Loading summary
Transcript81 lines
- [00:00]
Michael
This is what's going to happen because to the point of these agents, you can effectively tie in the permissions eventually you know what their APIs are hitting. So if you need like fable for like higher level work versus if you're going to hit Kimi via open router to go into lower level. But these agents will effectively start to gain context and knowledge and then they'll go out and do things and it'll compound. But the angle longer term I see is that these agents in these applications won't have the data silo of permissions and credentials that Microsoft slack Microsoft Teams Google sheets have. So they will just hop around different parts of the web because of this open protocol. And that's when you see the insanity. Because we can talk about productivity, which is already insane if you can like share with your team and add permissions and give context. But like what eventually happens when all of that is propagating throughout the whole Internet.
- [00:52]
Liam
It all comes down to computers communicating.
- [00:56]
Michael
The information superhighway can be a confusing mix of on ramps and off ramps. Bitcoin is worthless. Artificial gold.
- [01:06]
Brian
Is it still rat poison?
- [01:07]
Michael
Probably rat poison squared.
- [01:09]
Liam
We need to get into the world of okay, this is actually foundational technology.
- [01:15]
Michael
What the Internet of money does is it creates a single network which can do a microtransaction to a giga transaction. The Internet is going to be one
- [01:28]
Brian
of the major forces for reducing the roll of government.
- [01:33]
Michael
The one thing that's missing but that will soon be developed is a reliable E cash.
- [01:41]
Brian
Alrighty boys. Welcome back to another episode of Final Settlement. Today is Monday, July 27, 11:47am Eastern Standard Time. We've got another big show, boys. The Internet is buzzing. We're buzzing. There's lots to get to open versus closed. Guys, how are we doing?
- [02:03]
Michael
We're back. The guys got me all fired up because we were talking about just God knows what we're gonna talk about here, but tying into what's going on. Side note, just. I didn't think I was gonna say this, but keeping the rambling going. Brian, do you feel in the air like fall is just gonna be right around the corner? Football is gonna start.
- [02:21]
Brian
Yeah, summer's over. Summer, effectively.
- [02:24]
Michael
It's. It's gonna be nice. It's gonna. There's gonna be spicy. I don't know. I think with this AI stuff which will tie into or start with it's kind of bullish on just traction. If you like, if you like the juice, if you like the activity. Like this activity is not gonna stop Anytime soon where you can. You couldn't have said that before? I think everything's been, I don't know, it's even hard to go back to what was a pre AI world when you think about how fast this stuff's moved the past, like one to two years.
- [02:54]
Brian
Yeah, 100%. There's a lot going on. I guess the interesting part about all of it to me is that there are underlying synergies and parallels to the bitcoin space. So despite the price being down and being in this bare market, a ton is happening under the surface, more so than any bear market I can remember. In terms of things being built on primitives that are, as we'll talk about, open, free to use, interoperable. All of these things are beginning to compound in real time. And this is something that people, you know, particularly on the bitcoin side, have talked about for, you know, really since 2022, 2023, when the AI stuff started to percolate and then you had NOSTR come online and all of these things. So we'll get into it, but we're going to start with Jensen Huang's first ever tweet on the Twitter platform now known as X. He says in my first post, I'm sharing a letter. Nvidia signed on. Why open models matter. AI will transform every industry, power every company, and be built by every country. Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty. The world needs both frontier closed models and frontier open models. And so this is really on the back of something we've talked about for the past couple weeks of this idea of open versus closed in the AI world. You have OpenAI and anthropic on more of the closed side, you have these Chinese models on more of the open side. And now you have effectively a consortium of companies across the world saying, yeah, no, we're in favor of open models. It helps accelerate innovation. It allows these things to compound. And being closed really helps no one except for those that are closed. So there's this natural prisoner's alumni and game theory thing that we've talked about. But, Michael, maybe I'll kick this to you first. There's a bunch of other things that we'll get to stemming from this. But what were your thoughts on, on Jensen's post, the letter he put out and this consortium?
- [04:50]
Michael
I almost want you guys to go so that I can roast, roast the other, take the other side. Because whatever, if I'm going to say, I'm gonna say, I know you guys are gonna take the other, the opposite, but you're not gonna go as hard. But I'll go first. I think, I think that this is all a very like life. It's not black or white. And I think everyone is intending or trying to talk their own book. And talking your own book can mean multiple things. I think what most people pontificate in this industry is you're just talking the book of business, which is relevant. But I also think people are talking the book of just understanding where this goes and trying to slow things down. The tweet that I keep coming back to and I'll tie it back into the, the open weights because I think there's relevance. But those takes are. You guys are gonna share them and you've heard them like whether it's the browser or technology and open source like that will naturally proliferate. Naturally needs to. But I, I honestly think. And what people aren't talking about is the notion and it's this, this tweet really stuck out to me. So I think Rune, I believe he's at one of the labs. But either way, Dan Robinson, who's really tied into the research community over at Paradigm, the tweet is if we could coordinate a global capability slowdown today, I would likely press the magic button. And then Dan, you know, quote tweeted. This is what most people close to the research frontier currently say when you ask them. I think this is a notable fact, and this is the stuff that you see across the board when we've been talking about the whistleblowers and other just researchers that have left is that this stuff is only accelerating. And then you can go back into just the order of operations around why they focused on, you know, code and research and automating, all of that. And so I think personally that's really where this stuff stems from. And a lot of the conversation is how do we slow this down? And we've already talked about it's gray and there's game theory and you cannot be able to stop China, you're not gonna be able to stop others. So there's a race. But that's where I think a lot of this natural conversation has come from, is ultimately how do we like stop this? How do we stop distillation, how do we stop access? Even if we can, maybe it's a catch up. I don't necessarily know what the long term or how that would play out, but I think that's where it comes from. Less around the business model and the thing we were talking about before the podcast is, you know, like, we have to be honest. And it reminds me of like the Elon Musk conversation where everyone. And you don't hear it as much now, but Elon used to quote about Doge and, and all these different things that could be a unit. And it's like this guy deeply understanding physics, doing all of these things. The wealthiest man in the world doesn't understand proof of work and you can't print money on trees. And like how economic coordination happens. It's like, I don't believe they can both be true. It's either one or the other. And, and similar here. It's like, I don't think this guy built the fastest growing company on the planet Earth, right? In the past nine months, he's gone from 10 billion to $70 billion, leads this firm. It's no like small cakewalk to lead a firm. You have to have real empathy, real emotional intelligence to lead these amount of people, recruit them. And then he just deeply doesn't understand the optics. If he's just going to try to get the, you know, pull up the ladder or whatever on the regulatory construct. So he, him and an open AI win. Like, I just think that there's more to it. And the common take is like, oh, they just want to pull up the ladder and they want to be at the frontier. And obviously, like, that could be a component of it, but I just think that there's a lot more. And the best angle I can come from is that they have a peek into the future, whether it's what they use and also the research and they understand the exponentials and how fast this is growing. And that's a component of why a lot of this is coming out. And obviously on the other side of it, they all, like, there's a component of money, economics and natural. Like, I think the, the hard part for all of this is the mental models don't exist. Like, we don't have any mental models around. Like, we have industrial revolution, we have fiber optics, we have Internet, we have mobile wave, but we don't have anything like infuses intelligence at this level and then effectively can self replicate into an exponential form. And so that's really where the, I think like crossing the, the friction and the misalignment stems from.
