How To Maximize Gains With A.I. Trading Perps 💰 Liquid CEO INTERVIEW Transcript, AI Summary & Key Points

00:00 How is AI trading the market today? We're going to be breaking this down, taking a look at how crypto is impacted by what's going on in AI and it's going to be a good one. I do want to thank our sponsor today and that is Tandem where you guys can get into self-custody. It is the last day you can take advantage of getting a wallet and the next one 50% off.

00:19 So, do that. If you're not in self-custody or if even if you are in self custody, you want that extra wallet. Check out tandem.com. Use our link. It's going to get you moving and get you a little bit more of a discount. So, it do it does stack. So, make sure and take advantage of that and let's get going. Uh, listen, we've had Andrew Wang on or Franklin Wang on before, the founder of Liquid.

00:39 And we wanted to bring him back on because I think the the point today that we wanted to get to is really understanding what is happening around the idea of AI as it pertains to crypto. So, I want to welcome Franklin Wang back into the show. How are you, Franklin? >> Hey, listen. I want to first of all, we've had you on before and you guys will probably remember this video as what is that two months ago.

01:02 >> Uh and it was really talking about AI integration a little bit about where Liquid is. >> Can you tell me a little bit about Liquid and its growth so far? >> Absolutely. So when we did that video, I think we had about 1,500 Co- Invest users. I don't know the exact date. We were growing really, really quickly back then. Y >> uh we are now at 10,000 Co- Invest users.

01:24 >> Wow. Okay. Um, so we've grown 7x since then. >> For some of our audience, helping them understand Liquid for the first time. They've never used it. Give them kind of a rundown of really where your benefits are, some of the features that really make you guys special. >> Absolutely. So, Liquid's goal is to build the best possible trading interface for everyone.

01:42 First, we started with what we think is the best instrument, which is perpetual futures. I think at the time uh these were starting to get warmer and warmer as like okay historically these haven't really been legal in the US but they're clearly trending towards the direction of legality. So we think that perpetual futures are like the best possible instrument.

02:01 They let you get leverage they run 24/7 and so the largest part of Liquid's mission today is thus in bringing perpetual futures to every interface imaginable. So you can think of mobile apps that was our first flagship product. uh we brought them inside of Claud and chatgpt. Uh since we made that video, we've actually kept expanding. So you can now even interact with co-invest via text message and you can also interact via a Chrome extension that will sort of show liquid assets in various things like X and Reddit and

02:31 let you buy directly from there. >> So when you look at some of the big wins, especially when it comes to AI calls, you know, since the launch, has there been anything that sticks out for you guys? >> Absolutely. Uh there's one yesterday that really really sticks out. So I'm going to go talk about that. Uh one of our users uh it's JC Merllo used co-invest connected it to his system and made 20% returns yesterday trading crypto.

02:56 He chose some very obscure crypto names. Like one of them was Seeker and I didn't recognize the other ones. He chose one crypto that went up like a 100% and it was all thanks to him. It was all thanks to co-invest like it wasn't a crypto that I recognized. Um, so he was able to do this and what's really interesting is that he was doing this in more of an agentic way.

03:15 So back then when he showed co-invest, right, you could basically ask it like what should I buy and maybe let you buy one thing or like four things or whatever, >> right? >> But he runs co-invest with a routine. There are now these things called scheduled routines inside of Claude which let you run one operation like every hour for example. So he has his like routine and he runs it every single hour and when that happens it makes the trades.

03:39 So he doesn't need to like importantly he doesn't need to like look at anything. He just types in his routine and it just keeps going right and I think this really highlights something which is that if you have a process like locked and loaded in the past you would have to apply this process every single day right but now with AI you can just give it the process and tell it to run it every single hour and it'll keep running for the end until the end of time which I think is really really cool.

04:06 Are you at all concerned about AI and the use of wallets andor being able to really get exposure to your portfolio in general at all for in terms of safety? >> Obviously, there have been a lot of security issues recently in the crypto space. Many of them are caused by AI because a lot of these methods people are using are very very likely to create like issues.

04:27 Like one of the ways we can prevent it is actually by using relatively low tech approaches. So if someone hacked your entire claude and had access to your co-invest, they wouldn't be able to withdraw money. I think that's like a very very important criteria. Someone hacks your co-invest, they can't withdraw money. The worst thing they can do is maybe they can place some trades and even that's like a little complicated.

