.329 --> 00:24:46.950 stablecoin. They completely abandoned the UST 00:24:46.950 --> 00:24:49.849 concept. So it was a tacit admission that their 00:24:49.849 --> 00:24:52.829 core product, their whole grand experiment, was 00:24:52.829 --> 00:24:55.059 fundamentally flawed. It was a complete admission 00:24:55.059 --> 00:24:58.019 of failure. They airdropped the new LNA tokens 00:24:58.019 --> 00:25:01.539 to holders of the old LNC and UST to try and 00:25:01.539 --> 00:25:03.880 make them whole. But it was pennies on the dollar. 00:25:04.079 --> 00:25:06.319 And did the market forgive them? Did Terra 2 00:25:06.319 --> 00:25:09.839 .0 succeed? Not really. The new LNA token launched, 00:25:10.039 --> 00:25:12.559 spiked for a day as people sold the airdrop and 00:25:12.559 --> 00:25:15.259 then crashed again. Trust is a very hard thing 00:25:15.259 --> 00:25:17.720 to rebuild when you've just incinerated the entire 00:25:17.720 --> 00:25:20.359 GDP of a small country. But while the investors 00:25:20.359 --> 00:25:22.059 were licking their wounds, the founders were 00:25:22.059 --> 00:25:24.750 making exit plans. This brings us to the manhunt, 00:25:24.809 --> 00:25:27.309 because this story turns from a financial disaster 00:25:27.309 --> 00:25:30.490 into a crime thriller very, very quickly. It 00:25:30.490 --> 00:25:33.009 does. And the authorities were looking at both 00:25:33.009 --> 00:25:36.380 founders. Let's start with Daniel Shin. He tried 00:25:36.380 --> 00:25:39.400 to play the I wasn't involved card. He put out 00:25:39.400 --> 00:25:41.420 statements saying he had stepped away from Terraform 00:25:41.420 --> 00:25:44.279 Labs years prior to focus on Chai. But the prosecutors 00:25:44.279 --> 00:25:46.839 in South Korea didn't buy that. Not for a second. 00:25:46.880 --> 00:25:49.299 They found regulatory filings that showed he 00:25:49.299 --> 00:25:51.519 still held significant ownership in Terraform 00:25:51.519 --> 00:25:54.220 Labs right up until the end. And they raided 00:25:54.220 --> 00:25:57.259 the offices of Chai. They alleged that Chai wasn't 00:25:57.259 --> 00:25:59.359 actually using the blockchain for payments in 00:25:59.359 --> 00:26:02.140 the way they claimed. They were just mirroring 00:26:02.140 --> 00:26:04.599 data onto the chain after the fact to create 00:26:04.599 --> 00:26:06.960 the illusion of volume. It was alleged fraud 00:26:06.960 --> 00:26:09.380 on top of the catastrophic failure. He was eventually 00:26:09.380 --> 00:26:11.579 indicted in South Korea. And then there's the 00:26:11.579 --> 00:26:14.480 main character, Do Kwon. The captain of the ship. 00:26:14.980 --> 00:26:18.349 Just days before the crash. April 30th, 2022, 00:26:18.750 --> 00:26:21.829 to be exact, Do Kwon filed paperwork in South 00:26:21.829 --> 00:26:24.490 Korea to dissolve the Korean entity of Terraform 00:26:24.490 --> 00:26:27.089 Labs. Wait, say that again. April 30th. That's 00:26:27.089 --> 00:26:29.549 barely a week before the depegging started. The 00:26:29.549 --> 00:26:32.150 timing is, to put it mildly, incredibly suspicious. 00:26:32.730 --> 00:26:35.529 It strongly suggests he knew a collapse was imminent, 00:26:35.670 --> 00:26:37.829 or at least highly likely. He had already moved 00:26:37.829 --> 00:26:41.309 his family to Singapore. And then, as the investigations 00:26:41.309 --> 00:26:45.609 heated up in late 2022, he went dark. He vanished. 