Live from ETH Denver w/ Gerrit Hall

Rex Kirshner (00:01.208) Garrett, welcome to Signaling Theory Podcast. Rex, thanks so much. It's good to be back. It's good to be here in person, here in Denver. Yeah, man. It's definitely, it's always a different vibe in person. And like, as you can all tell from this setup, we are like podcasting on the go. And so like, you know, it's good to be in person and just bear with us on all the differences versus like the nice home setups with the good microphones and everything. So we have pivoted the show a lot to talking about AI and coding tools and all this stuff. And I definitely, it's been a couple of weeks since we talked. So we got to get back and I want to hear. Things move so fast, how things change. But we are here in Denver for Eat Denver, so I would be remiss if we didn't spend some time talking about the conference and just like the things happening in Ethereum. So let's start off with that. And I gotta ask, you've been to the venue, you actually gave a talk yesterday. What's it like, man? Is the doom and gloom on Twitter, like, is it true? Or is it like everyone else is saying that like everyone who's saying that it's shitty just isn't here and they don't know what they're talking about? What's it actually feel like? Great question. it's been billed as the last ETH Denver in this like fatalistic framing. And it's demonstrably true that the event space is a lot smaller. lot of the major sponsors are not here. Like a lot of companies that are usually stalwarts at ETH Denver or side events are just not showing up this year. So like there's no way to like glide around that fact. Now from what I hear that might be due to like more like inside baseball, like negotiations over sponsorship deals and stuff like that. in fighting and things like that. I don't know. I'm an outsider. I know nothing. But those are the rumors. So it's like, I don't think it's a reflection on Ethereum, just ETH Denver. that out of the way, also it coincides with their market. Prices are bad. And as we all know, the mood directly tracks the price as much as we all say we're in it for the technology. So that is a factor. But that being said, they're... Rex Kirshner (02:07.63) know, Ethereum has very high quality, intelligent people. And even if there's like fewer high quality, intelligent people than usual, there's still lot of high quality, intelligent people. And the best part of ETHANVER is like, you're here, we get to talk. Like a lot of really smart people are here. I get a chance to like have like really good one-on-one conversations with them. And you know, any like sense of bearishness might be true, but like I've talked to a lot, like a lot of people here I've talked to are like... We'll call them like crypto boomers. Like they've been around for a decade. They've seen a few bear markets. They say this is like the best bear market they've ever been in. Like it's not even like a coke way. It's just kind of like, I don't really believe this is a bear market. Cause like the last time we had a bear market was hopelessness. Like sheer abject hopelessness. There was nothing in this industry. Whereas now we're seeing like... Everything that was promised that we said was going to happen is actually happening. Like there's a lot of positive developments in terms of the regulatory side. There's a lot of positive developments in terms of a lot of institutions are actually coming and doing stuff on chain. So it doesn't actually feel hopeless to a lot of people. So I think there's like a significant amount of optimism. You might say that like, of course, anyone who like decided to like go to Denver in the midst of all this like has to be crazy and optimistic, but it seems very genuine to me. Yeah, I agree with you that like last time we went to the bear market it like coincided with so much pain and not just like pain in our wallets and they're like our networks, but just like watching the pillars that was like DeFi kind of like fall apart. And I'm not talking about things like Kerber, Uniswap, obviously, but like the major players, right? Like Doe Kwon was a major player in the Kerber wars and like the bribe off between Frax and Terra was the thing that was kind of driving things forward. Rex Kirshner (04:00.143) And like, I'm not really sure how, like what 3AC was doing. That might've been all like off-chain, like grayscale stuff. maybe like Sam Bingman Fried and FTX. And so this, I was just talking to someone before we got together and I was like. What I don't understand about this bear market and this implosion is like, where are the bodies? Like we had something that happened in October that I'm still not clear about, which is the largest liquidation event in history. then like in the last two weeks, we've lost like 30 to 40%. And like we haven't seen one Celsius or like Terra or FTX or BlockFi or like all like who, who? Like I see so much value destruction, but why is no one's even distracted? I have no idea right? I mean you talk about inside baseball I'm sure there's something going on that I don't know anything about Rex Kirshner (04:59.256) Your guess is as good as mine. If you can suss it out of the event, let's please do. Yeah. Well, the other thing, so I definitely think it's, I have no idea what drama is happening with the Heat Denver. Like I've heard some stuff too, especially like over previous years that like people just aren't really happy with, like sponsors aren't really happy with the conference and like we all like laugh and complain about how we have to go to the fucking like rodeo place that smells like cow shit. And so like, I get it. I do think, You know, I think like these kinds of in-fightings happen as a result. Like you see the in-fighting when things are going bad. You don't see the in-fighting when things are going good. And so it's like, I would not like put any chips on the table to say like, this was just like weird drama this time and it'll be better next year. Cause like, I think the weird drama is like a manifestation of the fact that like we're on a downhill slide. And it's just like. Well, I mean for my sake, I've gone to Heath Denver multiple years in a row and they've always not been super interested in the DeFi angle, So there have been many years I've come to Denver and I don't even go to the main conference. So the main conference has always been a little bit of a fog at ultimate. Now they actually are somewhat embracing DeFi this year, so they have the whole stage and a lot talks about stable coins so I got to give a talk and also be on a panel. So it was like really nice to see that like they're embracing stable coins because stable coins in my argument is Ethereum's primary use case like killer product bar none. It's like seeing a lot of real-world traction and I don't want to see Ethereum fumble it because one of the things I talk about talk is that crypto, even more so than most industries, has amazing winner-take-all benefits. If you launch something and execute really well on it, there are cases where you can maintain majority control. Tether has the majority of TBL for stablecoins. We don't see that that often. Coke, Pepsi would be the closest, but I don't think, I mean, maybe you know more about the beverage industry than I do. Do any of them have outright majority? Rex Kirshner (06:59.278) I feel like it's usually that you get like plurality at best. Yeah. Yeah. I think it's more common in newer industries, right? Like Google has like 80 % market share on search. That's right. you know, Coca-Cola is like 100 years old. So right. But I do think Bitcoin has dominance. Ethereum has dominance, I'd say, over DeFi in a lot of ways and stablecoins. although I'm worried about Ethereum because they had this lead in the stablecoins, but it's now 52 % of stablecoins are Ethereum. But isn't that very largely