04:49 >> Okay. >> So co-invest itself has very narrowly scoped permissions. It cannot withdraw your money. It can only trade. So once that's the case, there's very limited things that uh a hacker could do even if they could access your co-invest. And obviously we have a lot of things in place to make sure they can't access it. So yeah, >> I'm looking at Coinbase uh their announcement in terms of their own AI, Coinbase for agents.

05:11 Um rebalance my portfol. I'm just kind of curious why wouldn't someone use Coinbase AI to trade versus something like a co-invest? >> Well, there's a few reasons. So Coinbase is custodial, right? If Coinbase gets hacked, you're an unsecured creditor. If Liquid goes under, you can still access your funds. Secondly, Coinbase's fees are like exorbitantly high.

05:32 Coinbase's fees are like 10x those of liquids. I think Coinbase's liquidity is much lower than liquids as well. Um, if you want to make big trades, I don't think you can really make them on Coinbase. >> Okay, the liquidity isn't. How do I feel about all these other platforms like Coinbase and Robin Hood launching these agentic products, right? And I think it actually validates that this is clearly something people really, really want, right?

05:54 This is absolutely what we see. This is what we think we see there. We've done very little marketing for Co- Invest and the usage is very very high. People are just kind of catching up. I actually don't know if this Coinbase product is like in beta or whatever, but we are kind of already thinking about the next step. We think these guys are sort of actually copying us.

06:10 I don't think it's any coincidence that Robin Hood Agentic came out the day after Liquid co-invested and it took them like two weeks to actually launch it. >> I think we'll we'll always be devoted to providing the best possible experience. Um, I think liquid security is just much superior to Coinbases in this front. >> Wow. Okay. All right. Let's talk about meme meme coins in general.

06:31 I'm looking here on Hood Chain, and this kind of surprised me even. 431 of the top 500 tokens right now on Robin Hood chain, >> which actually uh dwarfs >> Salana. Now, this is in reference to the 500 tokens. That's not all crypto because Salana would be huge in comparison. Yeah. But when you look at memecoins in general, when will Co-Invest be smart enough to actually trade meme coins, do you think?

06:57 Is that something that you guys are working on or is it something already in in effect? >> We have built some products in this category. Suffice to say, I do not think it would be technically very challenging to write these things, but I generally think that if you ask Co- Invest to buy a memecoin for you, it usually is like pretty hesitant. I think that memecoins are kind of kind of a net negative to society.

07:19 It's not really participating in like capital formation. It's all like some attention game. Like if you really tell co-invest to do so like co-invest is about helping people make smarter decisions with their money. I think buying memecoins is not a very smart financial decision. >> So in that sense we don't let co-invest buy meme coins. It would not be technically hard to let co-invest buy memecoins.

07:37 It'd be very very easy. But the choice you guys are making in reference choice. >> Okay. So just restricting the access to memecoin at least from an AI standpoint. AI would probably go crazy on it. What What about this? I want to kind of get into a couple of points here. Do you think AI right now is calling the crypto bottom? >> Well, it's up a lot in the last uh last few weeks, right?

08:00 So, uh is the recent market low a bottom? It actually doesn't say that the bottom is in. Uh it's a little >> Yeah, it says >> it's not yet. It's not at a bottom. >> All right. Not confirmed. So, if you're trusting that, then maybe that's something to consider. Guys, what about how would AI trade the Clarity Act right now? Passing or failing? Do you think it would would fail it or pass?

08:21 I've actually asked that question of two different AI uh functions that we use and I got two different answers. >> So, are you asking how co-invest would trade it? I think >> in general, >> how would AI trade clarity >> in terms of the act passing or failing? Do you think they would trade it to fail or trade it, you know, trading on the side that it would fail?

08:44 Trading on the side that it would pass? >> We can ask it this question. I admit I haven't asked it this problem. I think it would probably I think my guess is that AI can really only draw upon what it can see, right? >> So, it might look at prediction market data. It might try to see are there on one side of the market that can often tell uh what's going on, right?

09:04 >> I'm I'm looking for the AI that understands that. If the AI truly understands the flow of sentiment in crypto, that to me, I think is is how you would really start to get ahead of some of these uh some of these. >> If you ask Coin Invest right now what it thinks of the Clarity Act. >> Yeah, let's do it. >> So, yeah, it's looking at the prediction markets in particular.