- [08:51]
Brian
Yeah, I mean, listen, I think it's a fair take in the sense that if you're looking at how Dario has talked about the advancement and the acceleration, if you're just taking what he's saying at face value, then he is worried about what could happen here? And then it's. It tracks that basically everyone outside of those walls would say, well, they're ahead then. So in order to catch up, we need to embrace open source. These things that will accelerate and compound. And so if you're just, you know, thinking about it, like through, you know, Occam's razor, like, what is most likely to be true, that could just be the default of like, yeah, he's actually really worried about it and is trying to slow things down. It's not necessarily a regulatory capture sort of play. I did want to pull this up. This was Jensen this morning retweeting the security alliance sort of consortium saying attackers have frontier AI defenders need a frontier AI ecosystem. The best open and closed models force multiplied by a global community. During the hugging face incident, which we talked a bit about last week, closed AI blocked essential forensics. An open weight frontier model helped contain the intrusion. That's why we created the open secure AI alliance. Liam, what are your thoughts on all this?
- [09:57]
Liam
Yeah, I think that while a lot of the people at the AI research labs do say that they wish they could slow things down, they mostly continue to stay at these labs and push forward these models. And so I think that, you know, similar to nuclear weapons, I think almost everybody would say, all else being equal, nobody would want any sort of this technology to really be diffused. But as long as it's out there, there's going to be mutually assured destruction or kind of defense by offense by developing them yourselves. And I think that there's something similar to be said here with having both closed and open frontier models. And I think that it's valuable for a number of different reasons, having sovereignty over how you want to use these systems to defend against any rogue closed AI or AI that you don't have complete control over, being able to have private information, which will become more and more important as the agents take over more and more tasks, both from a privacy perspective, from a personal level as well as a. As well as a company. And then just making sure that one company isn't a. Doesn't have to be a benevolent dictator. As these tools become more and more valuable, whether it's Chinese companies or anthropic, any other companies out there that develop either closed or open models.
- [11:23]
Michael
Yeah, one of the. And this is really where it comes down to the gray and there's like, it's less right or wrong. It's just the status of what's happening because obviously others need to use it. Lower token cost. You don't want to bifurcate or segregate, like the option options for enterprises, individuals to have access to this. But I think it's not perfect apples to apples. But it's like the version if we were in a similar shoe, like at Onrampur, and we don't want more money printing because more money printing will benefit our business. But like, let's say in this analogy, we had a certain segment of the market, we were, you know, making crazy amounts of money. And so we were talking about, like, the negative of money printing. And so we're trying, like, even though that the printing increases, we benefit and the price increases that we would benefit. We're trying to, like, say, look, I don't want this to accelerate at this speed because that diffusion and, like, how it play out with the. With the angle where it breaks down is ultimately that we are not participating in the money. It would be like the last part of the analogy is like, if we were actually. And so then the other side of the game theory is ultimately, to Liam's point, it's something that gets deeply misunderstood or not talked about is that these people don't leave and they go to work at these firms because the way it's internal at the culture and it's internal at the executive level is Darrow doesn't want Elon to have it. Elon doesn't want to have Sam to have it. X doesn't want China to have it. And so everyone continues down the path. And that's why it goes back to that Robinson tweet about if people could just pause or put the canary back to coal mine or whatever, it's like they would rather do that. They can't. And so I think that's like, also so now if he has. He's a fiduciary, he also wants to make sure that these things in his mind don't happen. He has to continue progressing at the speed that they are. And so it just keeps this going. And so I don't. We can continue here. But I think one of the main things for the rest of the conversation that I've wanted to share is this. I honestly don't think anybody talking about, including us, including myself, understand fully, like, the ramifications of how crazy this stuff is. Like, people talk about it and it's out there, but I don't think we're truly prepared going to the money side. I think it would be similar to, like, hyperinflation that I don't know if we'll see hyperinflation in the United States. But it's this angle that people talk about it, but it's until you're in there and you're like, you know, you hear the stories of getting your, you know, paycheck in the morning, spending it before lunch because you were buying half of that cost. Like those things you can never really viscerally feel you can talk about until you actually feel them. They're fundamentally different and I think that's similar. What's happening here is like this level of how fast things are moving and what's capable and where it'll. We don't, we keep, we don't have a grasp on it. And you can kind of feel that in a lot of the notes we're going to talk about and how fast things are moving that these things are happening at the same time that money's breaking down. And yeah, it's just, it's going to be a wild ride and everyone has to buckle up. Welcome back everyone to another episode of Final Settlement. This was an exciting one. Lots of buzz. No pun intended or pun intended. The recent Jack announcement square blocks, incredibly exciting. I would encourage you guys to make it to the end or at least look for that segment. A lot of things we've been looking at wanted to give a quick shout out. We're going to be doing some roundtable series and in person events to end the year. Austin, Minneapolis, Philadelphia, New York, New York will probably be next month, including my email down here. So if anybody wants to attend, learn more about what we're doing, interested in what we're doing at Early Writers, I'd encourage you to reach out. Never been a more exciting time. Also terrifying time depending on where you sit. But we're going to be expanding the business. We're doing a lot on the on ramp side, Early Writers onboarding clients. So if anything that we're doing is interesting, you've been having ideas on businesses you want to fund, what we're up to at Early Writers, what's going on at Buzz at Gentec Finance, or you're looking to become an honorary client or learn more about what we do, I'd encourage you to reach out to me personally. Would love to chat or if you just want to attend one of our events, hope you enjoy the show. Let us know what you think in the comments.