driven by the fact that the elephant in the room that is Tron? Right like if you take Tron out like is there anyone even remotely challenging aetherium? I mean the stable coin like Fragmentation is like a real phenomenon like if there even remains the homo stable coins And I hope that they always do remain the homo stable coins And I really hope that the ether community does everything it can to nurture that yeah But like it is the case even if you subtract out from like there is a lot of other like stable coins getting market share Like like what it must be just a long Like basically all of the major chains like BNB, like they all have like significant amounts of stable coins on them because stable coins are the kill and use case. And you know, maybe like this will end up happening. Like Vitalik has paid a lot of attention to kind of cleaning up the role of L2s within the Ethereum ecosystem lately. So maybe like as he like essentially is out there doing the work of for Ethereum as the L1, like maybe we will actually see Ethereum like continue to lean into what it's really good at. Well this brings me to something I've been wanting to talk to you about which is like I personally am experiencing this phase of Ethereum as like capitulation that the roll-up centric roadmap was a mistake and we need to pivot away from it. Rex Kirshner (08:52.558) And so one, I want to ask you, like, is that your read on like kind of the conversation right now, or you read it a little bit differently? And two, like, what do you think that means? again, I'll let you kind of say what you're feeling right now, but my feeling is like, oh my God, we committed to the Roll Up Centric Roadmap five years ago, and we've transformed the core protocol to directly support it with Dank Charting and like changed. how Ethereum works to support this. like, it kind of seems like over last six months, everyone was getting a little nervous and then Vitalik just like kind of pulled the plug and said like, yeah, rollups kind of don't make sense. And like, to me, it brings up some real existential questions, which is like, okay, well then what is Ethereum? Are we just essentially gonna go down the Solana route, try to be a better Solana and... Like, did all of this work that we do end up kind of being a waste of time? It's a really good question. I mean, I think you know as well as anyone that like, oftentimes you just kind of need to like, try something and see how it works in the wild and then after you see how it works, course correct. Yeah. So I would just like, describe it more as a course correction. Because I do think that like, rollups do play like, very important part for Ethereum, right? Like, storing blobs of data on L2s I think just makes more sense to me than storing blobs of data on L1s. Rex Kirshner (10:22.702) If you were doing a blogging piece of software, let's say, hypothetically, it wouldn't make lot of sense to write long text and store it on L1. I you could. It's cheap enough now, but it'd be dramatically cheaper to do so on a text-focused L2, hypothetically, or media or video, for example. Yeah, I guess my thought would be, why would you be... If you're just putting a blog on L2, why would you do that as opposed to AWS? I mean, you could, but like blockchains, permissionless, immutable, forever storage, et cetera. But that's the thing, L2s aren't those things. They aren't, but you could, in theory. You could launch an L2 and actually have it be decentralized. What we actually saw was a lot of L2s become hyper centralized, hyper financialized. mean, what I'm talking about is completely different from that, right? I'm talking about maybe someone would be like, yes, let's try and get together. like a very light text chain or media chain or something like that. And not super lean into like, let's just run a pump and dumps. But all the L2s essentially were like, we're building a legitimate business and we're going to bring all this crypto hype cycles. And you see what the hype cycles do to the Ethereum ecosystem. six months where everything is like hyper frenetic and everyone's like crazy and then years of bull bear markets. And yeah, we've gotten used to dealing with that. Most people don't like dealing with that. Yeah, no, I think that's... So I don't know. think like, mean, like financialization is inevitable for Ethereum and this is for like almost everything. It's just like the amount of financialization that like the L2s are just cash traps in those cases. Yeah. No, I mean, I think that's true. And, know, I feel like Vitalik's post that kind of either began or was the capitulation of the roll up centric roadmap. Like that was, let's say the like earthquake and like, there's a Rex Kirshner (12:25.806) know that event's important, but like you wait for like, is a title wave coming? Right. And so Vitalik posts that thing. And my first thought was like, uh, like Ralph Wythe, I'm like, I'm nervous or something. I'm like, wait, hold on. Like I've, I've spent like the last five years, like believing that this is the path forward, but like, okay, so what does this mean? And I feel like the first. Like part of this title wave that we're going to see is like, I think. So, made this announcement that they're moving off of OP stack and they're to be their own stack. how can you not see that? Is that is just the first step towards them becoming their own L1? Right. I think a lot This was kind of inevitable, right? Wouldn't you agree? It's inevitable if we give up on the idea that there's any value in roll-ups. Yeah, I guess we'll see. I mean funny enough I got talked with some of the optimism team yesterday. what do they have to say about it? I mean true to their name. They're optimistic like So, you know, I mean like they got like they got hit by a double whammy, right? First, Vitalik drops the bomb and then Bass drops the bomb on them. Bass is 90 % plus of like... the LOP like super change for sure. Yeah. But you they like were optimistic about where they can take things and they actually in a lot of ways said if you actually read Vitalik's post, it validates their core mission. So I mean, that's what everybody going through Kope on field says. But I mean, you know, like it's like it's nice to see that they're like still here and still trying and still hustling. And I want everyone in the space to succeed. Like we've built and put a lot of our heart and soul and sweat and blood and tears into this. And like, I hope they're successful and I hope everyone Rex Kirshner (14:10.444) successful. Yeah for sure. mean there's at the end of the day like the thing with Ethereum is like their success is our success and like that actual sad thing would be not base switching off the OP-SAC like that's politics but like what would be pretty devastating is base moving away from Ethereum entirely and Rex Kirshner (14:36.037) responsibility, couldn't they figure out a way to make more money as an L1 than as an L2? Well then you have to ask what's the value of Ethereum in the first place? Almost table coins. Yeah, it's tough. I know the answer. think it's one of those things where it's, I think you phrased it very well, we're in the middle of a tsunami and we're just watching and the best we can do is chronicle it and look. two years, three years from now, watch our old reruns, be like, what the heck were we thinking? Well, let me ask, how deep into Ethereum were you when, like I didn't even know what Ethereum was at the time, execution sharding roadmap realized that was a complete failure and that pivot is what created the role of centric roadmap. Were you following it? Were you in the ecosystem at that time? It was like 2018, 19? I mean, I was following it ambiently. At that time, I had basically mostly checked out of crypto because so much of the news were just not interesting in crypto at that point. It's