Should DeFi Platforms Block 'Insane' Trades? - Uneasy Money

Transcript

So there was a uh per issue um an Oracle issue or articles and per um one of my favorite topics uh last week where um SKHEX uh per on trade XYZ fell from uh 1128 to 917 $917 um and then uh almost uh immediately recovered. Um, so, uh, this was like some thin pre-market print, um, that happened. It was a real trade. Um, and then, uh, a bunch of people got wrecked.

Um, which again, like oracles. Um, uh, actually, um, we have a special guest, another special guest, uh, who's going to just join us, uh, Mike from, uh, Ether. So, um, let's pull him up on stage. We're going to we're going to talk Ethereum later. Uh, so, um, so I think let's let's get this, uh, get this rolling. So, um, welcome. Um, thanks for thanks for joining us. Um, so yeah, trade.xyz. Um, I think the the, uh, the maybe, um, most interesting part about this is that they were like, uh, actually, sorry, we're going to make you

guys whole. Um, and then they were like, well, but just this one time, like never again. This is not a precedent. Like, don't come crying to us if this happens again, which it absolutely will. Um, so uh yeah, I don't know. Uh takes on on uh this um this Oracle problem. >> No. >> Oh, sorry. From uh from me, >> from anyone. Uh you're on a talk show, sir. So uh whatever whatever uh very weak uh opinions you have, throw them out. That's what we want. So uh >> Oh, man. Are you a trade XYZ guy? Uh >> no, I not I'm I've somehow missed the

the boat on Hyperlid. Um right. >> So uh so no, unfortunately I just I haven't been paying attention. I didn't know that it had uh oracle issues or anything else going on there. I'm mostly focused on this Ethereum EIP that's been making >> we're gonna get to that. We're gonna get we're gonna get to the Ethereum EIP. We're very excited to talk about that as well. Um but yeah, so so um I think you know the other thing that was quite interesting uh that um this was a Kobe post that uh that came out around about the same time as this um where he's like uh trade.xyz XYZ is not raising money, which is always like a weird like whenever Kobe tweets something randomly like you know that there's some like

chaos going on in the background that uh like some some crazy [ __ ] going on. So, so basically uh the the setup here is you have Hyperlid and then you got trade.xyz um and trade.xyz XYZ is a call it ecosystem participant that offers RWA per right um and you know RWA per maybe like a little in in the gray if if that still exists in this day and age um in terms of like what you can get away with doing. Um, so, uh, so you know, Hyperlquid has like core crypto assets and then they've got all of these like

other ecosystem partners that offer like different, um, different other, uh, asset classes and and, uh, trade types, you know, options, etc. Um but again, you know, the the challenge with RWAS that you don't have uh to the same extent with like Bitcoin is someone has to tell you what the price of this thing is. >> And RWA's stop trading at random times for random reasons. Like when synthetics was building this like this took us like a year of dealing with oracle providers like chain link and and a bunch of people chain links people that they were talking to had no they're just not you they're like when the market is shut

down the market is shut down like why would we care about the like outside of market times right um but pre-market closed markets um it's it's a really challenging thing uh to manage and and so like a real trade it happened in pre-market that someone dumped all of their SKH highex uh which maybe was a good trade actually in hindsight um and uh and you know that's the price right like and and and so there's you know even the fact that trade XYZ was like okay we this probably shouldn't have happened for reasons whatever even though it was a real trade is like itself a pretty opinionated take on

asset pricing Right. And you know uh whenever I mean this like you know to Mike like what is the price of ETH versus Rapy? Is it the same thing? Like what about when it dislocates for some weird reason like is it still that price? Like um you know or like LST LRTs, right? Like um you know this is this is a thing that you must lose sleep over, right? Like you know not all flavors should be We're uh somewhat of a preview. We're going to uh make trading a lot better in uh in Etherfy over the next uh couple weeks. Uh and it is something actually

right now, you know, we're beta testing and um uh some of this functionality. And yeah, we we actually had the exact same situation where a couple of trades went through where, you know, not massive sums of money, but like users lost a bunch of money because uh they were allowed to do something that I think um uh reasonably reasonably uh we should have not allowed them to do. Now, if somebody goes to unis swap, right, goes to the UI uh picks some, you know, low liquidity token, you know, this happened not that long ago, right, with a someone went to the UI >> thing. Yeah. Yeah. I want to send 50 million each for 500 bucks, >> right? It's just madness. Now, on the one hand, you could say, "Look, sorry, bro. This is, you know, permissionless smart contract like you you you pressed

the proof that would you saw in the UI." >> Um, but like I think if you're hosting the UI, uh, and you're, you know, um, uh, I don't want to say facilitating the trade because, you know, we're not facilitating the trade, but you're you're playing some role. Yeah, you're involved. You're playing some role, right? >> Betrayed, right? >> Yeah. >> I think ethically, and and I do mean ethically, like you have a responsibility. I mean, you have a responsibility to prevent the user from doing something that's just like wildly insane. I think that's how brokerages operate. Again, not that, you know, we're we're a brokerage, but if I go to Interactive Brokers, uh, and I do try to do some insane thing, it does, you know,

displays or stop me. And I think I think unis swap should do that. I think uh u you know we're planning to do that as we roll some of this stuff out. Um and and I think trade XYZ should have should have done that. Like this isn't you know this isn't rocket science, right? It's actually it's not that hard to know, hey, somebody tries to do a trade with 70% price impact or some crazy slippage. You probably should just shouldn't let the user do that. Forget warnings and red text. Like you shouldn't let them do that. Um >> uh what if they really need this is you know this is the the thing right like what if that guy really needs >> and they can do some go to ether scan go to ether go to ether scan yeah

>> like contracts right >> yeah this feels like the same conversation that we've had before with >> we have it all we have it like every few weeks >> oh okay so it is it's not this is this is literally the same >> this is like what is there's like a philosophical epistem logical question of like what is the price of something, right? Yeah. Like there is no price. The price of SKH Highix was like was it the closing price like 12 hours before? Is that the real price or is that like what some idiot on a Saturday wants to sell because they're like getting liquidated on something else and they need to, you know like >> they need to.