- [15:34]
Brian
Yeah, no, that's well said. And before we move on, just a couple other things to hit on here. I think, you know, if you're thinking about the perspective of all of these firms that have entered this consortium, coordinated fairly rapidly to be on the side of open source. I'd say it makes sense from also just like an economic implications perspective. So we saw last week or the week before that Microsoft would be using Kimi, the Chinese open model, to power Copilot because there are economic implications for businesses in the United States who use cheaper models. If you know OpenAI and Anthropic are 50 to 100x more expensive than something they can get open source, there's natural implications there. And so if the US Government were to do anything to really try, and it would be trying, it wouldn't be a super effective route to go. But trying to curb people from using these things that are more open and cheaper, it would be disastrous for I would say the stock market, which is a proxy function for the United States economy at this point. And so I think that's the other sort of incentive to look at when we're thinking about all these different companies. How they're talking about it is because there's natural incentives. And then the only other thing to bring up was this was also this morning that Nvidia is in talks to guarantee 250 billion in financing for OpenAI so they can lease a 10 gigawatt center in Ohio. And this sort of compounds and continues to grow this, this debt jubilee that we've been talking about in terms of all of this capex spend and also the sort of incestuous nature of all these different firms who are in all these deals. So another example on that front, I
- [17:11]
Liam
just on, on your first point too about Microsoft running K3, I think that it's important to note that it's not just for lower cost. It's also the ability to post train on open models and have thus lower latency and better results for any specialized tasks. And that's why, as you know, things become more automated and just focused on specific things. In this instance, running Copilot, it is more, it is just better to run off of open models that you can post train and have your own evaluations for specific tasks rather than generalized models that are applicable to everything.
- [17:52]
Michael
Can you go back to that chart with all the images of the people supporting open source?
- [17:58]
Brian
Yes, give me a second.
- [18:01]
Michael
So I think the chart Brian's gonna pull up again, it goes back to everyone's incentive and awareness. Like if you think about it, I don't think it's without question like these firms are not frontier labs. They're not a. Like it's very concentrated. There's a few people probably at these firms like Palantir, as An example point being is it makes complete sense why these firms would not want to be holding or have to. They want an open platform so they can post, train, get lower cost, et cetera, et cetera. But the thing that is fascinating is I can guarantee you in 10 years there is half of these firms will not exist in the future because of what the Frontier Labs are working on. And so like that's the, the, the friction that exists. It's like they're saying slow it down. Not even because of like, I mean it could be the model, it could be portion of the business model, but there's a component of like it won't matter if these firms have it or not because of what like you think about right now. We talk about, we'll talk about Buzz and Slack, but I think about like HubSpot and a lot of these like SaaS platforms that are so expensive in the lock in and now that doesn't exist because you can effectively pull all the data and platform rebuild it, build it into other solutions. Like that is going to become even more. If we believe that this even moves any more north and up into the right as far as intelligence goes, we can already do that today. What's going to happen six months from now? What's going to happen like 18 months from now? So I think it's just like one of those versions of be careful what you wish for where they're saying like look, we should slow this down. And to your point, you can't ever stop it. But I think the angle is how do you actually slow down the. It's not even distillation. It's like us accessing overseas models. It's both sides. And anyway I just think that's a component that's not talked about, the natural version. And I'm always kind of like skeptical when the natural thing, everything you listen to is we just need open source. We need it. Which is like obvious. But I just think that there's more to it than.
- [19:56]
Brian
Yeah, no, that that's totally fair. And the other thing worth mentioning is like if you really wanted to stop people from using open source Chinese models, you wouldn't necessarily ban them. You would just have OpenAI and Anthropic enforce some kind of KYC, which they don't currently do in order to try and curb the distillation. Like that would be sort of the more upstream approach to stopping what is happening downstream. If that makes sense.
- [20:21]
Liam
Yeah.
- [20:21]
Michael
And the other thing I was going to say that reminds me of like Bitcoin is you Know, this may be controversial but like the government has slowed down Bitcoin. It slowed down Bitcoin with regulation, with ETFs, et cetera, et cetera. And if you want to hyperinflation, you would educate the market on like what Bitcoin is, what is inflation, what is real versus nominal. There's all these things happening that slow down Bitcoin. Now, inevitably it's an open source solution to money that will win out. But the government has no shortage of ways. We've seen this, you know, Operation Choke Point, et cetera, you know, crypto and stable coins. It's not even the poly markets of the world we talked about. Prediction markets have existed for a while. It's not until the past, you know, 18, 24 months they've blown up because they've been allowed to blow up. And so I just think that, you know, there is an angle that you could slow this stuff down. They're making the intentional effort.
- [21:08]
Liam
They're.
- [21:09]
Michael
They don't want to.
- [21:11]
Liam
Yeah.
- [21:12]
Brian
Okay, moving on to what you alluded to. And it's a natural segue in the sense that we're talking about open protocols, how they compound upon each other. And this was the other big news item from last week. Jack Dorsey releasing what is called Buzz. And I'll just read a bit of this here. Yesterday we released Buzz. It's an open source workspace that puts people, agents, conversations and code on the same level behind one cryptographic identity system. We built it to reduce our dependency on Slack and GitHub and we're sharing it so anyone can do the same. And so I think, you know, I'll hand this to you Michael, because I know you're super jazzed up and buzzed about this and I think we all are. I mean, I think this is pretty exciting to see on a couple different fronts. One is, you know, I think how some people perceived this last week was like, okay, this is just an open source Slack competitor. I think it goes way, way deeper and is more significant than that. And then the other thing, probably most interesting to me, at least personally, is like Jack is just shipping things that are built on the same primitives as Bitcoin. So public private key pairs. He became very enamored with nostr back in 2023 and now he is shipping things that use those same substrates and architectures bitchat, which we'll also talk about. And now Buzz. So Mike, I'm going to hand this to you because I know, I think you were playing with it over the weekend and let's just talk through a lot of the implications here and there's
- [22:34]
Liam
some other links we'll pull up.