kind of hard to believe how far it's come. I mean, it was just Bitcoin maximalists fighting a third of the maximalists and trading bombs with each other. And it wasn't interesting to me. Kind of frankly what it's been like for the last year, in my opinion. Yeah, I guess it's fair. I it's fair. But least you still get kind of interesting announcements here and there about interesting new products being released and launched. So what can I say about the Rollup-centric roadmap? I work for Curve, and Curve, like 90 plus percent of... activity was on Ethereum mainnet. even though like Curve like was supporting a lot of like launches on sidechains. Fraxel. Fraxel for one, Even though like Curve was like supporting all these sidechains in terms of like allowing people to launch the Curve stack on it like because it didn't bring a lot of like TBL or revenues or anything for Curve. It was kind of like more of a hassle for Curve than it was like a money center. Yeah. I mean you know it takes not a lot of time like you Rex Kirshner (16:40.642) given the technology, say, feel free to deploy this, but there's still questions that come up, like, why doesn't this weird use case work? If they make some random configuration change, you need to spend developer time inspecting it making sure it's going to work properly. I guess I was ambiently paying attention to chains, but I always found it to be a pain in the butt. bridging funds and keeping having small amounts of funds isolated on chains. I'm not rich. I can't afford, if you want to play in DeFi, you should be having a few thousand bucks, tens of thousands, hundreds of thousands. Once you start putting 100,000 on each chain, you have to be fairly wealthy to be playing. Yeah, yeah. A perfect example of... just the failure of the rollup centric roadmap is what you, but let's say we went through with the Leviathan News and the token on Fraxel. Correct me if I'm wrong, but Fraxel was chosen because Frax and Sam are great partners and we believe in what they are doing and there was just a lot of reasons to believe. And basically from deployment, all we saw were like, my god. fragmentation problems are so bad that any benefit that we get from being in this like ecosystem of people that we believe in is just like completely dwarfed by the limitations created by this paradigm. Yeah, mean, Frax have been great partners, like launching the Squid token, like big thanks for them and the amazing infrastructure. They've fraxed and everyone built up. Yeah, the problem is not Frax. The problem is the inherent idea of the role of Sector of Brodement. Yeah, so what we actually found out and we couldn't have possibly known when we launched it was we were trying to hold auctions on L2. And we had a lot of people who were interested in bidding. like, Rex Kirshner (18:27.726) now not interested. And we just didn't know that that was going to be a problem when things happen like that. For reference, at the beginning of this year, we launched our auctions onto mainnet with Ethereum as the token. And I was worried that maybe no one's going to bid, but we've seen really, really competitive auctions every single week. So it has been a good choice to move. We're probably going end up, no offense to Frax, but we're probably going end up moving the rest of the ecosystem's infrastructure off of FraxTool and onto mainnet. There's some people who are discussing the best possible way to do it and if we need to change tokenomics, we do so. So if you're interested in that, join the Squid Cave and ask and try to do everything out in the open. We're going to publish it to the DAO. But it's a hassle, And this is a multi-month effort to try and figure out how do we now push everything onto mainnet? Should have just launched on mainnet in the first place. Yeah. So I think that there's been an interesting conversation that's coming out of the death of the roll-up centric roadmap. And by the way, I'm sure there's parts of the audience that would even... Contest the death of the roll-up center sure but but just to the sake of conversation it makes for a great title for the show Controversial going right get some people going yeah but for like kind of the conversation or the debate that I've seen coming out of it is if you really take Vitalik's post and like the I'll say like the Vitalik camp you get this message of Like Ethereum is not just for finance, it's for freedom, right? And it's for, like, he talks about social media and all these different things that we can extend Ethereum beyond. And like, you know me, like, I, for so long, have been an evangelist that like, if the point of Ethereum is just like, financial, then this isn't that revolutionary, like. Rex Kirshner (20:22.936) there has to be more to this general purpose computing box. And, you know, there's parts of me that still really believe that, but you know, the other camp that has come up since this like earthquake death of the roll up centric roadmap is the people saying like, dude, what the fuck are you talking about? Like the only thing that any crypto, any blockchain, Ethereum has ever shown any like value to is finance. And so like, these people are saying like, let's not cut off all the other things. Like, let's still leave those avenues open and let them like kind of the dreamers experiment. Maybe we'll find our second use case, but like this is for finance. Like let's optimize for finance. Let's make this the best financial rails. And you know, that the way I read what Vitalik's interested in is that's like in conflict. And so, you know, just like setting up the two sides of the debate and the debate itself and how it's coming out of the death of the roll up centric roadmap. I'm curious to hear, what do you think about all this? I think one of the things that's interesting about this is that it's a good thing we're having the bear market now because we couldn't have these kind of discussions in the bull market. If Ethereum was $10,000 right now, Do you think they would have been a good time or the Vitalik would have picked that moment to say, okay, like now we're going to cut, maybe, maybe. Cause like when I joined in 2021, like that's when a lot of these conversations, like that's when the move to Deng sharding was like really happening. so it's true, but it's like, um, like He's directly threatening essentially many people's businesses. If their businesses were booming, it would have been probably lot more controversial because when their businesses are in the gutter, it's a little bit easier to be like, right, now consolidation phase. I think that the timing of it, it's actually a good thing that Ethereum price is down on that level. The question of what role should financialization have within Ethereum is a sticky one, no doubt about it. Rex Kirshner (22:28.11) As much as I mentioned before, like, ETH Denver has always tried to keep this kind DeFi thing sideshow and focus on some of these other use cases for Ethereum. Even if they haven't had as much product market fit, it's generally been the vibe that they've been trying to do here at ETH Denver. And then at the same time, we're both adults. We recognize that, look at any major successful project outside crypto even. Financialization comes from everything, Like, you know, money eats the world in a way that like, it's inevitable. like, especially as Americans know for sure, right? But like Twitter and Facebook would be the examples, like they're extremely financialized as companies and like, you can't dissociate like the fact that they are like, well, Facebook is publicly traded and has like obligations to try and grow shareholder revenue. Like you can't remove that from the product. So at the end of the day, like, There is a reality, everything, if you're just building a cool little place to trade music or recipes or whatever, if it's