00:26:45.630 --> 00:26:48.210 But not completely silent. He did that bizarre 00:26:48.210 --> 00:26:51.650 interview. The coinage interview in August 2022. 00:26:52.170 --> 00:26:55.559 It was surreal. He's on camera looking relaxed 00:26:55.559 --> 00:26:58.160 in a T -shirt, maybe a little humbled, but still 00:26:58.160 --> 00:27:00.980 with that undercurrent of arrogance. He admitted 00:27:00.980 --> 00:27:02.920 his faith in the project was super irrational, 00:27:03.240 --> 00:27:06.279 but he flatly denied it was a Ponzi scheme. He 00:27:06.279 --> 00:27:08.059 tried to frame himself as a failed inventor, 00:27:08.319 --> 00:27:10.619 not a fraudster. While he was a fugitive. While 00:27:10.619 --> 00:27:13.079 he was an international fugitive. Interpol issued 00:27:13.079 --> 00:27:15.819 a red notice for his arrest. South Korea revoked 00:27:15.819 --> 00:27:18.319 his passport. He was technically an undocumented 00:27:18.319 --> 00:27:20.779 person moving between countries. And he was on 00:27:20.779 --> 00:27:22.980 the run for months. Months. There were rumors. 00:27:23.049 --> 00:27:25.569 He was in Dubai, Russia, Serbia, of all places. 00:27:25.789 --> 00:27:27.849 Turns out he was in Serbia for a while. But in 00:27:27.849 --> 00:27:31.109 March of 2023, the net finally closed in. Podgorica 00:27:31.109 --> 00:27:33.750 Airport, Montenegro. This is straight out of 00:27:33.750 --> 00:27:36.710 a Bourne movie. A private jet is waiting on the 00:27:36.710 --> 00:27:40.390 tarmac, ready to fly to Dubai. Do Kwan and his 00:27:40.390 --> 00:27:43.450 CFO, Han Chang -joon, walk up to the passport 00:27:43.450 --> 00:27:46.170 control desk? And what documents do they hand 00:27:46.170 --> 00:27:49.049 the border control officer? Costa Rican passports. 00:27:49.269 --> 00:27:52.670 Fake ones. High quality forgeries, but fake nonetheless. 00:27:53.130 --> 00:27:56.150 The border guard gets suspicious, runs the numbers, 00:27:56.289 --> 00:27:58.849 and it comes back as fraudulent. They search 00:27:58.849 --> 00:28:00.990 their luggage. They find another set of passports, 00:28:01.049 --> 00:28:03.700 Belgian ones, also fake. And a stash of electronics. 00:28:03.980 --> 00:28:06.740 Why do you need two sets of fake passports from 00:28:06.740 --> 00:28:08.240 different countries if you're just an innocent 00:28:08.240 --> 00:28:10.519 startup founder whose project failed? You don't. 00:28:10.519 --> 00:28:12.700 That was the smoking gun that proved intent to 00:28:12.700 --> 00:28:14.700 flee, not just travel. They were arrested on 00:28:14.700 --> 00:28:17.720 the spot. And that kicked off this massive diplomatic 00:28:17.720 --> 00:28:20.819 tug of war. The U .S. wanted to extradite him 00:28:20.819 --> 00:28:23.299 for securities fraud. South Korea wanted him 00:28:23.299 --> 00:28:26.059 for financial crimes. Montenegro had him for 00:28:26.059 --> 00:28:27.480 the fake documents. She's got a good amount of 00:28:27.480 --> 00:28:29.519 time in a Montenegrin prison while they fought 00:28:29.519 --> 00:28:31.640 over who got to prosecute him first. Which I 00:28:31.640 --> 00:28:33.680 hear is not quite the same level of luxury as 00:28:33.680 --> 00:28:35.579 the Terra Club lounge at the baseball stadium. 