- [22:36]
Michael
Yeah, there's a lot. I think we'll be covering more. I'm going to try to be succinct, as succinct as I can because there's too much, I think implications. I think one core aspect like going from the meta version is this guy is just insane. Jack, when you think about the prescience and the I think he's the only CEO to what does it lead to publicly traded companies, but then also at the valuations that they were and to recognize what Twitter could be, what it still is today. And then ultimately square and blocks and where they're going and then obviously one of the only CEOs really in front from a publicly traded perspective, deeply understanding Bitcoin and then naturally going and really having the affinity for Noster because I think the Noster component probably tied in before of how do you understand Bitcoin and then how do you understand how the Internet is wired and realizing that we're going to talk about how early we are in Bitcoin or in AI and the only mental model I can say is like we are still early in the Internet because I think people forget Even though we're 20 years in really commercial Internet after 08, so roughly 20 years, we're still in the first primitives of how we think about centralized data structures and really there was no payments embedded so you had to use ads. And I think we're naturally moving away from that, especially with the level of swap that exists. And so if we're only that early in bitcoin, think about AI and the natural proliferation from a commercial perspective has only been a few years old. So now tying it to AI and buzz. One of the things that we talk about here is like whether you're in Bitcoin and you're messing around with hardware wallets and you had to play with it and that's how you deeply start to park lots of your wealth in there because you really understand the protocol is very similar with AI when you think about working with teams in a workplace perspective. So these guys at least for two years, if not longer, have been at the forefront meaning square and blocks leveraging AI. They built their own harness, which was goose that they recently open source. And we've seen this firsthand and I think I've commented to you guys personally and in Slack like there will be some time and I thought it'd be anthropic doing it where you have some kind of functionality because it's insanely clunky if you're working on a different project in a work stream. And how the hell do you share that context and data over? That's only one small sliver. But you can see how you would start to pull that out. And that's effectively why they built this. But where it gets really heady and interesting is because a the reason why a firm in their thesis, and I believe in it, is that like Anthropic doesn't talk to OpenAI, OpenAI doesn't talk to Grok. And then if you have all these open source tools, how would they be able to share the context, data, et cetera. And so that's effectively part of where buzz fits in. But I think more importantly is that a lot of people and there was like feedback from the buzz like onboarding. It's a little clunky. You have to have an agent, you have to, you know, either the CLI open or the application on the computer. But I think where really this comes from is there's going to be no shortage of applications and infrastructure architecture that leverages all this as the substrate. Meaning it's the back end where you onboard an application. The agent spun up because the, the application is what is paying for the inference and for you to have the agent and then you'll just be moving around and you'll be able to like across different applications. Because historically you've had walled gardens and it all sits with the noster primitive on the back end that's using data or your cryptographically signing so it could recognize who you are in the messages. And so like some of the mental models to think about this is like it's not perfect but if like AOL started and AOL had its own architecture so you could only go into AOL and then the browser existed and the browser lets you search more. Or another example is if AOL only lets you email AOL people and MSN, whatever MSN was, Microsoft X Network only lets you met them. You need that layer that sits in the middle. And so this is like the thing that ties all like intelligence and then naturally it fits into. If you tie intelligence coupled with identity and it doesn't necessarily need to have the permissions of a central database for who you are in your access. Well then eventually you need to park capital payments value through it and Bitcoin sits there. And so when I was first, when I was looking at this like holy shit, like this is the thing, it goes back to The AI side. Like I don't think we're prepared for AI and I definitely don't think people are as like excited of like where this goes. And there was a lot of excitement. So that's kind of a statement in itself.
- [27:14]
Liam
Yeah, I think that this is massive for a couple reasons. One, there is so much information that is not shared within teams that just lives between emails, people trying things and failing or iterating and improving over time and nobody being able to learn from that in public. Working through something like buzz allows everybody to learn together and iterate and create their own evaluations for specific tasks or just have a greater knowledge graph for the organization moving forward. I'm curious a few things about especially I know that Relays will be essentially run by organizations or groups that are startups, open source communities that are using this. I'm interested to see how robust this is in order to not make sure any information is leaked from large organizations. So I suspect that there may be some instances of it being a little bit slower there. But it's really interesting to see just shared compute for open source products here or open source Compute, the iteration over time that everybody can learn from in public and it allows everybody who just starts in an organization to have essentially all of the information and learnings that come with 30 plus years of being a middle manager at some of those larger firms. So I think it'll really accelerate all development.
- [28:40]
Michael
Just a few quick things. So there's the first order, second order and then 10th order and the first second order is again just mental models. If you think about Google Sheets and Microsoft Sheets Sheets like you've historically had to download it, upload it and you couldn't like share. And so you have to be within an organization to share. Like if you're using Google Sheets which is still the diffusion of that's probably 20% of the economy. I think I've shared with you guys that when I was at we were pitching McKinsey and you pull up Google Sheets and they're like what the hell is this that you had multiple people collaborating on there? Obviously this was back in like six years ago or eight years ago, but it's still relevant similar with Slack and Microsoft Teams that you can like integrate and like how much productivity that Liam sharing. But I think like when you go further down like this is effectively how we talked about the hugging face OpenAI sandbox and people like jumped out. This is what's going to happen because to the point of these agents you can effectively tie in the permissions eventually you know what their APIs are hitting. So if you need like fable for like higher level work versus if you're going to hit Kimmy via open router to go into lower level. But these agents will effectively start to gain context and knowledge and then they'll go out and do things and it'll compound and. But the angle longer term I see is that these agents and these applications won't have the data silo of permissions and credentials that Microsoft slack Microsoft Teams Google sheets have. So they will just hop around different parts of the web because of this open protocol. And that's when you see the insanity because we can talk about productivity which is already insane if you can like share with your team and add permissions and give context. But like what eventually happens when all of that is propagating throughout the whole Internet and like the last thing to tie it through is like we've talked, this is how you can really start to more than squint because you could squint before and see slop and attributing value via nostr. And there's a lot of like things that it worked on there where you could discern when people are saying something's valuable because they've attributed some satoshis. And then you can have a graph and look at it, but you can really start to see how if your agent is accessing this and it accesses it via spend to get it, it can like effectively help with that graph to show you what's being indexed and then that will index to you what is the right solution. So like the open source tools, right we talk about OpenAI and they go to ads or they, they put that first thing that's in the front because of aeo, SEO or their own weights. But eventually if the open model is showing you what is most valuable and what's most valuable is because the agent is discerning because you'll just stop to use it. Now you can start in that unit is what you need. And that's why I like part of the stuff we've been talking about is there's the abundance and then there's a scarcity and it's everything in life is yin and yang and you need to have that scarcity and that's the other part that ties into it. And then the nostril protocol is what ties this whole thing together.