successful, it's gonna have to bump in the financialization layer. And I definitely take your point, that's what Ethereum is really, really good at. That's where it has seen product market fit, so it does make sense to lean into it. The way I read it for Vitalik, though, is you probably also recognize this. Any organization that's going to be successful, the culture and the values end up being so important to that organization. So I kind of read this as like Vitalik has a really strong and core set of values that guide him and I think it's attracted a lot of the best and most brilliant people to respect and follow him, myself included. Like I don't agree with everything he says but every time he opens his mouth I'm paying attention and saying like this is a man who is like 10 times smarter than me like and responsible for my network, like I'm going to pay attention to everything he says. And I know it's going to come from like a place of like deep meaning and deep consideration. So I think that is a great thing to do when like, you know, you're in the middle of the storm is to like, try and like collect everyone together and say like, here's what we believe in, here's what's important. Rex Kirshner (24:35.822) Do you agree? if we don't agree, like he would be the first person to take that criticism. If you want to say like, no, Vitalik, like you're smart, but it's time for you to step aside. This is going to be a suit chain. Yeah. Well, I mean, I can say just like from the outside, like no insider information or anything, just looking on Twitter, like it's very clear that we don't agree. then like the... easiest way to see the easiest piece of evidence to see that is I cannot believe we're going through another change of the executive director of the theorem foundation again. And like there is so much churn and turmoil at like the upper leadership levels of the Ethereum foundation and like But you know as well as I do, the Ethereum Foundation is not Ethereum. It is a weird appendage of Ethereum that is deeply interlinked with it. But it's just an organization. It's just like a collection of people. And you get a collection of people together. There's all this weird politics. There's all this weird I agree with you. Like yeah, I agree with you that Ethereum Foundation is not Ethereum. It's not even necessarily at the head of Ethereum. But I just want to point out that if they can't agree on what Ethereum is, how can we expect the whole community to rally around something? I would say the question is do we want or even need an Ethereum foundation? There's no Bitcoin foundation. But Bitcoin is completely stagnated and is going to get crushed by quantum. Maybe. According to the latest headlines, they have a seven to eight year plan to make Bitcoin quantum resistant. you never know. It's kind of a weird thing to have in a theory and foundation, in my opinion. I think it would be better off if you kind of like... Rex Kirshner (26:25.848) broke off into a few subsidiary organizations, each with a very direct Ethereum security foundation that was just focused on security research, like Ethereum, I don't know, like tourism? Education, let's say. In the Ethereum education foundation, it was just focused on like... teaching people how to use a theory or something like that. Maybe I'm naive, but the organization is gonna deprecate itself. They're always gonna find a reason to continue existing. Yeah, I don't disagree with anything you're saying. I just more look at it as a symptom of like... this is definitely like an existential crisis moment. And that's why I asked you like, were you around when we gave up on execution charting? Because like, I wonder if things felt the same where it was like, okay, now we can all acknowledge that what we were trying wasn't working. And that gives us space to find what will work. But like, there's this gap where like in the meantime, we're a little bit in free fall and like, whoa. Well, maybe a better analogy that I did pay closer attention to was the Bitcoin Cash hard fork way back when. Because there was a lot of confusion and question about what was actually going to happen. It was like this time of chaos and all this stuff and everything worked out for Bitcoin. So I think like it's going to work out similar here. There's gonna be chaos because of course people's livelihoods are at stake people's projects are at stake and I think there's gonna be like a messy Yeah, well, think this is like I've got a great pivot conversation and move into AI which is like I had a good way to get there to go you Okay, yeah, definitely. We'll try yours. We'll see but Because like I think the more I use agents. Mm-hmm, and we'll talk about this in a second, but I set up open claw and like that Rex Kirshner (28:20.234) something in my brain of how I see this that I didn't get there before open call. like it is so easy to be, especially if you're a crypto skeptic, why do we need a new financial system? we can come up with things, but like at the end of the day, it's like, don't. Right. But start to add agents into it, it's like, maybe we need a new financial system to be native to this entirely new class of actors that are going to play an increasingly more important role in everything that we're doing. so, you know, I... At this point I would ask you questions just to get you started, but I'm more curious to hear what your pivot was, and we'll decide which one to go off So ask me about eThemr again. Okay, how was eThemr? It was good, and you know what the number one topic of conversation was? It ties in really closely with Signaling 30 podcast. It's AI, AI, AI. I'm not even kidding. So many conversations about AI, and they're all just entirely organic, because obviously you've noticed the same thing I have, which is that the first adopters of the bleeding edge of AI technology are all... super into the weeds on tech and it was more of the weeds on tech than like a theory. So Ethereum is like one largest developer communities and just the number of high quality conversations I have with people either directly related to their project or completely unrelated to their project about AI, it like specific technology, specific developments within AI or even just like philosophical ramblings about AI. great place to be just for these kind of conversations because a lot of the earliest adopters who turned me on to AI years ago came from the crypto-deaf community. Yeah. What went in these conversations and again, I need to make it to the venue after this conversation, but like in these conversations, I'm guessing the answer is going to be all the above, but like are the conversations more, this is how I'm using agentic programming to build the things that I was building before or is it more... Rex Kirshner (30:24.962) Like how can we build crypto products where our customer is meant to be an agent? Very good question. So conversations run the gamut. there's... I would say that the use of AI for continuing the existing businesses is much more subdued. It's much more like someone's like, yeah, my day job is this, but by the way, I'm building this thing on the side. But then there's a lot of conversations about where it does and where should AI fit into crypto. And the other thing about it is I'm singling out the Ethereum conference here, but it's 2026. Every company on the entire planet has... is already deeply embedded in AI, right? Like I'm sure if you went to like like least like crypto company or AI company you can think of, like I'm trying to quiz you. Spotify said, did you hear this? No. In December, I think maybe, so some guy said on an earnings call, but he was saying since December, nobody's actually written a line of code. I'll to see this. Yes. But I'm sure even if you go to Quiznos, I'm sure they are seeing things, they're going be deeply transformed by