Ep78 Slop, Stunts & Millennial Politics

The whole gang's here. And it feels like we're back in 2006. We got Spencer Pratt lighting up LA. We got the relaunch of Digg. How'd we wind up here, guys? I can give you my LA take as I've spent a lot of time thinking about Los Angeles over the course of my life. Yeah, I'd be interested in it and maybe like to tee it up. What I find the most interesting about it's not the politics. It's actually he's leaning into these very timely, very quick- AI videos. It's like obvious AI swap videos and it seems effective. And I wonder if that is, if he's like, basically like carving out a new playbook that we'll just see other folks begin to use. It just feels like it's kind of like even more so than what we've seen these like kind of first-person politicians that are just kind of streaming their life or like creating really cool short, you know, form video. It just feels like he's taking it to another level, just like walking around LA and showing its like elements, which I think is also kind of interesting too. It's a really good point. I'll maybe I'll try and do like a 60 second recap on how we got here, but I think maybe we can just like go around the horn on what you're describing, Aaron. LA is my favorite city in the world. I know some of you guys in here have lived in LA for part of your lives. I will say since, since, you know, over the last 15, 20 years, there's been a ton of problems. I mean, we have homelessness is like a total issue that really got aggressive after COVID. We've had a slew of poor governmental decisions around both the insurance and also just like the, the kind of like the handling of how to prevent fires. As a result, I think California broadly has been hit pretty hard over the last 10 years around some of some of the fires that are that have been popping up. And, you know, some of this stuff is of course natural disasters and there's only so much you can actually prevent around, but there is infrastructure decisions that have been missed both on the insurance side and also on the preventative side that have led to some of this stuff. I think the Palisades fire broadly was kind of like a come to Jesus moment because, you know, when you start messing with wealthy people's property, it's like they start to make some pretty loud noise. And so, yeah, I think, you know, LA has been kind of in this weird transitional states and like stuck in suspension around like, is it going to improve? Is it going to not improve? Is it going to get worse for, for the better part of like a decade at this point? And it's left open this kind of like vacuum for someone to step in and kind of like radicalize, you know, a movement around trying to improve what is like, you know, one of the largest economies in the, in the world. I would say this has maybe dovetails really nicely with some of the stuff that Aaron was bringing up because like we now have the cost to produce content at like an all time low. We have the, you know, very clear track record of someone like a Donald Trump being able to kind of catalyze a social online movement around his beliefs, which then allows somebody who's like outside of the political engine to jump into it very quickly. You've got kind of like a very low hanging fruit around everyone. I mean, everyone wants cleaner streets. Everyone wants faster housing permits. Everyone wants kind of like less, you know, decay and bureaucracy in the government. And so it's like a very easy sell for someone young and who's has somewhat of a celebrity status to be able to kind of like chant the obvious and do it in a way that that's effective and powerful and had his own house burned down and is using that for content. And yeah, I would say maybe just to like bring it full circle to the point you were making Aaron. It's like, it feels like LA has kind of set the perfect preconditions for what is very clearly happening around our political apparatus, which is anybody can run. Anybody can create content. Anybody can catalyze a social movement. And we're no longer require like these checks and balances to like have somebody rise over the course of their career to like be fit to run for office of any kind, because you can kind of bypass those gatekeepers now with some of the new technology networks and some of the new social networks that exist. And I think AI is probably fuel on that fire. And we're seeing that kind of play out in real time. Yeah. And I think, I think video may have been the missing piece. Right. And it's interesting that it's coming out of LA, you know, the cradle of Hollywood. And it's interesting that it's coming out of politics, but, you know, you can just run these short, you know, one, two minute, you know, ads pretty much. And they just seem pretty effective in conveying a message, whether you agree or disagree kind of with that message. So to me that that's like the most interesting piece. And it's almost like they don't, he doesn't seem to care that it like looks sloppy or it's not like high production. It's just about having like a timely response or just kind of staying in the meta. So I thought that that piece is kind of interesting and mixing it with, you know, just like the real time, you know, first person kind of perspective that I think we've seen a bunch of politicians increasingly, you know, increasingly kind of do. It's also interesting just to see the continued power of the reality star in American life. So I think it's a pretty fascinating story. I mean, this isn't even the continued, continuing power of the reality star. This is like a staple of California politics. I mean, this is Ronald Reagan. This is Arnold Schwarzenegger. You know, this, this is actually an ongoing tradition in California. California, it's