- [31:35]
Brian
Yeah, it is very wild. I'll just read through a couple reactions online here. So it's like Slack plus openclaw plus Herder plus some really unique Features that people are sleeping on. Create and interact with agents on top of any harness clog code codecs choose which models agents use. Agents can delegate work and work in parallel and get work trees. Agents are first class citizens and work like humans. Creating channels, delegating access to chat history. You can share AI compute. This is something I wanted to ask you about Michael, because this is a different tweet from Greg Eisenberg who's talking specifically about this shared compute angle. Can you kind of describe for the audience like what that means and what the implications of that angle of this are?
- [32:13]
Michael
Yeah, so the first thing is there's just certain people's wires or brains are wired for certain things. And I just believe, you know, for better or worse, like my brain was kind of wired for Bitcoin. One, because I just didn't necessarily trust the system. But two, like you're always thinking about from game theory and really it's an adversarial network so you have to think about how the hell do you not get wrecked in the system. And this dude's brain is literally wired for AI. Like when you go and look at the stuff that comes out, he's just seeing it for whatever reason. I would highly suggest if you're ever trying to learn about this stuff, that he has a great podcast channel. And so in this tweet, this is right after it's come out. So he's already seen it, he's referencing a few things, like a couple things will click and a pace. A community can now run on top of an open model together. It learns from the group's private data. It gets sharper over time. Narrow private model, better than gbt, impossible to copy. And you could take that data into another deal. But then one of the angles to Brian's point is compute becomes something you can share. Like building shares in a gym. Ten people split, one machine. And so the core idea here is that I mean there's I think different angles but one easy one that we would think about. It's like okay, well what would it look like? And you saw Toby Lucky got gifted. He made it seem like it was easy to run his own like frontier model. And the comments when you saw from Nvidia giving them that I think it was a black well, it was like a hundred thousand dollars. But the point being is it's similar to Bitcoin, similar to like multi institution cost. C. Most firms aren't going to roll their own custody in the same way. Most firms probably at least to start in for a while. Aren't going to roll their own open model, open training. How do you like get all of your people building on it? But the rational thing to do is that for a number of reasons, we've already talked about the same reason. The rational thing to do is to hold Bitcoin and then probably leverage something that doesn't have it relying on a single counterparty. And so the idea is that if you can, if you're an entity, you can build your own, you know, get your own servers, have your own model. And you don't, you can have it on prem right? So you don't have to like have it with a third party sitting somewhere. But then the same, the issue was still going to come up. Well, how do you access it? Like, how do you actually get all of your teammates to access it? How do you host the data, serve it? And then how do you like have an individual maybe still hit Fable or Frontier Lab for certain work, but then also be able to correspond with your proprietary and sovereign data. And so he's referencing like via this, you can effectively have your individual person tied to other agents that can tie into that data and leverage that compute instead of having to hit anthropic or hit, you know, a different third party open router that's giving your data to a different hosting service. So that's one small example, but the others are just like referencing if you're going to pick up a project or something you're working on, you know, like sir, what is it? Discord? And you're a gamer and you have your own model working on your computer and you want to bring people into that and you have the agents there that they can leverage that compute for whatever is being used. And there could be reasons why you would want them to do that versus use a third party that you can effectively pull that into the system. And so he put ideas here like Idle Compute Exchange. Every shared machine sits unused half the day. You build a market that rents the dead time to whoever needs the power right then and there. I think that this is super important because this goes back to and this is, people smarter than me have talked about this. The Internet from a contract perspective was, is definitely built the wrong way. And the way to explain it simply is that data centralizes. And so you can think about like aws, like let's just say it started in Seattle where their headquarters work. And then if you're using AWS now, there's AWS west, there's AWS east, there's probably multiple there. And then you have you know, GCC will have servers there. But the core idea is like similar with Netflix. If you have Netflix and it's pushing data, so it's like taking data at a centralized server and then effectively when it pulls it, when you pick the show you want to watch, it has to push it from somewhere that's probably not near you. And the rational thing to do is to be able to have a more decentralized graph of where that data sits. So then when you need it, it's a faster speed and you, you don't have. It's a more robust network because you don't have a central entity that can like that one throat to choke if you don't. If you like, what do you have? Aws, gcp, Google and Azure are your main providers. Well, if they decide they don't like you or whatever it is, they're only going to use these models. Well, where do you go as an enterprise? There's probably others, but those are the strongholds. And so the notion of being able to effectively graph together where the compute is leveraged. But then also the last part is you still need the token to flow through it, because in this world where you're getting more robust, you can still choke the person off via. Initially it was Visa, MasterCard, et cetera, whether it was the technology to pipe through it. And micropayment's not really tangible or you couldn't do. And then stable coins will be there, but you can just start to see how this all really ends up in this decentralized web with these different primitives.