00:28:36.000 --> 00:28:39.000 Definitely not. And finally, looking back from 00:28:39.000 --> 00:28:42.559 our perch here in 2026, we have the resolution 00:28:42.559 --> 00:28:46.059 to that legal battle. We do. Terraform Labs itself 00:28:46.059 --> 00:28:49.240 filed for Chapter 11 bankruptcy in January 2024, 00:28:49.740 --> 00:28:52.220 but the real hammer fell on Duquan personally. 00:28:52.420 --> 00:28:55.519 He was eventually extradited to the U .S. and 00:28:55.519 --> 00:28:57.960 the trial. Well, it never got to a full trial. 00:28:58.119 --> 00:29:02.539 He took a plea deal. In August 2025, Duquan pleaded 00:29:02.539 --> 00:29:05.210 guilty to two counts of fraud. And the sentence. 00:29:05.410 --> 00:29:08.230 On December 11, 2025, he was sentenced to 15 00:29:08.230 --> 00:29:11.130 years in federal prison. He also had to forfeit 00:29:11.130 --> 00:29:14.250 $19 million of personal assets. And the company 00:29:14.250 --> 00:29:16.769 had billions in civil judgments against it that, 00:29:16.789 --> 00:29:19.269 of course, it could never pay. 15 years. It's 00:29:19.269 --> 00:29:22.549 a long time, but for $45 billion in damage. I'm 00:29:22.549 --> 00:29:24.130 sure there are victims who feel that's light. 00:29:24.289 --> 00:29:26.470 The U .S. prosecutors argued for more, but the 00:29:26.470 --> 00:29:28.950 judge weighed the plea deal. Still, it sends 00:29:28.950 --> 00:29:30.970 a clear message. You can't just code your way 00:29:30.970 --> 00:29:33.529 out of securities fraud. Eventually, the law 00:29:33.529 --> 00:29:35.920 catches up. We can't wrap this up without talking 00:29:35.920 --> 00:29:38.000 about the enablers, because Duquan didn't do 00:29:38.000 --> 00:29:40.900 this alone. He had help, whether witting or unwitting, 00:29:41.000 --> 00:29:43.079 from some of the biggest players in the industry. 00:29:43.339 --> 00:29:45.500 This is the part of the story that makes the 00:29:45.500 --> 00:29:48.700 big institutional players squirm. Let's talk 00:29:48.700 --> 00:29:52.960 about Jump Trading. Jump is a massive, secretive, 00:29:52.960 --> 00:29:54.980 high -frequency trading firm. They're like the 00:29:54.980 --> 00:29:57.940 plumbing of both the stock market and the crypto 00:29:57.940 --> 00:30:01.700 market. In October 2024, the SEC reached a settlement 00:30:01.700 --> 00:30:06.440 with Jump Trading for $123 million. And the details 00:30:06.440 --> 00:30:08.339 that came out of that settlement were damning 00:30:08.339 --> 00:30:10.759 for the Terra narrative. What did they do? The 00:30:10.759 --> 00:30:14.819 SEC alleged that way back in May of 2021, a full 00:30:14.819 --> 00:30:18.019 year before the final collapse UST had de -pegged 00:30:18.019 --> 00:30:20.500 before, it dropped to something like 94 cents. 00:30:20.779 --> 00:30:23.559 The algorithm failed then. Wait, so the death 00:30:23.559 --> 00:30:25.779 spiral almost happened a year earlier? It almost 00:30:25.779 --> 00:30:28.210 did. But nobody knew because according to the 00:30:28.210 --> 00:30:31.049 SEC, Jump Trading secretly stepped in. They made 00:30:31.049 --> 00:30:33.609 a deal with Do Kwon. Jump would buy massive amounts 00:30:33.609 --> 00:30:35.990 of UST on the open market to prop up the price. 00:30:36.170 --> 00:30:39.170 They artificially restored the peg. So they painted 00:30:39.170 --> 00:30:41.130 the tape. They created the illusion that the 00:30:41.130 --> 00:30:43.069 algorithm worked and had self -corrected when 00:30:43.069 --> 00:30:45.569 in reality it was just a secret centralized bailout. 