just adapted, you know, maybe to a different form of star who maybe, you know, doesn't quite have the pedigree of the prior to. But, you know, there's a lot of parallels between when Gray Davis was governor and got recalled and got recalled by Mr. Total Recall on Schwarzenegger. And, you know, this situation as well, like there's a certain institutional complacency, you know, that is really prevalent in California in which, you know, you're kind of dealing with like a state that's very hard to do business with, you know, and has like an entrenched, you know, democratic majority. And then you're pushing that down to the city level. And, you know, they start getting, I don't know, maybe so disconnected or like so interested in preserving their own organizational power that, you know, it leaves them open to this sort of, you know, maverick. Like, hey, I'm a dude and I got greats and other people can relate to me and I'm going to do it in a more timely way than, you know, you're going to go about your business. Los Angeles is also, I don't know, I just feel like as a city, it's a really strange place because it's, I never really got the feeling of like a Los Angeles wide solidarity. You know, like LA to me, it's just so big and it has so many different neighborhoods that like the vitality of the place is often dependent on, you know, a few locations or a few industries or a few scenes. And that like, you know, you can go into the valley and just feel like you're in 1965. And especially when things are slow and a lot of storefronts are shuttered, like it can be a really sort of jarring place because it doesn't have, you know, that same level of, I don't know, like immiscible uniformity. Right. Like, let's just take like the furthest possible extreme, like Paris, like wherever you are in Paris, you know, you're in Paris. When you're in Los Angeles, it really like is just kind of unmooring. And unless you know the place, well, it's hard to know exactly where you are. Yeah. I'm going to say one point on this, which is, and I think you have to, I think if you don't know LA and you come here expecting one thing, it's really difficult to be pleased because like ultimately it is like 10 different cities baked into one. And it's the, it's the knowing of LA and like the different pockets and these different places that makes it a really special place and understanding how to get around and when to get around. And there are these tricks of the trade that you only pick up by living here for a long time. And it is maybe one of the hardest places to visit in the world. So unless you're just going to park yourself in Malibu or just spend your time on the beach. So yeah, anyway, it's a very, it's, it's hard to love for short amounts of time, but easy to love once you're here for a while. Do we think we just, we see more of this though? Is this like, there was also like a new model that Google came out with today. I think it's called Omni. So it just looks like the video stuff is finally hitting an inflection point. I wonder if like, I expected it to come out of, I don't know, like some influencer, not like a political influencer, but I wonder if this is kind of what continues to push that entirely to the mainstream. I mean, the political ad memes, I mean, Trump apparently, allegedly someone on Twitter saying this like internally, he's like obsessed with watching AI videos of himself, which is not surprising, honestly. But what the whole Spencer Pratt phenomenon has had me thinking about is actually just like, and I'm actually writing something about this. So it's kind of timely that we're doing this because I'm literally in the middle of putting together an essay. But I was thinking the AI slot videos really also the other big political ad that kicked off was like the Mom Donnie Ken Griffin apartment one, which is more like a scene, you know, scene cut to Ken Griffin, no AI, but is on the streets of New York pointing at something. And so you have like the right kind of leaning more into the like absurdist AI generated kind of core entertainment. It almost feels a little bit more like LA making movies, whereas New York making scenes. And so I was like kind of thinking about it from the lens of like LA versus New York with their approach to political ads. And it also is a right and left thing where it's like the right seems to be using more of the AI stuff to generate like memetics. And there's a couple of reasons for that I could go into, which I'm thinking about as well. But why do you think? Well, I guess just to tell you, a lot of like that, like 4chan culture and stuff like it's very memetic, like the right has always been better at memes. So the notion that they would use AI to create these like absurdist kind of compelling memes is not surprising that they would immediately graft on to that because there's like a lot of historical precedent for like the right online, like using. Just being good at memetic culture. And I mean, we can go into that separately, but like that's like part of the reason why I think that the AI stuff is more has been used more for like the right politicians. But what I'm curious about this and what I'm like kind of trying to understand is like how much the AI stuff will be used in political advertising or if it kind of just gets bifurcated where like the AI film, movie, ad, political ad will be more of a right coded thing versus like using these more like on the street scenes and UGC kind of influencer content that feels frankly a little bit dated to me becomes more of a left thing because the left hand still is leaning more anti-AI anyway. Like I wonder if a progressive candidate did like an absurdist AI video, if that would be well received or not. Maybe it would be. But there's a