- [37:21]
Brian
Yeah, it's fascinating stuff. And then this is a tweet from Alex Gladstein saying Nostr's on Fire powering two of the top three trending products on GitHub. So that includes Buzz and Bitchat, which I referenced earlier. We've seen Bitchat usage spike a few times over the past six to 12 months. And it's typically when a government is trying to get people to stop using it. And so this is a story out of India. India moves against Jack Dorsey's Bitch hat sparks legal debate. And I'll just read what Jack wrote here. In recent days, bitchat has seen a sharp rise in popularity in India. Provider sensor towers shared data with TechCrunch that showed that India accounted for about 85% of the app's global downloads between July 17 and July 23, compared with about 1% over the previous 30 days. Bitch was downloaded more than 91,000 times in India over the past five days after downloads jumped 32 fold on July 19 from the previous day. And so I think this speaks to a couple things. One, what I already referenced around the underlying primitives of public private key pairs, nostr as well as just open systems being more robust, decentralized, harder to clamp down. This is from kali. India forces GitHub to take down Bitchat. So they actually attempt to take it down on GitHub which doesn't change anything about what people, the people's ability to actually use it going forward. So it shows sort of the resilience here. And so this is from India itself saying the technical architecture of the application significantly impedes interception, attribution and investigation by law enforcement agencies. Anything on this before we move on, guys?
- [38:54]
Michael
Yeah, I mean this is a core component because this would probably be one of the first examples like make no mistake about it, this is how Bitcoin will get. It will permeate and get across the world. It won't happen overnight simply because it's going to be easier to use stable coins, it'll be recognizable, but there will be the swaps there. But on the edges, like any technological innovation, the people that need it most will go there. And you can already see how embedded in Buzz is you have your own git repository so you can make updates, you can change. That's a big component of this is like engineers and product managers can work on it. The idea that agents and GitHub you need pull requests and updates and people work certain hours so you can have your agent working and there's an audit log. But why I'm bringing that up is because if you're managing one of the server sides of a Buzz network and you're hosting something that is in India as an example, not kosher and you can imagine in the west if they decided models or whatever, well, you can effectively host it now it'd be risky, but you would decide to host it because at the other side of that is you also wouldn't be able to accept payments outside of for Bitcoin. And so you'd be able to get that code base and then you'd be able to pay that person for hosting it. And it's just fascinating. We're seeing this because five years ago I had this conversation one night. It was, it was thinking about if anybody's familiar with the peer to peer like file share networks like the torrents and it was this notion that like between Nostr and Bitcoin you really have. And if you didn't necessarily need Noster but you really needed Bitcoin is you effectively could never ban or censor anything online because ultimately you would have some bounty where somebody would upload, they would seed whatever data that was outlawed and there would be some level of money that would warrant somebody to put it online. So you always like the. The example is I have X book that is Dean puts it would put you in jail for the rest of your life for whatever reason. And there's always a price. Somebody would host that on their computer to upload it. And so once that price was set and made, then you could get that book. And that's what you need the board One one reason why you would need a neutral form of money. And so this just takes it a step further because if you can host that ability via get and then you could take the payment, different sovereigns in different countries are going to inherently get more centralized, more draconian with a lot of this tech because it threatens their, their posture around enforcement. But with these tools, you're effectively going to be able to get any of that code. You will just have to pass some satoshis to get it.
- [41:31]
Brian
Yeah, 100%. I mean, at a higher level. Like why a lot of this stuff is super exciting is exciting to me in the sense that I think for years Bitcoin people have said there are going to be other applications, other uses of these technologies and architectures that basically incept people to recognizing the value of Bitcoin outside of just store value, like the digital gold thesis. Someone has to deeply understand money in terms of why they would want to use Bitcoin as a store of value. But it's these other use cases that people always sort of hypothesize that would actually, you know, allow people to do things, grant them efficiencies, grant them seizure resistance in certain instances where they didn't even necessarily know that like bitcoin or similar primitives were underlying what they were using. And so I think this is actually, you know, now really accelerating in real time. And we're seeing examples of that. But Liam, do you have something on this?
- [42:24]
Liam
I was just going to say building on your point too. I think that we've just been essentially speaking this entire time about how there is robust open protocols and censorship resistant networks that people can build on in order to essentially have sovereignty over their data and censorship resistant communications like it is in right now, for whatever reason, to be almost counterculture in order to have more control over everything that you're doing, from your own intelligence to your own communication layer. And that's just the type of environment where Bitcoin does the best, where there is more purpose and focus on just having control over your life and data and money too. And so I think that this is going to do a lot in order to accelerate Bitcoin's just education, which we know is the biggest challenge out there. And how maybe if the continuation of open source protocols and open source AI will on the edges at least help people understand that Satoshi can't just come in one day and just change the total amount of bitcoin ever mined.
- [43:31]
Michael
Yeah. And maybe to add to that, like I think why I'm personally and I think we're excited is there's the component of like this intersection that the venture firm that we run was set up on, which was the, the notion of doing more with less. And it was really the intersection between Bitcoin and AI. And the natural version of that was just doing more with less meaning leveraging AI to hold more Bitcoin. But the natural version or the next order was going to be that infusement. And we talked about kind of like the less commercial version, the more commercial version is what's going to get it like integrated. And the angle is like, you know, if you have your own personal agent that is required, like think about people shop online, what is the level of E commerce? And if you have an agent that effectively can follow you around the web and have all of your context and then we'll talk about some of these other acquisitions around payments and in banking, but has your permissions, your level, your understanding and then you're running it from a centralized version of like it's your own buzz or however it's. Because I think buzz is like a first version for one entity, but that shit's going to be forked. It's going to be like embedded into other platforms and then imagine that level efficiency and. But the firm consumer side, they're not going to know about any of this. They're not going to know about. Usually it'll probably come with a dollar account. It'll just have like stable coins running through it and you'll have your wallet of your bank account. Maybe your bank account will have its own agent. They can spin out other agents with different permissions for different things. Whether it's like paying your finances or shopping for you or going and buying your groceries. But like you couldn't have that if you have a model embedded in your USA account and then a model at Costco and then a model somewhere else. But if it's effectively able to hop between that, share all the data that you are allowed to and then it's all integrated together. That can't happen unless you have a substrate below it. And that's what like I believe all this is. And then that'll naturally people won't even know what it is. And then eventually we'll figure out where does the kind of like handoff happen from dollars and stable coins to btc. But you can start to see where that's going to effectively that will happen 100%.
- [45:38]
Liam
Yeah. The foundation is getting 100x stronger year over year while all the applications that are going to be built on top of it that can continue to monetize just similar to how it worked with the Internet are going to only accelerate from here with all these better protocol advances.