00:30:45.839 --> 00:30:48.700 Exactly. And in exchange, Terraform Labs gave 00:30:48.700 --> 00:30:51.740 Jump's sweetheart deals on LNA tokens at a huge 00:30:51.740 --> 00:30:54.099 discount, allowing them to make over a billion 00:30:54.099 --> 00:30:57.019 dollars. It was a symbiotic relationship built 00:30:57.019 --> 00:31:00.119 on deception. If investors had known in 2021 00:31:00.119 --> 00:31:02.859 that the algorithm had already failed once, do 00:31:02.859 --> 00:31:04.740 you think they would have poured billions more 00:31:04.740 --> 00:31:07.880 into Anchor in 2022? No way. That's a material 00:31:07.880 --> 00:31:09.990 fact that was hidden from the market. And what 00:31:09.990 --> 00:31:12.490 about Binance? The 800 -pound gorilla of crypto 00:31:12.490 --> 00:31:15.329 exchanges. They were an early investor in Terraform 00:31:15.329 --> 00:31:18.609 Labs. They listed UST and LNA. They promoted 00:31:18.609 --> 00:31:21.150 Anchor Protocol's 20 % yield on our platform, 00:31:21.390 --> 00:31:23.569 calling it safe and stable. Which it wasn't. 00:31:23.650 --> 00:31:26.049 Which it was not. There have been numerous class 00:31:26.049 --> 00:31:28.509 action lawsuits in the U .S. and formal complaints 00:31:28.509 --> 00:31:30.609 in countries like France accusing Binance of 00:31:30.609 --> 00:31:33.589 misleading advertising. It reinforces the idea 00:31:33.589 --> 00:31:35.509 that the exchanges weren't neutral platforms. 00:31:35.829 --> 00:31:38.450 They were active participants profiting from 00:31:38.450 --> 00:31:40.819 the trading volume. even as the ship was clearly 00:31:40.819 --> 00:31:43.519 taking on water. So we have the crash, the manhunt, 00:31:43.579 --> 00:31:46.039 the prison sentence, and the regulatory crackdown. 00:31:46.240 --> 00:31:48.759 I remember South Korea revived its special financial 00:31:48.759 --> 00:31:51.920 crimes unit, the Grim Reaper of Uido, just for 00:31:51.920 --> 00:31:55.319 this case. That name is so metal. But yes, the 00:31:55.319 --> 00:31:57.960 regulatory landscape we live in today in 2026 00:31:57.960 --> 00:32:01.440 is strictly defined by the Terra collapse. Algorithmic 00:32:01.440 --> 00:32:03.880 stablecoins are effectively banned or regulated 00:32:03.880 --> 00:32:06.200 out of existence in most major jurisdictions 00:32:06.200 --> 00:32:08.619 now. So let's land this plane. We've been through 00:32:08.619 --> 00:32:10.559 the mechanics, the madness and the aftermath. 00:32:10.799 --> 00:32:13.799 What is the takeaway? Why does this story matter 00:32:13.799 --> 00:32:16.920 to the listener who might just have a 401k and 00:32:16.920 --> 00:32:18.910 has never touched crypto? I think there are two 00:32:18.910 --> 00:32:22.150 massive enduring lessons here. The first is what 00:32:22.150 --> 00:32:24.529 I call the innovation trap. Explain that. We 00:32:24.529 --> 00:32:26.630 live in a world, especially in tech and finance, 00:32:26.869 --> 00:32:29.990 that fetishizes complexity. If someone explains 00:32:29.990 --> 00:32:32.210 an investment to you and uses a bunch of jargon 00:32:32.210 --> 00:32:35.109 words like burn and mint equilibrium or algorithmic 00:32:35.109 --> 00:32:38.349 peg or synthetic derivatives flywheel, your brain 00:32:38.349 --> 00:32:40.569 tends to shut off and you assume, wow, they must 00:32:40.569 --> 00:32:42.430 be smarter than me. This must be legitimate. 