lot of thoughts here. I'm kind of juggling a ton of different thoughts together here. But this has been something I've been thinking about as we're talking about the Spencer Pratt thing because it was like AI videos basically. I mean, yes, and the debate like catapulted him. I started paying attention to Spencer Pratt probably a couple weeks ago because of AI videos. Like, I think it's like working. Like it took a nobody reality star who was kind of a villain on the hills into like actual contender, which is really powerful. So Priya, I think you opened up like a huge canvas to talk about. There's like so much going on in here. And I'm going to start with positioning and what role Mamdani occupies versus the role that Pratt occupies. Mamdani won. He is now the machinery. And he's handicapped in a lot of ways because there's only certain levers he can actually get his hands on. And I think he's actually been very tactically smart in his early days in office by focusing on like his presence in the city and all of his actions up until, you know, that wealth tax on second homes has been very Mr. Fix-It oriented. And making life easier, right? Like in a lot of ways, he's actually doing a bunch of the stuff that Pratt is campaigning on, at least in his early days in office. And so we're constantly seeing Mamdani on a city bike. Mamdani walking home. Mamdani talking to guys, you know, in bodegas telling them like, you know what, we think it'll cost you $5,000 to fix your sign. Like that piece on the secondary homes was like the first time he actually stepped outside that bubble. So that's like one part of this going on. And then I think the other thing is once you start getting into the political spectrum, right? Like the Democrats, Democratic Socialists, the DSA, like their attacks or like their focus, right? A lot of it is on how ineffectual the Democratic Party is itself and that like they're basically a uni party, right? With the ruling Republican majority and, you know, distinguishing themselves through the lens of like these guys have lost touch with reality. It's hurting everyday people. Therefore, we have to be very like brass tacks. Like, yes, we have these huge, you know, big ideas around, you know, like the communal embrace of the collective and, you know, redistribution of wealth taxes, right? And all this shit. But where they actually really like connect and are landing on is more around the fact that like the Democratic Party doesn't do shit for anybody. They're completely useless and ineffectual. And I think a lot of like the surprise around Momdani, right, is not that he like wanted to put like a 60% tax on every single one in New York. It's that like, you know what, I can just get shit done. And whereas I think if you're Spencer Pratt, right, all you have to do is go, Jesus Christ, this whole thing is so freaking absurd. We just torched a whole neighborhood. Like two homes have actually gone through the permitting process to, you know, to actually start rebuilding. That if your insurance payment, assuming you actually got to the point where you got an insurance payment, is going to leave you, you know, it's only going to cover maybe 60% of your replacement costs. Because the cost of rebuilding in this environment is so egregious, right? Like Pratt has like this, as an outsider, a different like lens of attack and the huge surface area that like the whole institution and the whole way we're going about doing things like just doesn't apply to reality anymore. And therefore, like AI is a great medium for it. That's interesting. So in a way, like it's like what they're actually messaging. The medium, the medium is more about the message than the medium itself. Like I think I was maybe over indexing on like the medium, like, okay, like the left is more like in person trying to do these like videos versus like, you know, Spencer Pratt's communicating through medium of we'll call it like AI slot videos. More because of the message they're trying to propagate versus the type of person that they are. Is that kind of what you're saying? To some extent, I also think it's just as simple as horses for courses. Like what works best and what plays the strength of each of these people. And like no one knew who the hell Mondani was. I mean, you know, apparently he was, you know, the assemblyman or a state senator or whatever for like Astoria, which is next, you know, next door to me. Like I can walk to Astoria in 15 minutes. You know, I had no freaking clue who he was. And he had to build this whole thing by going door to door and standing on street corners and getting ignored. Whereas like, you know, we basically were, oh, we're back in 2006 and Spencer Pratt's top of the wire, right? Like he had this, you know, pre-built celebrity that just wasn't doing anything. And, you know, he was able to activate it. And then, you know, he didn't have to do that hard work, nor can you really do that hard work in LA, right? It's very different terrain too. So, yeah, I think some of this might just be like, sure, ideological. Some of this might be like very situational. Right. But yes, like can you imagine a DSA candidate trying to run AI slot videos? No, because their base would just reject it as wholly inauthentic. Whereas like this is a guy. But I wonder if that puts them at a structural disadvantage in the long run. Because I think that like timeliness, that timeliness I think is one of the major drivers here. And so not being able, and yeah, you can like hop up, like open your phone, record something reasonably slick. But, and there's a nice folksiness to it. But I think that there's like a sharpness to what you're seeing out of the Platt campaign. That I think is probably going to be a long-term benefit of just