09:27 Um it's also looking at see what Congress is saying. It's looking at news. It's looking at all these sources. It says that it gives a very little low chat. says it probably won't pass. >> Yeah. So that that's poly market speaking, you know, we'll just show the the clarity act right there. So when you look at this, I mean many people would say, "All right, oh, I'm going to you could if you're going poly market, you're going to trade against it."

09:49 So that could be a variety of tokens. Would co-invest be able to select tokens on the basis that clarity does not pass? So the assumption is clarity does not pass based on the data we have right now. >> Yeah. What tokens would Co- Invest say, "All right, this is the position we should take." >> Let's let's just ask it. >> Well, it is saying that if it passes, crypto is likely to like go higher.

10:17 >> Well, sure. Yeah. >> Yeah, that seems I don't know if you need AI to tell you that. Okay. Seems like something >> It seems to be generally I think bearish on it generally thinks that it's already priced in and if it passes it would go up, but >> Okay. All right. So, let's move on from here. So, a couple of things I wanted to kind of go off of there.

10:33 Uh, you know, we talked a little bit about memecoins, what the possibility here is, you know, on the clarity side. Uh, when you look at something that's going on with the CFTC, because this is something that >> may start to shift how investors work here in the US, especially if the CFTC brings hyperl legally to the United States. I want to play a clip for you.

10:57 Actually, this is a clip that's talking about the response during that last meeting because this makes a lot of sense for you guys if you're if you're looking at how to understand what's going to be happening in the markets. Let me play this clip for you. I would have to ask the commission what they are doing to police everybody's VPN where they are not accessing these markets.

11:19 That's horrible for our industry. We're not a bunch of carnival barkers at a circus. >> This occurred offshore and that's fake news. No, I'm just asking a simple yes no question. Have you got more people in my regulatory department than you do in your entire company? Okay. Thank you. >> Maybe maybe you should learn a bit about efficiency then. Um >> well maybe you should learn about credible markets.

11:39 >> Okay. >> Um >> well let's let's try to keep this productive. Okay. >> How do you resolve disagreements amongst a very you know vast array of different players in the industry who obviously have different agendas and different interests? >> Well, you don't want to resolve them. You want those things to be aired out. I mean, I think it's actually productive to hear where the dividing lines are.

11:58 I think you want to understand where the battle lines are. It got a little spicy for a second, but that's okay. You know, that's what makes these meetings interesting and that helps us to distinguish, you know, what we have to work on going forward. >> All right. So, what we saw there was the uh the clarity or excuse me, the um uh the meeting last week.

12:18 it was with the CFTC that was chair Celig and Duffy uh kind of going at it along with Hyperlid there and do you think Duff like for instance Terry Duffy right now he's wanting to ban anything that has 5x leverage why do you think that they're concerned about that >> of course these the CME like the CME doesn't offer more than 5x leverage >> so of course he's worried about other people offering more than 5x leverage >> so just simple market preservation this is the angle they're looking at right No, cuz he doesn't want

12:50 it cuz like hyperlquid, right, is technologically like much more advanced than the CME. It's 247. It offers a like high amounts of leverage. It offers automatic liquidation. So that like that's why it lets people have more leverage. It's much more advanced I think in many ways than the CME. So I think to that extent yeah >> if if Celig is successful onshoring hyperlquid >> first of all how would that function and what do you think would be the impact on markets if if hyperlquid were to become a real thing here in the

13:21 US >> I think if hyperlquid is legalized uh you you've probably seen the CME stock price react negatively to hyperlquid being legalized I think that's very bad for the CME because right now many US funds want to trade on Hyperlid but can't Right. >> Yeah. >> And so once it's legalized, you can imagine they'll want to trade on both Hyperlquid and CME.

13:40 And the CME will finally face a challenge to its historical monopoly. >> Yeah, for sure. When you look at um what's going on with PERS in general, Hyperlid obviously kind of playing this a huge role in this market. Uh Liquid out there now also playing a huge role. When you look at something like Anthropic, this is a good example right here. Do you think that we will see liquid and others start to really put a dent in the potential for what is coming and that is an AI IPO season?

14:10 We're going to probably start with Anthropic. OpenAI is coming as well. There will be others. Do you think this is going to have any kind of impact on the on those markets? >> Absolutely. Right. Like right now the bankers and Anthropic are negotiating what Anthropic's price should be. Right. And having a market like this gives both sides something to point to.