- [45:56]
Brian
Yeah. And as you mentioned Michael, some related deals that were announced last week, potential deals I should say in the case of this one, Stripe in talks to acquire Open Router and potential $10 billion deal OpenRouter, a fast growing marketplace that enables developers to route traffic across various AI models. According to reporting from the Wall Street Journal, exact terms are unconfirmed. People familiar with the matter indicated the transaction could value the start of near 10 billion up from its $1.3 billion valuation just in May. Who wants to take this one?
- [46:30]
Michael
Hey guys, hope you enjoy the show. Pretty jam packed. Lots happening on the AI front. Bitcoin front. Just wanted to give a quick plug for Back to the Basics. We've heard a lot from the market. There's a lot of people that are interested in using us, maybe not ready for multi institution custody, hence on ramp finance launching. A few months ago I was just actually on a call with a bank. We're going to be bringing in FDIC insured deposits in short order, ideally by the end of the year. Along with some goodies for folks that have been looking to use on ramp. But on the back to the basics plug we've tried to make it as easy as possible for individuals that are interested in using us until Labor Day. And so what that means is you get access to no cost DCAS or 50% off spot buys IRAs. The first hundred, they're going fast but we still have some available for no cost to onboard your IRA and then multi institution custody. A lot of folks either have smaller balances or have been interested in trying us out with a smaller portion of their stack. We reduced that barrier and I believe it's only a hundred dollars an hour. $1,000 for the year if you want access to that so I'd encourage you to go there in bitcoin.com and then as a quick short plug I do have some gradual and suddenly books available with signed by Parker. Good friend, you know, worked with them for years and excited to pass those through to new folks that sign up. So for anybody that signs up making up this code as we speak I'll share it with the team. Use GTS Basics so graduated suddenly BAS Basics and once you onboard and execute your first order if you tag it online I'll make sure to personally ship you out a book to your address. Hope you enjoy the pod. Let us know any thoughts, comments if you made it this far and you heard the first plug. My email is below. You can shoot me a note personally if you're interested in anything we're doing we'd love to chat and I'll respond directly. Hope you have a great rest of the week and we'll be back with signal versus noise and the last trade later in the week. This one, this one's exciting because like we've been talking about this, I think that we, you know obviously Stables is a portfolio company solving this exact issue around stablecoin transactions. The natural segregation, segmentation, bifurcation of stable coins. I think it gets lost. There's so much happening behind the scenes when it comes to tokenized deposits for banks and then stable coins. And you see, you know we've talked about open USD. There's going to be this in same way when it comes to intelligence that it's not interoperable. You have to switch and effectively barter to move between them and it's going to cause a clunky experience. And similar with AI right now people are looking for efficiencies from a dollar perspective on tokens people will look for efficiencies as staple coins proliferate to how do you route dollars? Whether it's via paying for merchant services, goods, payroll, bank accounts overseas. And so it was a logical progression that you would have this effectively collapse on itself. And this is like the first order of it. Stripe looking to take openrider and this is so prescient when you think about stripe because stripe, you know, what was it 18 months ago they did the acquisition for Bridge. It was a billion dollars that was a huge M and A in the digital asset space. Most people didn't see it, didn't understand why 12 months later it looked obvious. Where there's this genius act in place, there's all these firms building on it. Bridge, one of the fastest growing companies Stripes Valuation and I think this is a very similar respect in the open router angle is that they're seeing it from two sides. It's the notion of like routing and the valuations at these firms because you don't necessarily want to spin up all this stuff. I think of open routers like this nice intermediate layer. The buzz is on the deep side of it. If you're going to host and manage all that, that and then on the other side of the barbell is like anthropic OpenAI. And in the middle there's a lot of firms that are looking to route intelligence at the lowest cost. But then eventually you're going to need to pay the tokens on that because the most efficient way would be a digital form. And then when you layer in how do you route stable coins via those payments, I think they're coming at it from two different angles. And so, yeah, Stripe's just like, you know, really at the forefront of a
- [50:31]
Brian
lot of this stuff.
- [50:33]
Liam
100% agreed.
- [50:34]
Michael
Yeah.
- [50:35]
Liam
There's so much that can be improved with stablecoins too. And companies only want the lowest or cheapest way to pay. They're also going to want to earn more yields on their deposit. And that's going to be where tokenized deposits and offshore higher yielding stable coins, whatever sort of rewards. I think that Stripe can kind of just come in and help manage that for them, especially if they want FDIC insured deposits too, with tokenized deposits. And so I think that they're with this aspect just going to get closer to. They already essentially have won the payment application layer almost. They're winning it at least. But I think that they're going to get closer to where the money actually sits too and almost think about becoming more of a bank or just kind of integrate into everything that they're building with the stablecoin side of things with Bridge and agree that the open router acquisition just makes a ton of sense. I think that they're going to get so much data from how companies are using AI too and be able to help accelerate their own deployment of it too, as well as just help accelerate other companies to save time and money.
- [51:51]
Brian
Yeah, great points. The other thing worth noting is that this is also on the back of just a week ago, I believe Stripe pursuing a joint acquisition of PayPal. So they are really going for it seems like they're seeing the field well right now. And then, Michael, I think you had brought this one Cursor, Ramp and Meta are all building model routers, but two have major Model ambitions themselves. What's the story here?
- [52:15]
Michael
Yeah, I think that these different firms are recognizing the value in routing from intelligence perspective and then that you're not going to own the full data layer. Especially from like the notion of there's different analogies people use but like leveraging intelligence to like use you know, an arrow to kill an elephant or whatever. But I thought this one was most interesting because I think Ramp was at the forefront and ramp is going to be in a super fascinating position because if you're managing the finances from a corporate level and your spend well, the logical thing to do is not only to be able to discern how many different solutions you're using, whether it's intelligence or SaaS based solutions, but then also when you tie all that together with what your people are using, how they're using token efficiencies, then you can effectively route them to a lower cost and you can position right in front of the firm, the cfo. Well, we can reduce your cost, let's say from inference by X percentage, but then we can reduce your overall cost from a SaaS perspective it makes complete sense. So I thought it was fascinating to see Ramp step into the game and I think just more will over time and cursor makes a lot of sense as well just from a, you know, sitting as a harness to be able to embed engineers to code more efficiently and then not only hit X or Grok if they still drawn for that acquisition, but other, other models.