leaning into the technology. Sure. But you can also be timely and real. I mean, Ryan Reynolds does it better than anybody, right? Right. So it's not like that, that approach, you know, maybe it is harder. But like, it's not like you can't do that in this day and age. But yes, like you're speaking to different groups of people who have like a different receptivity. Is that even a word? To messaging and the packaging matters. Yeah, it's going to be interesting to kind of. I was trying to do some research to see which got like more impressions and which kind of had a larger effect. And I need to look back at like the article. But I think, I think that that Mamdani video, the Ken Griffin one, we all know which one we're talking about here. I think that got about 50 million views. The Spencer Pratt one, I think in 72 hours. So it's more quick hits. I think it got, you know, something like millions of views in like a seven, you know, 60 to 70 hour time frame. The effect of it at the time, because I mean, Mamdani obviously pissed off a bunch of billionaires with pitotiers in New York City. But like those AI slot videos kind of like gave Spencer Pratt a huge amount in the, I mean, the debate did too. But like he kind of went up and he's been, you know, moving up in polymarket over the last couple of weeks. I think partially as a result of these videos that like have gotten him insane attention. So it took like a nobody into somebody. And then with Mamdani, it just like obviously created huge news waves. So there were like different effects. But I'm just trying to think like which is most effective, honestly. And maybe the answer is just both. And it's circumstantial and it depends on like the audience and whatever. But I do think like some combination of that is probably going to be the future of political media, to say the least. Let's get our L.A. resident back in here and talk about the Pratt hat. Derek, where does the Pratt hat sit in the repertoire of L.A. lids? The Pratt hat? You mean, what do you mean by the Pratt hat? It's got the Dodger. He leans into like the Dodgers. Oh, oh, oh, oh. I see what you're saying. I mean, would I like you? Are you talking about like his own Pratt hat? Like that crappy like L.A. logo that he uses? Yeah, yeah. The Pratt hat. I would you could not catch me dead wearing that thing. And by the way, like huge tactical mistake. Why the hell is it in Giants colors? I don't know. I don't know. The whole thing is a little wacky to me. Call me a purist, dude. But like if you're going to rock an L.A. hat and it's just two letters, it needs to be the Serif Los Angeles logo. So anything less than that is just cheesy to me. That's just my two cents. But yeah, I mean, this whole discussion is fascinating, honestly. And like I do think that this is kind of the lightning rod for the idea, Aaron, to just like circle back to your point. It is it does feel like this the ingredients that are making this possible are just L.A. feels like the first battleground, but it feels like we're about to see this in the midterms. We're going to see this at state level elections and national elections. It just feels like the surface area of how people are going to message political campaigns is about to kind of be kind of like very different over the next 10 years than the previous 10 years. Yeah. And I think you're getting you're like seeing them play off each other, which I think is interesting, too. To me, it's just more signal that like the media landscape's evolving again. I felt like for a while it changed. Right. We moved from like tweets to like reels. And then it prior for the since COVID, it's kind of stayed in that same slop feed. And it feels like it's going to shift again. I don't know to what, but I don't know when you see kind of like the like evolution of playbooks, something's going on. I kind of feel like we're in a setup for for some serious reform movement that, you know, we're we're just like getting independent people out there having success that is going to build towards like a real, you know, real challengers to the two party system. I mean, we we get reform movement, I don't know, every hundred years or so. I feel like maybe we're due for the rise of one of those. Yeah, I can see that. I mean, it feels like to Derek's point at the beginning, right? Like you can just go direct and that probably will have some some impact. I guess we saw like a glimpse of that, Chris, but whether you agree with the politics or not in the UK, right, where they did have a new party kind of emerge from their normal consensus or at least the consensus for the past, I don't know, 50 plus years. I literally called reform. Yeah, I mean, I think there are parallels, you know, between Great Britain and California, right? Like especially around, I don't know, maybe like institutional encroachment and the feeling of you've created a dysfunctional nanny state that is just getting out of hand and we need to push back against it. I think the UK has a much different dimension going on as well. All just, you know, simply around like immigration and like, you know, it opens itself up to like a whole bunch of other layers that, you know, maybe California is isn't exposed to. And California is really just about like, we can't do what we want to do anymore. What the hell is going on here? I don't really want to like jump into the deep end of that pool. But I definitely think like the UK is kind of speed running shit we're going to be seeing here. Feels right. I mean, wouldn't be the first, right? So super interesting. Do you guys also catch those just to pivot a little bit? Those announced massive crypto rounds with ARK and Canton? I thought it was, I thought they were kind of interesting. One, I thought it was