00:32:42.769 --> 00:32:45.750 It uses complexity to mask a lack of fundamental 00:32:45.750 --> 00:32:48.779 value. Exactly. Innovation is great, but financial 00:32:48.779 --> 00:32:52.720 gravity is a real undefeated force. If an investment 00:32:52.720 --> 00:32:55.880 claims to be stable but isn't backed by anything 00:32:55.880 --> 00:32:58.119 other than faith in its own future growth and 00:32:58.119 --> 00:33:00.539 a complex algorithm, it's not an investment. 00:33:00.720 --> 00:33:02.839 It's a confidence game. And the second lesson? 00:33:03.140 --> 00:33:05.759 The yield warning. This is the one I want everyone 00:33:05.759 --> 00:33:08.400 listening to burn into their brain. 20%. The 00:33:08.400 --> 00:33:11.359 golden number. If someone, anyone, anywhere offers 00:33:11.359 --> 00:33:14.400 you 20 % interest when the risk -free rate, what 00:33:14.400 --> 00:33:16.319 you can get from the government, is 1 % or 2 00:33:16.319 --> 00:33:19.240 % or even 5%, you need to run in the other direction. 00:33:19.559 --> 00:33:22.779 There is no such thing as a risk -free 20 % return. 00:33:22.920 --> 00:33:25.559 It does not exist. The risk is just hidden. And 00:33:25.559 --> 00:33:26.880 if you don't know where the yield is coming from? 00:33:27.140 --> 00:33:29.700 You are the yield. Your new investment money 00:33:29.700 --> 00:33:31.579 is being used to pay the interest to the people 00:33:31.579 --> 00:33:34.819 who got in before you. You are the yield. That's 00:33:34.819 --> 00:33:37.119 a chilling thought. It's the simple truth. The 00:33:37.119 --> 00:33:39.039 people who made money on Terra were the early 00:33:39.039 --> 00:33:40.960 investors and the ones who got out before the 00:33:40.960 --> 00:33:43.559 music stopped. The people who provided the exit 00:33:43.559 --> 00:33:45.559 liquidity were the regular folks who came in 00:33:45.559 --> 00:33:48.359 at the end chasing that impossible 20 % promise. 00:33:48.660 --> 00:33:50.839 It's a hard lesson, but at a very, very high 00:33:50.839 --> 00:33:56.240 price. A $45 billion price and 15 years of Duquan's 00:33:56.240 --> 00:33:58.640 life. So here's the final thought I want to leave 00:33:58.640 --> 00:34:02.099 you with. With Duquan in prison, with the enablers 00:34:02.099 --> 00:34:04.400 having paid their fines, and with Terra Classic 00:34:04.400 --> 00:34:07.730 still... Bizarrely, trading is a kind of zombie 00:34:07.730 --> 00:34:11.329 meme coin. Have we actually fixed the structural 00:34:11.329 --> 00:34:13.530 issues in finance that allowed this to happen? 00:34:13.929 --> 00:34:15.949 Or are we just waiting for the next charismatic 00:34:15.949 --> 00:34:18.389 founder with a complex revolutionary algorithm 00:34:18.389 --> 00:34:21.030 to repeat the whole cycle all over again? That 00:34:21.030 --> 00:34:23.309 is the question, isn't it? It is. Thank you for 00:34:23.309 --> 00:34:25.190 taking us through the math and the madness of 00:34:25.190 --> 00:34:27.170 this story. It was a pleasure. And to you, the 00:34:27.170 --> 00:34:30.590 listener, stay curious, but stay skeptical, especially 00:34:30.590 --> 00:34:32.989 if someone offers you 20 % interest. We'll see 00:34:32.989 --> 00:34:33.969 you on the next Deep Dive.