>> There is no such thing as the price of an asset, right? It is like a a some kind of a construct. So you need to have a set of rules that says and you know the the like if someone is willing to sell enough SKH highix at $1, like is that the price for that fiveminute period where they're like just dumping into the market, right? Um and and like these are these are like non-trivial problems. You can't solve it because markets just do weird [ __ ] right? Um and so you know uh I think trade.xyz XYZ like part of the UX is like if something goes wrong as the venue and you know NASDAQ does this right like we were supposed to have uh you know uh like

stop triggered on this and it didn't or it happened three times and we're just going to roll back the trades or like this thing happened someone fat fingered a thing it's not the real price we decide what the real price is the the challenge for for DeFi is like you're not even supposed to be involved in this right like it's supposed to be contracts and stuff and all of a sudden you're like making post hawk judgments about what the real price was. Um it's okay for the NASDAQ to do it. I don't know about uh about like trade.xyz that's >> well could say like if you're operating a wallet right like whether it's let's say meta mask or phantom and a user tries to interact with a known malicious contract like it's the same kind of argument. You could say, well, listen,

it's permissionless, man. Like, just the user should be able to do whatever they want. But the fact is, look, you you're operating an app. You have a list of malicious contracts. Uh, at minimum, there should be a multi-stage thing that tells you if you do this, you're just absolutely going to get destroyed. And then in the end, I I think it's a discretionary call to say, "No, we're just not going to let you do this. No matter what, someone steals your phone, tries to do something, we're just not going to let you do that." And I think either decision is right but uh sort of throwing one's hands up in the air and saying like hey listen it's it's not our responsibility like look just objectively to some extent it is um you're running the site responsibility.

>> Exactly. Yeah. I mean, you know, let's not let's let's be realistic though. Like part of the reason that people were like, "Eh, I'm I have no responsibility or like I can't do things or whatever." Like we talked about this a lot today is like, you know, if you said yes, I'm going to stop people from doing something. There was a guy called Gary Gendler who would show up at your house and blow your brains out. And so everyone >> maybe some Bitcoin maxis who are gonna like >> yeah there are like two two groups of people that would come philosophers are going to come and philosophers who come and get you. Um but yeah so all right let's let's uh let's go >> I was going to add one thing um this

actually reminds me of when when I was at stakehouse we implemented this thing called metaoracle um Mike you might be familiar with this so I remember um >> you know in lending protocols when you integrate an asset um there's a question of do you use the fair market value um which can you know whiplash around like we've seen here or do you use the primary redemption value um or do you use both and try and figure out when there are deviations? So, it's it's a real um I mean, you know, in different lending protocols, stable coins, stable coins have been hardcoded at $1. Is that right? you know there's so much

different opinions around this and um I think it's quite prudent to have you know uh like a metaoracle like the one that stcast implemented on uh on Morpho um where you actually get notified and there's a deviation when there's great enough deviation then um you can kind of switch off between the two. So um >> the the challenge is right with articles everything you do creates like downstream consequences right like you make that design decision right and it's like oh that's got to you know okay we don't want people to get wrecked in this case which is like let's call it you know the like two sigma case right but then you switch that thing on and then in the three sigma case everyone dies

right like there's just you know like like it's like ah okay I see why that was a mistake now right like It's it's really really hard to [ __ ] with oracles and not have layers of consequences that are like built into all those decisions that like as you get further out into the tail >> of like you know uh possibilities like increasingly bad things happen, right? Like you know we we used to talk about this all the time of like what happens if everyone tries to get out and the you know exit queue meanwhile it was actually the problem is like the entry queue was the was going to be the like limiting factor right

you're like ah like people are going to want to get out of staking right now we're we will talk about this in a second right now we're like actually what if we stop this entry queue this entry queue is actually like really annoying there's too many people trying to get in right um and so often times times the thing that you're trying to prevent ends up not actually being the real thing. It's some other thing that you didn't think about and then the consequences of that are like significantly worse than the thing that you were worried about and trying to prevent. So, um yeah, mechanism design is hard, I guess, right?

>> If you like this clip, tune in to Uneasy Money every Wednesday at 400 p.m. Eastern. If you hold crypto on your phone, your biggest vulnerability isn't your wallet, it's your carrier. AT&T, Verizon, and T-Mobile have been breached again and again, and SIM swaps are still one of the easiest ways for attackers to drain accounts. That's where Cape comes in. America's privacy first mobile carrier. Same premium service, but Cape rotates the identifier on your SIM every 24 hours, deletes your call and text metadata after a day, and protects against SIM swaps with a 24word recovery

phrase that only you control. You also get two middle toend encrypted secondary numbers for banking and signups, so you stop handing your real number to every app that asks. Go to cape.co/chained co/chained and use code unchained for 33% off your first six months.