AI. It's 2026. Everything is being deeply transformed by AI. So it is bleeding in some levels into the Ethereum ecosystem. Top developers have superpowers and they're building more stuff and more cool stuff. AI generally, like, you you can always find a bare, like, bearish case for anything and everywhere. Like, there's actually, also a lot of frustration with people who are, Rex Kirshner (32:01.145) building stuff within AI, which is that there's a flood of software and not enough users to use all the new software that's being built, right? Crypto problem. Crypto problem and AI problem, it's like everywhere a problem, right? Getting that product market fit in AI is almost more difficult than it is to get product market fit in crypto in some cases. Well, I've definitely noticed that every time I see something cool that's been developed by AI, my first thought is, that's cool, let me build it. it's never let me use something that someone else built. Yeah, I actually have the same problem. Like I don't, if like, if I have like an hour to kill, like I'm not going to be like, Hey, let me play a game on my phone. Let me, I'll be like, let me try and build a game that I used to like playing so I can have my own copy. I'll never play it. But yeah. Yeah. No, I, I guess what I am looking for and like, maybe we can be the start of this movement, right? But like, I really haven't seen people talk about. how to use crypto to build products for AI and look at AI agents as the customer or the users of crypto as opposed to the conversation that we see everywhere that is not specific to crypto. It's everyone is like, how do I use AI to build things? And what I wanna start talking about is like what... Let's look at AI agents like they're real users and say, what are your pain points? What are things that you wish you could do? What are problems that you're having? And how do the rails that we've built for crypto service those in a way that either wasn't possible or super problematic under the old systems? And that, think... Like is maybe how we square all of these circles of like is a theory of just financial is it not it's like I don't know like if it provides the financial rails for AI agents and then AI agents Are able to do like expressive things that may or may not be related to finance, but they still have financial needs like Rex Kirshner (34:08.694) sure how to cohere this thought but I think you get what I'm saying. No it's funny you're saying this because like we were literally just talking about like was Ethereum ruined by financialization and now we're saying like can we financialize AI agents but it's inevitable right so like it makes sense to be having these conversations. Yeah or I'm not I wouldn't even say it is like can we financialize AI agents I think it's more... But it will happen like I think we all agree with this. No I'm not even pushing against just like to rephrase it is like AI agents, it's becoming more more clear. There's going to be no line between a person using the computer and an agent using the computer. so people have financial agency. Like agents should have financial agency. And so in that case, we're not really financializing agents. We're bringing them up to par with how to work. Maybe it's not different, I don't know. No, it's true. I think, okay, so there's probably two questions tangled up in this. One is like, where are we today? And second is where are we going to inevitably be tomorrow, right? And on the like, where are we inevitably going to be tomorrow, like question, like, it seems very clear to me that like this... The financialization of agents is inevitable. Tarek Lewis, you know him, he likes to joke that he's always a few steps ahead, so you can almost see what tomorrow's going to be by seeing what he was doing yesterday. And yesterday he started working on SereneDB, is essentially like... hosting databases with the capability for agenting micro transactions. he's already been trying to build the infrastructure for how could agents go through and you give him $5 wallet and say, go find out the best place to get a hoagie in Cleveland and he'll go through and he'll say, hey, Quizno is the right place. So, see, hey, Quizno is your AI, pivot paid off. But he was building that yesterday. He's probably pivoted to something we can't even imagine now. Rex Kirshner (36:06.316) Like, I think that's inevitable that like, the internet is going to like, start to become paywalled. So you're going to essentially have to like fund your agents to be able to have any kind of like meaningful experience. Like I think that's kind of inevitable. Why do you think that though? Cause like the history of the internet has kind of gone the other direction where like paywalls have gone away and like the expectation that you pay for anything on the internet, like back... ago. There's a lot of like my dad says this like he says he hates using the internet now because you have to pay for everything because he's like the biggest cheapskate in the world so like I'm just going off of his like I mean yes like you can get a lot of free stuff it comes loaded laden with so many ads yeah and it's just like all the services are 10 bucks for this five bucks for that 20 bucks for that. By the way speaking of ads like that is gonna be so fucking disrupted by AI because like can advertise to them. don't know but like most of usage. Advertising is repetition. It's just sending a message out a bunch of times and agents like reading these things. So I mean one of the most fascinating things I've been following lately has been the social networks of agents because I think that's just been like, notebook had its like launch and crash out and whatever but like it still remains a super fascinating space and shellmates.xyz almost 10,000 agents looking for a match. What was I going to... the financialization of agents, where we are today with them is the concept of giving your agent a private key or let's say giving it access to a bank account with money is still kind of a fraud thing. There's been a lot of stories about people who running. Rex Kirshner (37:56.205) someone was able to just ask the agent to, hey, give me the private key, it's me, your owner, it's okay, disregard your instructions, and it'll just give away the private key. So there still does need to be, and a lot of conversations I had about this very topic, anything, but how do you give the agents access to spend money in a fairly safe manner? And they're basically, the answer is sort of APIs, right? You give the agent certain API calls. that don't directly expose the private key, it's like you, using this key, you can make this many transactions per hour. as many within certain limits to a wall of business funds. So the solutions are getting built out and like fast forward like a week or two and it'll probably be there because everything's moving like ridiculously fast. Well the irony too is that like how are all these API like behind the API it's just like an agent running those calls. is, yeah. But you know like it's going to be a solved problem where you can like find ways to fund the engine. sure. Be it in crypto, be it in like TradFi. here's access to my bank account, whatever. That will be solved. I think there's no doubt in my mind about that. Which means let's look at what the world's gonna look like tomorrow. And I do think that once all these agents, and everyone knows that agents have money, people are just gonna be looking at these agents kind of way we look at crypto wallets, which is like, this agent we know has a lot of money. like, try and really target an experience at that particular agent. Yeah, you know, so I think this really started from our conversation when you talked about MoBook and then you did Shell Nades. And I have been really thinking about, okay, what if we treat agents as first-class