interesting that a public company in Circle is purportedly raising more capital from venture. And two, like I just still don't understand anything about Canton and, you know, why there's so much interest in that. That was confusing to me. I saw that online and I'm like, why? That's what I'm trying to figure out. Well, the Circle thing, like I just saw a little bit of it today, but it's a token sale, which I thought was a little interesting. And then the size of it, how are they going about doing that ahead of Clarity? Question mark. Sorry, I don't know if anything else there. I guess, I mean, I guess it shows the confidence that that will go through. Like you're curious why they're able to get away selling, doing whatever, a token sale publicly like this before Clarity is announced? Yeah. Yeah. I think they joined a long lineage of other companies that have done the exact same thing over the last five or six years. So I don't know if it could be confidence that like we're moving into a, it could be a signal that we're moving into a different paradigm around how we treat these tokens as either part or in part, kind of like next to or at different times part of the securities regime. The truth is, is like these things do need, I mean, tokens do need to be decentralized ultimately. And maybe they feel confident in their path to have the vast majority of the network be owned by the public once that thing launches. So hard, hard to know entirely what that looks like. Yeah. Pre, this is the, this is for circle, right? This is the ARK product. It's ARK and, and the Canton, Canton network. Canton is a little different to me since that network already exists. But at least for ARK, it's like, you know, a lot of the, a lot of the times these tokens get sold are part of securities transactions. And because the network's not live yet and not trading, it's like you do have some latitude around, you know, these, these things are largely part of like the securities regime before they start trading. And so they, they're, I guess they, they feel comfortable with that approach. Canton's a little different. Like Canton actually has a network going. And so when they, they do these deals, I think structuring them can say a lot about what kind of risk they think that they're availing themselves to. But I think broadly speaking, like it does feel like a new day. I think people are willing to take on more risk and hope some of these messy questions around how tokens can be valuable, get sorted and answered over the next couple of months, both with clarity. Cause I don't know how much within clarity we're going to get around some of these topics, but hopefully after clarity. Is that your sense to you, Aaron? Yeah. I think that feels about right. Yeah. I mean, it feels like at least for these, these base, you know, networks, they, it feels like that was contemplated, at least in the last version of, of clarity that I saw. So that didn't seem like the, the bigger issue. I think, you know, for circle, sure. Yeah. Like they want some more support for it, but it's interesting just to kind of see them raise capital, not from the public, you know, public markets where they're permitted to raise capital, but from here. And two, they are a publicly traded company and they're going to have like a, looks like a tradable token. Yeah. I mean, it's like, I don't know if we've had that yet. Right. Like there's been rumors that Coinbase may do that, but that, that hasn't been announced. So I don't know if that kind of foretells what we may see from Coinbase too on that front. For Canton, it's just like, like what? Like, you know, like, like if we zoom out five years, like it's 2030 or is anybody going to be talking about that project? I just would be surprised. But what do I know? I hope no one was talking about it before and they got acquired. I was like, what? I haven't heard that word in years. I was like very perplexed. The other one that kind of came out of nowhere, didn't like bullish acquire like traditional broker dealer or something like that? Like, did you guys see that? Again, like I'm kind of behind on my, my news, but like. As you should be. Yeah. I mean, but the bullish public company, because obviously they're, they're, they acquired Aquinity, which is a top transfer agent to bridge traditional markets with tokenization. I don't know if you guys saw that one, but that one felt pretty big too. I mean, I think it was like a multi-billion acquisition or something. Yeah. I think it, I think it was. I can't remember what the details were, but it was like one of the largest deals of this type, I think in crypto. So yeah, it does feel like a lot of folks are all kind of converging to the same surface area, which is like more assets on blockchains, getting licenses, getting approvals, possibly launching tokens for these networks. I know Kraken has the, their blockchain coming up and that Coinbase has base. Robinhood has their, their own circle. So ultimately, like a lot of these public companies will probably have tokens on balance sheet support some of them. Yeah. Underwriting like the economic activity that happens on these ledgers. So I think there's a, there's a, yeah, there's definitely a story here. It wouldn't surprise me in six to 12 months to see a bunch of public companies with tokens on balance sheet. Putting my AFL CIO bag holder hat on. This is extremely bullish for XRP. Ripple has been leading the way here. I think you're probably right on that. Actually, Chris, I think that's a good call out. It's interesting how I think maybe they failed in terms of product market fit and won bigly in regulatory navigation and market structuring. Yeah. So what do you think happens here, Chris? Like they're going to, you think