14:29 So like if anyone wants to point to a price that's radically different from these uh they either side can point to this right like if anthropic is like we want 2.5 the bankers can say this place says you're worth 1.95 and conversely if the bankers want 1.5 and anthropic can point to this and say no it's 1.95 >> is there enough data sample there to really do price discovery out there prior to an IPO offering do you think >> I think that it's probably not terribly wrong right like these markets markets are generally

15:00 wellknown enough that like most people who have information can kind of contribute it right the markets are not huge right now like I expect that a real anthropic market will have 1,000x the current volumes of enthropic preipo but there is no enthropic preipo market on like official rails right this is the only thing we have >> so I think that historically if you've looked at the preipo histories of companies like liquid and hyperlquid we've basically gotten it right within 1% every time as to where the IPO will

15:31 initially trade at, right? Or at least like the opening print, not the exact IPO price. >> Yeah, we'd have to give credit to that, you know, to that even if you just looked at SpaceX because it was accurate, you know, to that point. So that's some history on a major IPO, which Anthropic is most likely going to have a very similar uh scale uh to kind of go in that.

15:51 So, okay. So, how fast do you think, Franklin, before we would see everyday investors, now I'm not talking about somebody that maybe is active in crypto, but a traditional investor that's been active in Tradfi for a while. How long do you think it's going to be before they start moving to liquid, hyperlid, etc., and truly making trades there? >> I think it will happen very fast.

16:14 Uh, we've seen very large growth. I think last time we talked, Liquid was probably at like I want to say 35,000 users. Today we're well over 60,000. >> Yeah. >> And so just in those two months, we've gained like tens of thousands of users. All of our metrics suggest that we are going exponential. We're finding better ways to tap traditional investors.

16:32 And generally when they hear they can get preipo exposure to anthropic, they're very excited about that. So >> um I think they all clearly want exposure. Yeah. >> What do you think uh in terms of hype? Do you think they should be or that platform should be worried about market share? right now because this is seems to be a little bit of a power grab.

16:51 >> In what sense? Like yeah, can you be more explicit? >> Well, in the sense of market share when when you look at per because PERPS are launching across a variety of platforms. We're going to see this everywhere >> and Hyperliquid has kind of been the only game in town. Do you think they should worry about market share here in the near future? >> Yeah, absolutely.

17:09 I think this was one of the big pieces of the thesis for Liquid. It's that we will have we have a lot of we aggregate many different perexes and per sexes right we we believe that perpetual futures are like the right solution. We think every single perpetual exchange has its own advantages and disadvantages, right? Some might have different fees, different liquidity, different instruments.

17:29 And our goal is to put all of them together and kind of let the user choose, right? Yeah. >> Products like lighter, like we were the first integration partner of Lighter. >> So we believe that more competition that layer is better for consumers. So generally we think Hyperlquid of course should be worried about market share and they should be worried that they have many competitors.

17:48 Now, >> yeah, >> I think it's a big win for consumers. Liquid will always provide the best venues to its consumers. >> Let's talk about the the chips hype that has been going on, you know, for the last really last year versus crypto. Do you think that is going to continue or do you think that particular narrative is starting to die off? >> AI just needs a lot of memory.

18:07 Uh so I think that's one reason why the chips have been rallying. Uh there's definitely like some aspect of like are we in a bubble territory? Um, obviously they've run up a lot. I think that was pretty reasonable. I think that memory was very underpriced for a long time. Right now, I still feel like it's slightly underpriced, but I think it's hard to know for sure.

18:29 And also, it's run up a lot, so you may see a lot of volatility. A lot of people might want to take profits. >> Well, I mean, I I listen, this could get even more exponential, too. I mean, you look at what Nvidia is doing with Hugging Face, the likelihood, I mean, they're going to essentially become a vertical stack. So I wonder the kind of impact that's going to have in general to chips and overall the AI narrative.

18:51 Uh do you feel like investors to today should be diversifying or do you feel like they just need to jump into a high, you know, pos high convicted position and go that route? For instance, I'm all in on digital assets. Maybe someone's all in on chips. Maybe someone's all in on on, you know, anthropic or other AI platforms. Where do you think that plays out?

19:11 I think that there's a few things going on, right? Like I think that if you've kind of been looking at chips, chips have obviously run up a lot. It's a little psychologically hard to bring yourself to kind of invest more into it. It's like what am I doing here? Why am I buying this after it's already run up 900%. >> I think that there's a really enduring there's enduring quality in those businesses.