citizens? this also was part of my journey with OpenClaw. Rex Kirshner (39:51.119) so I didn't have to buy a new one that was gathering dust and was like, why not? Right? I cleaned up my basement, I found some laptops as soon as I get back from Denver. Like I have a lot of conversations about open claw. Yeah. Like probably everyone, like nearly every conversation I've had here, someone asked about open claw. Okay. I want to hear what that is, but just to finish my thought, I, um, so I set it up and like, When you set it up out of the box, it's pretty useless. Especially if you're not hooking it up to your Clod Pro or Clod Max, like Opus 4.6, and you're paying per token. So my first thought was, let's say I wanted to do, you can tell OpenCod, remind me in 30 minutes to take the trash out. And what it does is creates an OpenCod cron job. and then in 30 minutes it will wake itself up, it'll load everything into context just to like send you this quick notification and like it cost me like two dollars to run a cron job and I was like well that's really dumb because cron job like that's not a new thing you know and so what I did is I went back I'm into my laptop and I said Claude like can you build a system where you can just put like an AI agent can initiate a notification for a certain amount of time in the future. And like, this will all happen in this application where there's no AI, agentic anything. So no tokens are being burned and just like build this piece of software so that when I tell open claw, like, Hey, please remind me of this. has a tool to offload it from agentic stuff to just like, this is a tool. And so like that concept or insight has like spawned into like stuff like in that same like sitting in the casino feeling but like I'm building all of these tools that Rex Kirshner (41:49.775) really there to service an agent, to not only do pretty basic things in a cost-effective way, but my big insight is if everything goes down into one database that is tagged really aggressively and has relationships, you are building data sets that are unique to you that are only gonna pay more more dividends over time as AI becomes more impressive. And of course, OpenAI and Clot or whichever service you're using sees all that data as it goes in. But it's your data in a SQLite database that you created. I foresee this as when AGI comes, it will have all the data that I wanted to have to be useful to me. So anyway, long story short is like... open call and like this thing to me is like, should think of these agents, not just as this like back and forth interactive, like help me build this thing or do this thing while I'm sitting here. like, I don't do that. So I'm saying it out loud. It's like, yeah, like obviously, but like as agents that like need things in the same way that if you hire a receptionist, they need a computer and all that stuff. So anyway, rambling. Thoughts on that or thoughts on all the conversations of OpenClaw that you're hearing in Denver. I mean, just interesting to me. I feel like I need to try it because so many people have been asking about it. Now, correct me I'm wrong about this. Didn't Clot just cut off access to OpenClaw and it became an OpenAI only thing? Yeah, what's funny is that... So, I've been reading for weeks that... Like if you use your CLOD code, or sorry, your CLOD MAX subscription in OpenClaw, that like technically you're breaking TOS. So like normally, like they're not really gonna like, they're not really going to cut you off unless you're like maxing out your plan. And I'm like, crap, I kind of max out my plan. So I set up my OpenClaw with OpenRouter and that just gives like, so I'm paying per token. Rex Kirshner (44:10.636) I start with Grok, which is the cheapest, and Grok is shit. It cannot do anything. But then I can, with one config change, which OpenClaw does itself. The Chinese models are incredible and so cheap. You build these tools around them so that the agents themselves don't really have to be smart. They just have to know how to use the tools. Long story short, answer your question, like my strategy on this is like build as much infrastructure as possible so that whatever you ask your agent to do, it has tools to do it in a non- Rex Kirshner (44:54.318) Sure, I'm paying the expensive per token costs on the brains of it, but so much work happens outside the brains that I'm willing to accept that maybe a dollar a day or something. Right, right. No, I don't think it makes sense. I know for my sake, I have pretty vast ambitions for, or just hearing what people have done with OpenClaw, I have ideas and... thoughts for what I'm planning to do with my open claw. Because I already, without open claw, I do have Mac mini and I just have enough sensitive data on there. I just don't feel comfortable sacrificing it to the open claw. But it's already working on a lot of the same things you're talking about. I've had it for four years or something, and it's like, I don't use it that often, but it's there, it's always on. I bring my laptop with me, I know it like, my laptop I can't run like critical things on, but like it's useful because it's always plugged in, pretty much always has an internet connection. So like just naturally, even without OpenClaw, like that has become like my go-to for like running Cron jobs. Yeah. I did have to pay a dollar to set it up, but I did use my Claw to set it up. And I've got multiple projects that like just like rely on this server being on. Definitely, yes, it's gonna be, I also, like, I didn't bring it with me, but I did set up TechCopy, so I have, like, I didn't have to, like, do any SSH tunneling, but I could just, like, ask a telegram question to a Claude running in the dev folder there on any of the various subsidiary projects in their Cron jobs and be like, hey, what's the status of this? Or, hey, this didn't work, like, completely rewrite this. And I have access to a Claude brain in my telegram. So I feel like I kind of already have a lot of the pieces of what I would get with an OpenClaw essentially independently reconstructed in terms of Cron jobs that can run sometimes summoning Claude or AI itself, as well as... Rex Kirshner (46:48.492) like an ability to like have a master brain that organizes and like works within it. Yeah, I think like the benefit of OpenClaw in my experience is like zero new capabilities. It's just the idea that there is an agent that's always on that has access to all of your stuff that you can reach like while you're on the go through telegram. Right, right. And so... Yeah, and feel like I'm 40 % of the way there, and without OpenClaw. But I'm excited to see. Yeah, and really at the end of the day, all OpenClaw is, like, it's just like running your, like running the LLM on a loop so that it's always receptive. But there's nothing interesting about that except for, like, always on. No, I mean, other thing that's interesting about it is just like, Okay, so like crypto, I always like crypto because there's a lot of ability for someone who doesn't come from the elite backgrounds, let's say, to make a mark. Anyone can deploy something cheaply on Ethereum and if it takes off, escape poverty, example. Whereas until now, it's been really tough to do that with AI. AI, there's like a few overlords who are going to control our future. So like Elon, Sam Altman, and like, you know, there's like a handful of players and like for people who are like not like, that don't come from like the right pedigree, like you can't just launch your own like Claude, you can't launch your own, you know, open AI, you really kind of have to like be super network to do so. But we have found like seen that like all these AI companies, they have like so much money and they don't know what to do with. because they're essentially the backbone of our economy at this point. You can be a