Ripple Inc. goes public and then they just compete in the same, the same vein or like what, what do you see for them? I don't know. I don't pay attention to Ripple. A friend of mine was working there for a while. And so I, I heard a lot of stuff secondhand, you know, but like they were doing all of this a year, two years ago, you know, as soon as they won their court case. I mean, you know, they have a massive balance sheet. They had, you know, their product was kind of dead in the water. But, you know, they had this institutional heft to them. And so they went on an acquisition spree and, you know, are starting to work their way into like the broker dealer stack, into the custody stack. Right. Consensus goes to Miami. Consensus goes to the strip clubs. Consensus brings in trad fi suits. What was people's feelings about consensus this year? I didn't, I mean, I didn't go. I, I happened to be in Miami for another kind of adjacent event towards the end of the week. I saw like the, I guess the end of it, just, just passively. Like there was some like suey banners that were like flying in downtown, I guess Miami beach or one of the strips of Miami beach. Yeah. I mean, I saw like some glimmers of that from the timeline. Not surprising. Right. Like if you look at where the interest is, it's definitely more institutional now. I feel like consensus always had a little bit of an institutional like veneer to it. So it feels about right to me. I mean, all the, like the shenanigans, like, I don't, I don't know. That's not going to wet my beak. The guy peeing, the guy peeing in the club and then slipping on his pee. Wait, what? I didn't see that. I, I, I had to see that video like two or three different times until I figured out what was actually happening. It was an insane video, Aaron. How did you not see it? I just was not. Consensus. I almost feel like you should watch it right now. No, I don't want to watch this. I know. I like. All right. Link me up. How did you avoid seeing this? I really curate my timeline, guys. Like if I see anything I don't like, I just mute the heck out of it. It's like, I don't know how you missed it. My takeaways from consensus via the timeline is only three things happened to consensus. One, people did BD at 11, five nights in a row till 6am. Two, while at 11, people got so trashed that they peed on the bar, fell over and landed in it. And three, crypto is so broke and skeezy that people are taking strippers' monies off the stage and pocketing it. What? I missed that, too. I saw that one, too. The video of that guy. Someone somehow zoomed in on a guy taking pictures off, taking money off the stripper stage, being like, man, he's down bad. He's stealing from these poor people who are doing their job. Isn't it amazing that this is the industry that we fight and bleed for? I know. I actually kind of, it makes me cringe because this is the stuff that goes viral and I'm like, ugh. Because it kind of just basically corroborates what everyone else thinks already. And it's like, damn, I know so many normal, smart people in this industry. I really wish that this wasn't what came out of that. I'm watching this video. Hey, amazing. Who's recording this? How did they whip out their phone to capture that moment? I know. I was wondering that, too. You know, like, you know, 100% pretty. I guess if we can't deliver on utility, we're just going to deliver on embarrassment and memes. Is that the state of things? I'd much rather read about some more useful things that we can do. Well, the good news, Aaron, is when crypto gets too cringe, there's another frontier technology that's just as cringe and just as unloved that you can give your attention to. AI? Is that what you're saying? Yes. Yes. Oh, God. I know. I just want, I mean, yeah, it's all bad. Well, here, let's swing over to AI for a second and talk about stuff going on. Because Anthropic is doing everything at once. I guess we could talk about the structuring stuff in which they're trying to reign in all this, like, second, third, fourth hand dealing and SPVs around their equity. You know, so that's one thing. Another thing is their strategy of, like, have good models and compute will find you seems to be very busy at the moment. And then, like, you know, your man there, Elon Musk is both now like a data center operator for Anthropic and in court battling it out against Dario's most hated mutual enemy, Sam A. Like, lots happening over there. Yeah. I mean, I was worried that they wouldn't find the compute, to be honest. I was surprised that that deal came together so quickly. And it felt pretty shrewd on Elon's part just because he has excess compute. And the bigger question I have is just, like, what does that say about Grok? Like, is he, you know, not going to be pressing as hard on that and he's just going to, you know, be one of the major purveyors of compute because that's really where their skill set is? Or does this, like, be, is this used as a flywheel to, like, help them fund more training of Grok? Like, to me, those are, like, the biggest questions. And I guess it's good to see them, to see our tech oligarchs, like, play nice with each other. I guess this is an instance where that happened. Yeah, I think there's a little bit of 4D chess going on and there's a lot of making up for, you know, rush decisions that, you know, are aligning in ways that Elon's particularly good at pulling off, right? Colossus 1 is a bad training environment because it works on multiple chip types and cannot, like, all training, training is gated, right? In terms of, like, work finishes here, has to echo to there, has to echo to there. And if you have all these different chip classes working at different speeds, right, you're really only getting, like, I