19:32 The chips are likely to do well. I don't know if I would really advise like I would probably advise you to keep some allocation towards that. I don't think I would say this is go all in on that. Like there are a lot of risks that I think are very real. Like is China going to make the chips at some point? This chip shortage will stop. Maybe the AIS get more efficient.

19:48 They use less compute. That's these are all real possibilities. I can't tell you like I cannot in good faith recommend like hey pivot all of your assets into memory. I think it's pretty strong as a narrative. Like I think it's pretty strong as a business. Like I've heard rumors that uh you know Anthropic believes they'll use even more memory in the future.

20:09 You look at the way Anthropic is signing like data center deals like it's going out of style. Like clearly there's a lot of demand for compute. Uh these companies like Enthropic are able to generate huge amounts of revenue for like one gawatt of power. In that sense every Enthropic is not really even being that price sensitive. Like you saw $45 billion deal for Nscale like obviously $1.25 billion a month at SpaceX.

20:32 huge deals, not really price sensitive, just going all in. So everything in that supply chain is going to be a huge beneficiary of that. >> Do you think we're ever going to have you know compute at I mean because we are starting to see some trends for local compute that obviously will have an impact on certain aspects of the market. Obviously we will see you know machine level I think AI now starting to scale.

20:58 Do you think that's ever going to have a component where where maybe that would potentially slow down that market? >> I think there's a real risk that local AI kind of threatens the market. I do think that cloud agents in particular, like agents you don't run locally are just much better. Like if you use right now, if you use like codecs or cloud, you can run them in like a local agent mode >> where you can run them on your own compute, right?

21:23 >> And I think that's just a very bad interface generally. Like it's it's okay. It's okay. But like having cloud agents, you just tell it to do something, it does it for you, right? Like in a very simple level, if you run it on your own computer and then you turn your computer off, your your computer won't be able to do anything for you, right? You probably don't want to keep your computer on all the time.

21:42 >> Have you guys done any testing with like Grockbot with their essentially they have a cloud agent tool that's similar I think >> uh going that direct. Have you guys done any testing with that yet? >> We use Grockbot for a lot of different like managerial tasks. Um, it's really quite impressive. We also use Cognition's Devon product, very similar product, cloud agents.

21:59 Uh, cloud agents I think work very well because the environment is very controlled, right? Like if you're using your agent on your computer and then your computer disconnects from Wi-Fi, it's game over, right? So like cloud agents really do solve a lot of problems. >> All right, last question to you is we covered this uh this project last week. gets arrow and essentially this is an you know an RWA you know platform essentially allowing you to earn some >> you know some profits within RWAS themselves but it but what's

22:30 happening is this is on Robin Hood chain what about liquid on Robin Hood chain at some point is that ever coming down the pipe where you know obviously we can see vaults all sorts of creative ways to utilize Robin Hood chain >> we're absolutely going to support we believe in providing like the best experience for users we don't really we're we're not like we're not married to a chain in particular.

22:52 If Robin Hood offers like high amounts of yield, we'll integrate Robin Hood chain. We'll obviously support deposits from Robin Hood chain. Uh we've listed many Robin Hood chain tickers like Cash Cat Ponds, major ones that aren't like, you know, super sketchy. >> I think that we aren't tied to a chain. Uh we aren't tied to a chain. We just generally want to give whatever is best for users.

23:13 So if Robin Hood chain offers massive incentives, we'll let users like take advantage of that. >> Okay. All right. Well, there you go. Listen, Franklin, we're going to continue to watch this because I think this is getting ready to heat up. >> We are going to see a lot of AI integrated pers. I think we're going to see a lot more just in AI integration into crypto.

23:32 And I don't know that this is the end of this. I think we're probably going to have >> Absolutely not. I'm sure that we're a big trend here. >> I do not think this is a flash in the pan moment. I think undoubtedly people one way to think about it is that everything kind of water flows uphill to downhill right and I think generally human effort also takes the path of least resistance and AI no matter what you think of it definitely lets you do things with less resistance so in that sense undoubtedly there will be more

24:02 and more AI that's an intrinsic rule of the world >> I like it Franklin Wang thanks for coming in today we appreciate it >> thank you for having If you like this video, hit like and subscribe. Drop a comment down below and also join our free private member group. The link is in the description.