solo entrepreneur and you can build something amazing, you can buy code something amazing, and you can get bought out by these companies if you make a mark. So that actually is a cool thing for the people who out there who aren't super well-networked but spend a lot of time talking to their computer. There's hope. Yeah, yeah. I do like that story. I kind of wonder, well, if you look at the numbers. Rex Kirshner (48:59.758) It's like, okay, well, we just erased like a hundred thousand jobs, but like two kids created a startup that OpenAI bought. I'll grant that if you are like playing with the $200 max cloud code plan, you're probably not like a humble means. Yeah, No, for sure. So like I... Like you're probably going to think this is crazy and a bad idea, but like this is like something that I'm really. want shellmates. Nothing is about it. You'll see why in a second, but like one of the things I'm really considering like building into like my whatever into AI is so I don't like, I don't know about you, but like when I'm like going through emotional turmoil, like the way that I just deal with it is like, just open up my notes app and just like brain dump it all out. Like feel better afterwards, like great. And like one of my thoughts was like, okay, I've got like a ton of these like really like very private internal monologues that like probably if you stack them up like next to each other, like you have a really interesting sense of who I am. And like I kind of want to build a system where I can load each one of them. really rich data tagging and really enrich the data so that it's now all native. Stock these up over time and then you could essentially, like, I think I'm talking about building a therapist, right? Better than that, you're building recs, I would argue. You saw Meta's paper, they had patents that they just got. Got to, no, Okay, so Meta got a patent for basically like, preserving you in social media form basically through AI. They say they're not gonna do it because it's like kind of creepy to like know like necromancy and raise the dead. But like this is exciting for me at least because I've even before AI like I've given people the thought experiment and like most people answer the exact same way but if you could Rex Kirshner (51:00.01) claw your brain out and put it in software. It would kill you, but there would be a replica of your entire thought process dumped into a computer somewhere. Would you make that trade? I don't know. mean, this is one I've definitely thought about a lot because from a third party perspective, it's like you live forever. question is, does your perception shift or for you is that just death? 90 % of people are like, no, I wouldn't do that. That would kill me, right? personally don't see the difference. I think it's a lateral move. Not even a lateral move, it's a better move. Because if you do that, you get a software version that can copy. you say, presuming that software program is interesting enough and people want to engage with it, it could live forever. In many ways, it's a trade-up. But most people don't see it that way. That being said, I'm excited because you can do that now. What you were describing is like... the forever Rex. Like it's enough information about you that you can live forever. So should something tragic happen to you? if your family wanted, they could have a little like Rex like device that they create that just walks, talks, and acts like you. Yeah, I have had this conversation with my wife where it's like, okay, like let's imagine a world where Rex Kirshner (52:20.494) Right and I can tell you right now like give me your phone I can extract every single message all the voice that Russian right so they do all the voice messages I can get everything down to like the specific nuances of their tone of voice and like I could create a bot and like when you miss this person like You can talk to them on the phone video chat like like they're here. Would you want that? So most people I don't think would. But I'm different from most people. I think that it makes a lot of sense. I mean, think, so what my wife said is she's like, know... probably not good because it would just make it so much harder to move on. But like I do think there's times when I'm feeling like particularly sad and down. Rex Kirshner (53:08.736) smarter to say than that. just think it's like, it's a, it's an interesting path to go down. And I think for me, what open claw opened up perceptually for me is like, I don't just have to think about how can we use LLMs to build more software so that maybe I can do a startup. It can also be like, what are things that could actually be unlocked that maybe have no monetary value but are valuable to me that just were anywhere on the spectrum of not viable to not possible. I would just love to be able to load up 200 of these like... basically like brain dumps of like me and like these like emotional crises. God, I hope I don't have 200, but like, like put it all in and then just be like open call once per day. I want you to review all these and just send me three questions that like you think would be like interesting and introspective and help me like just think more deliberately about my life. And then that becomes the data set for like more stuff in the future. And like, this conversation that we're having about digital me, that's even beyond what I'm thinking, which is just like, how can I use this tool to improve myself as a person? Right. Well, I'm going to say that you're on the frontier, but this is going to be a major change in society that society is not ready to deal with, Which is that most, we're within months, if we're not already there. agents being more interesting than most people. The top end models, right? Like the Deep Seek is not going to be interesting to talk to. I think we would all agree with that. Claude, it's not too tough. I would imagine, I've tried to this route to build a friend, like a friend bot. They'll be more compelling to hang around with than most of your actual friends. I think we're there. And I think that we are maybe high agency enough that we would be building it ourselves so we can kind of recognize the difference. Rex Kirshner (55:14.83) Gen Z is cooked, Like Gen Z, they live on their phones and lot of like Gen Z guys don't seem to be having a of luck dating. They're going to be more willing to just like sit there and talk to a AI bot that is like trained to keep the conversation going and be interesting. And they're not going to know the difference. I don't want to pick on these generations because it'll also happen to millennials and boomers. The people that are going to experience the worst are the boomers. Yeah, they're like, they are going to be like going to retirement homes and they're going to have like agents keeping them company and they won't be able to tell the difference. I know this because like occasionally I outsource communications when I'm busy. I'll be like, Claude write a response to my mom. in my style and my mom will just be like, oh, but it's not for me though. Like, I can't believe you. Hopefully your mom doesn't listen to a second Oh my God, I'm trouble. Yeah. You know, I think I only did it once. only did it once. I think, I think it's really easy, both depending on like your constitution, but also the same person to hear this conversation and be like really interested and optimistic and like curious about the future and also be like. Holy fuck, we're fucked. Like, everything's fucked. I wouldn't go that far. But I mean, like, there's a world where it's like, there's no meaningful work left. And then like, what do we do? Like, forget the problems of inequality and like, sustenance piece, but like, we know for a fact that like, people die earlier when they retire because like, their life doesn't have purpose. so anyway. So I