believe, like, I don't know, like a 10, 11%, like, utilization of that whole Colossus 1 core. And so it just wasn't really working well for training, but it can work really well for inference. And so that was a bit of maybe a miscalculation that, you know, ends up working out fine as a different business line. It also really helps ahead of the SpaceX IPO because now you've turned a cost center into a revenue center. I've been kind of thinking of, like, XAI as, like, gateway computer in a lot of ways in which it's just, like, a hardware reseller. You know, in this case, it's, like, a data center reseller because, yeah, they, you know, had a hard time with their model training because they didn't staff correctly. They didn't give it the attention. Like, that's a very, you know, human-oriented personality management business because it's such a small group of people. And there was clearly, like, a lot of, like, you know, either they didn't get the right people in or, you know, their defections reached a critical mass in which keeping the rest of the team didn't make sense. But, like, that whole team got thrown overboard. And now they have this option call on Cursor where basically they bought, you know, at a minimum $10 billion worth of, like, coding training data and given Cursor, like, free runway on Colossus 2 to go to work. And if that pans out, then they can just buy Cursor for $60 billion. And so, you know, Elon is, like, very heroic in, like, the classical sense of the word in which, like, he takes on great tasks or great, you know, great quests, gets himself in a lot of trouble in trying to perform these quests, and then perseveres. You know what I mean? And this kind of is, like, Elon on this, like, crazy hero's journey involving mountains of GPUs and personalities. Yeah, I mean, and I think if the news reports are right, even the Cursor team is already starting to work with the Grok team, Chris. I think I read that there was, like, some meeting where, like, the Cursor engineers were being brought in to talk to the Grok folks. So maybe that's a piece of it, too. Not just what I imagine is very valuable programming-based training data that's hard to get. But number two, maybe that team just is cooking and cooking more so than whatever Elon was able to assemble before. Yeah, I mean, Grok couldn't clear the waifu barrier, so maybe Fresh Blood and a different POV will, you know, get them over the hump. Yeah, completely. The other thing, just in terms of AI news, it looks like the interest in OpenClaw has been diving down, like, pretty significantly. At least there was a tweet that was circulating that, assuming it's true. But the interest in OpenClaw kind of peaked in mid-March, and it's way down from where it was initially. So I wonder if, like, that little period is going to get a little bit more quiet as we see kind of people run their course with OpenClaw and maybe the, like, adjacent products, like the Hermes agent. You think, like, the Hermes agent, you think those are, like, I can't tell if all of these, like, little AI projects are, like, very, like, feel very, like, crypto protocol coded to me, where it's, like, people are super interested in it, super crazy hype online, and then, like, no one uses it in three months. But I can't tell if this is just, like, science. I feel kind of like DeFi. Yeah. Like, to me, it's more like DeFi. Like, you know, like, there was real innovation that's going on there, and I think that there is here. I think that people wanted it to do the world, and it can only do some narrow things at this point. And then, you know, as more and more people begin to build and use these types of tools, like an ecosystem will develop related to it. But I think for folks that were around, like, DeFi very early, there was, like, you know, one to five really amazing projects. And then there was a lot of noise kind of outside of that. And then, you know, we kind of saw evolutions in a couple waves. So I wonder if that's what we're going to get with, like, the AI personal server assistant. I don't know exactly how people want to characterize it, but it feels like it could be similar to that. I, for one, am thrilled that the 100 parallel agent maxer guys are not the loudest voices in the room. You know, I do think, like, there's a lot of similarities in which, like, there's people who are attracted to novel things in action and want to be there, but then they don't really have anything to add to it. And once they realize, like, you know, the agent itself still requires input, direction, taste, like, you know, that, like, you need to bring something to the table because all this is a tool at the end of the day. You know, those people now have to, like, find something else to hop on. But then there's people for whom, yes, like, this unlocks something. It's just always a much smaller group of people. You know, I mean, going back to DeFi, there was a million copypasta projects. There were really, you know, only a handful of, like, truly innovative protocols. Yeah, and I guess maybe in a couple of years, there'll be, like, some AI event where some dude pisses himself and slips at the end of the day. Is that what I'm hearing here? As long as there's a robot taking money off a stripper's table. That'll be the scenes from 2030. It's just basically that. Oh, man. What are we in for? The only thing we can be sure of, Aaron, is that we will be embarrassed by it. We will regret lending our time to something that has turned so cringe. Do you think? I feel like the AI stuff doesn't have as much of a cringe veneer as crypto. If I had to, like, if I had to line it up line by line. I hope not. But I don't know. I mean, eventually there'll be a pullback here, right? And I think you'll see all the exuberance, like, you know, glosses over all the problems until it doesn't, right? Sure. I'll give you guys my quick take on the agent stuff. I actually think, listen, it's really difficult. So when OpenCloud launched, it basically had, I mean, it had command over the entire market because the market didn't exist yet. It was the only game in town. And what we have found is there's a lot of, I think it gave people the core insight that, like, hey, instead of single shotting inference requests via the cloud and kind of like using the human, you know, mind as like the gating function to building something, we could actually start spitting up these agents and as long as we harness them correctly and automated them in the right way, we could spin up five and 10 and 15 and 20. And it created a new model for by which, like, people could make things. And I think my general take on this is looking at, like, the interests or, like, the Google trend reports on OpenClaw. To me, it was always inevitably going to be down only because, like, they pioneered a category by which they were the only name. And now we're finding, like, people are building their own harnesses. People are using Noose Research's Hermes agent, which is the fastest growing agent in the world. I mean, it's like OpenRouter mentioned over the weekend that they just in sheer tokens have already overpassed OpenClaw. But then you've got, like, TinyClaw and Paperclip and ZeroClaw and Pico and Nano and also the proprietary stuff that people are creating. The data that I'm looking at is, like, this concept of agent orchestration via the harness is, like, it's basically just up only. But I do agree largely, like, these things are going to get tweaked and rebaked and there will be new products that version up, you know, the core behavior. I think my general thought, though, is, like, I'm pretty bearish on OpenClaw. I think that was a vibe-coded first pass. And to me, I don't know if my takeaway here is, like, OpenClaw losing market share being correlated with, like, the lack of interest in using agent harnesses. I feel like I hear that and maybe it's first mover. I mean, I think Peter was a pretty experienced developer. Someone was telling me this past week that, you know, he was from previous software projects, like, already independently wealthy. I don't know if that's true or not. So I do think that he's a talented person. It feels like some of it's also, like, diffusion from new tools from just Cloud itself or Anthropic itself that you can just do using Cloud code or some adjacent things that are available via code work. And I wonder if that's just going to, like, take a little bit of air out of that balloon, Derek? Yeah, I have my theory about this. I do think it has, for sure. I think most people were using, for people who love the best intelligence, and right now the best intelligence is coming out of OpenAI. Codex 5.5 is coming out of Anthropic models. And so, like, the vast majority of agentic harness stuff for people who want to, whatever, be on the bleeding edge can be usurped by people who just want to use the harnesses coming out of those closed ecosystems. My general view on this is, like, your harness is probably the thing that needs, or at least there will be massive benefits to kind of, like, a third-party agnostic harness. I think the telemetry that you're able to get, whatever, the value that you want to keep private, to me, and the fact that you don't necessarily, as models become commoditized or there are fine-tuned models you want to use or smaller models or local models when you want to run inference requests that are cheaper, just using an OpenAI harness to use OpenAI products feels like not where we're going. It feels like we're entering a world where, like, modularity and choice is going to play more and more of a role, especially as, like, we're seeing these things start to, at least the delta between, like, the open-source models and the performance of the bleeding-edge models start to close. Like, I think before it was, you know, 18 months, and then it was 12 months away, and then it was 8 months, and now we're, like, 3 months away. I mean, this last DeepSeek 4 model was, like, pretty freaking good. And so I do think the harness is something that I'm looking at and thinking this thing needs to be not a closed ecosystem. You want to use the harness on multiple models. You want to use it on open-source models. You want to use it on fine-tuned models, on proprietary models. And I just find it hard to believe that, like, everyone is going to want to use clod models when they grow up. Yeah, I mean, we agree with that. I mean, that's what, in part, when we designed kind of our research agent, it in chat. It was with that in mind. I still, the open versus closed, I think that's going to, I bet it follows a similar path. Like, I think most people, for this to go hit scale, they don't want to tinker. They don't want to have to maintain it. They don't want to have to become, like, you know, the equivalent of, like, a Linux engineer to, or a Linux user to, you know, go on with their day. My sense is you'll probably get, like, a pretty powerful, like, closed version of this, too. Yeah. I do think that that's the bet that these guys are making right now. Like, I've played around with the Anthropic and the OpenAI ones, and they're good. They're really strong, and I think they're going to only get better. Yeah, completely. Well, we lost one of our four. So maybe, Priti, you want to intro us? Yes, I'm happy to. Welcome to Net Society. Today, it's me, Aaron, Derek, and Chris talking all things crypto, AI, tech, culture, and more. Just as a quick reminder, these thoughts and opinions are not of our employer, and this is not financial advice.