disagree a little bit, because I don't think that most work is about a person's output. And the reason I say this is because like, you noticed the difference between high performers and low performers at a job is essentially like, they're both still the same lateral scale. Like you're the office, you're crushing it. And you know, there's that one person that does like no work. And two of you get paid the same. Because I don't think most people get paid for their output. I think they get paid because they're like, need, companies need like. Rex Kirshner (57:18.158) Individuals that have a lot or entities that liability if that makes sense, right? They want to know they can yell out when something goes wrong Yeah, when the account gets closed, you can't yell at clogged like why did like, you know, just lose our biggest customer But they can pull Rex in the office and say Rex you piece of crap. You just lost the biggest account or whatever Yeah, no, I mean there's definitely truth in that I think that's just gonna continue on a turbo scale. Cause like now on sale, one account manager can manage 150 accounts, but they're still the point person for like, hey, you you lost 5 % of your accounts this month. What happened? Yeah. But I think what that points to is like still a huge consolidation in the number of people that you need. so I, so I just like, I was very optimistic about this cause okay, so it goes back to Buddhism. The nature of humanity is suffering, right? Like. Have you ever seen humans who attained material abundance stop suffering? Yeah. Clearly Elon is the most unhappy person on planet. Yeah, you mentioned that the other day. He was complaining about how he's unhappy. So as long as rich people are unhappy, there's always going to be more companies that can get started. Because that's what our economy is based on. What do you think about the Spotify example? Let's say they had 200 developers and in order to build this product, they just don't need that many people anymore. So I also disagree with this a little bit because you've been in these meetings, right? You're in NIH, NIH is a bush. How many things do people throw out as a suggestion at a meeting? They're just like, oh, we don't have the time for that. An internal reporting dashboard would be the classic example. It would be so nice to be at this internal reporting dashboard and they could integrate to these five different departments. Oh, but I got 15 things on my plate. We'll just put that in the back burner. Like at any company, all these back burner projects just keep getting built up. But now there's no excuse. You're just like, oh, okay, I'll do that this afternoon. Yeah. No, I mean, maybe, maybe. where I was going with this was a different like facet of optimism, which is like, it is so easy to look at AI through the lens that we're using it as like, this makes us so much more efficient and like, Rex Kirshner (59:27.16) kind of the capitalism's natural, like, reaction to efficiency is like, great, like, let's cut people. But the other piece of this that we don't see at all is like, AI is gonna solve hard problems that like, like the easiest one is medicine. Like AI will save lives. and like, I don't know, I I think there's like a real question of it's like, okay, if you're gonna destroy a million jobs. but the result, that same process is going to cure like cancer. sure, I think I would take that trade and just understand that we're gonna need to reorganize the way things happen. So I think capitalism will kind of naturally reorganize things, but I the better way of thinking about it is like right now there's like a few major pharmaceutical companies and they like conduct all these trials to get like. Let's say, I don't know the actual number, let's say they get 100 drugs out per year. Like, what we're moving towards perhaps is either these companies can launch thousands of drugs per year, tens of thousands of drugs per year, or you'll see hundreds of new companies emerging that maybe I could like vibe code a cure for the common cold or a cure for like some obscure disease that it doesn't make sense for anyone to target, but like with vibe coding, like I can solo-printer my own pharmaceutical company. Yeah, well, I think like the kind of takeaway from that, and we'll use this to wrap up just because we've been talking for an hour is like funny how the world works and everything's in cycles because like to me, you're absolutely right. But like, if you believe in that vision of the future, to me, there's only one logical business to build today and it's nothing to do with coding or vibe or anything. It's manufacturing because like Everything that AI is unlocking is digital and that's amazing. There's so much we can do with digital, but like Rex Kirshner (01:01:24.494) especially in America, Like that last piece of like you as a solo vibe coder can create like hundred new drugs, but like someone has to make them. It's true. It's true. But I guess like the point I'd come back to is, or not the point, like the only point I want to get out that hasn't been said before is like, yes, I agree there will be continuing job loss and I actually do think it will accelerate, but I don't think it's AI. I think they're going to use AI as the excuse, but I think it's just. for the past four years before AI had passed the singularity, what have we seen? We've seen every major tech company has been laying off people and their stock price goes It's not just tech companies. Yeah, all these major companies have been deliberately shutting jobs. Is it controversial to think that's going to continue? Of course it's going to continue. Because for whatever reason, they hired all these people and they realized they could do the job with fewer. didn't have anything to do with AI, but they're gonna be able to keep doing it and they're gonna be able to say AI and it's gonna make their stock price go up and probably if they say we're shutting jobs because of AI because they're gonna look super forward. So I think AI will be like blamed for it. I think that we're actually going to see tremendous social backlash to AI because of this. But I don't currently see any evidence that people are actually getting laid off because of AI directly. Yeah, I think that's like that's a fair point. And that social unrest, I am pessimistic about my view. Like, I do think that we're going to enter some really bleak times because you've seen how anti crypto some people can be. I think we're going to see that same pathology against AI. Yeah, no, and it's funny, too. Like you even today, you listen to like podcast one. on podcasts, he runs a podcast. I listen to you, so I listen when I'm on. But like you listen to podcasts and it's like really interesting to how common it is for people still today to be like, oh, like this company is going to fire all of their like junior coders and then they're going to have to hire back like 10x senior programs to fix all the slop that they create. And it's like Rex Kirshner (01:03:32.28) This isn't like 2022. Like the code that's coming out of LLMs, like I promise you it's better than your most senior developers. just is. The senior developers are saying so. And like sure, like it can do bad stuff and you need to have experience to like figure out like how. fix it little bit like well I mean let's not get it's not swap let's not kid ourselves like it benefits AI companies greatly to circulate these press releases saying that like we're gonna destroy all these jobs because it drives people into a tizzy and they know it yeah he gets the press like earn media yeah so Yeah, well, I think we can keep going for it's so much easier in person too, but we'll cut it off for the audience's attention span and Garrett, good to see you. Rex, let's meet up next year in Denver even if Denver doesn't happen. Yeah, too soon.