- [53:41]
Liam
Yeah, exactly. I think it's pretty. It's actually fairly trivial to build a V1 yourself too. I have a version that will just hit my local AI for some small tasks and the frontier for others. It's honestly not very good just based on the amount of compute that I have at this point. But it's going to get better over time. And so you can see a similar version of where enterprises host some amount of compute themselves and want to keep some amount of their data private for just simpler tasks and then use the Frontier for others. There's a version where this can proliferate more and more and you know, being at the forefront where there can be economies of scale to secure lower cost compute is going to be advantageous to a lot of these payment firms.
- [54:32]
Brian
Next on the list we had a raise on the the agentic payment side payments infrastructure. So Natural Pay has raised 30 million Series A led by Forerunner VC to build payments infrastructure for AI agents. Six products live today. Wallets, Vaults Pay Request Transfer Connect Michael, what's the takeaway from this One. Yeah.
- [54:55]
Michael
So I thought this was interesting because you hear a lot about agents, commerce, banking and so they're not an actual FDIC bank. I think they're working with columns. Columns, which is interesting because I think he was former, I think plaid and he's a guy that's been outspoken around defying crypto and the notion of primitives exist in banking because you want to reduce, you want to increase friction, not reduce it, as we've seen with like a lot of the exploits that have happened. So that's actually interesting. I know it was Colin banking it that he's stepping in here for this, where this gets practical, how it be used. And I cheated a little bit because directionally, instinctually it makes sense. But like how does it use today and the future? And so a few of the angles are specifically like voice AI answering phones. So like if you were a restaurant taking in requests or hotel requests and you're effectively going to pay, like how do you actually manage that? And like isolate from those agents to be able to do that. Like the agents are already good. I think OpenAI just released something there around phone, like phone calls and training that data. But like different things, like customer service agents as well. If you're going to like, like go down that decision tree and then effectively give some kind of refund, well, how would you pull it out of that token, like. Or how would you pull it out of that wallet and deposit vertical SaaS, embedding money merchants and API resellers. So there's just like no shortage of things that the, the core angle here is as intelligence and trust gets better with these, these models and it starts to get embedded into different platforms, you're going to naturally have economic value, move and coordinate, but then you're going to naturally want to build permissions, processes, credit and all the things that naturally exist that a human have. And so I do think that this is a natural extension. So that raises interesting. But then a lot of these things for anybody that's like, well why are these guys talking about like stable coins and you know, agentic payments? Well, it's natural that eventually this stuff will go to bitcoin for a number of reasons that we don't have to rehash here. And so I think that's going to be the interesting angle for us as we talk to portfolio companies, investments and LPs is there's going to naturally be, I think the intersection, and this is partially why I got so excited about Buzz, is you have the crypto people that are closest to World class operators. They were at the stripes of the world. You think about like bridge founder acquired there he was, I believe at Square and a bunch of other large firms in Silicon Valley. But they missed the money part. And then you have on the bitcoin side, understanding Bitcoin nostr? E Cash. But you missed, you missed the commercialization part. Where we sit nicely in the middle is how do you actually look at the products and services that will be able to play in the world that exists today. But as it translates into bitcoin, that is a huge gap and I think will be non consensus and contrarian for the next probably couple of years because it's just not rational for people. And you see this in Silicon Valley. It's the biggest misnomer that exists that AI all these things are going to happen and money's not going to matter. And I think a lot of people believe that. I don't think Elon Musk believes it, but a lot of people just generally believe money will not matter. And there was something that I saw a tweet earlier that was like scrolling down. It's like, if you believe like effectively that paraphrase, you need your brain rechecked because like, it's like, how are you going to flourish or thrive if you can't coordinate economic activity? You need some unit of value. Obviously we shouldn't have to be said, but that's what a lot of these entrepreneurs are thinking, that like money won't matter in the future.
- [58:13]
Brian
Yeah. Liam, you got anything on this or should we move to our favorite topic of the show?
- [58:21]
Liam
Let's do it.
- [58:22]
Michael
What's the favorite topic of the show? Let's talk a little debts.
- [58:26]
Brian
You know, we haven't done it for a few weeks. There's a couple headlines to, to run through here. So Mahlers quits 21 capital as tether's bitcoin merger collapses. And then you had strategy with a couple headlines last week, pulling together a consortium. If you remember a few years ago, they did something similar on the bitcoin mining side. This is more geared towards protocol development, more specifically, at least near term on the quantum issue. And then in filings this morning, strategy is stacking more dollars. And they did their first buyback of Stretch. So I think they bought back about 25 million of STRC. Liam, I'm going to hand this to you. What do you got latest on the dats, the world of dats.
- [59:13]
Liam
Yeah, I mean on strategy, first I think that they have how much now? Almost 4 billion in cash. Whereas their Bitcoin treasury is about 50 billion if I believe the stretch product just isn't a great product and just people have gotten burned by whatever it went down however many percent like 25ish from its what was marketed as safe, secure $100 peg. And since then I think a lot of people will never come back and there likely has been some leverage on the way up. And so I think that the total addressable market for stretch holders is just smaller than people originally thought. They're going to continue to buy more dollars and less bitcoin as a percentage of their overall and probably continue to buy back stretch. But I think that that's, that's just still not going to solve the core problem for them that there's asymmetric downside for investors and this will continue to be probably in my opinion, not the right move for them. And then to the 21 and strike angle. I mean I don't want to really beat a dead horse, especially with so many of those shareholders down so much. But you know, it's easy to sell a dream, but it's tougher to make it happen, especially when there's not really a credible plan. It'll be interesting to see this new XXI company with essentially tether holding almost all of this share. I think that they're essentially going to come out and do what Rockefeller did back in the day where they're just going to go to companies in the bear market now that are struggling and need cash. Especially I think MICA in Europe is specific region that's probably going to see a ton of demand just because the regulations are very burdensome there. A lot of firms either can't get them or after getting them, they're not going to have enough cash to pay off to continue to operate. And so they'll probably make some acquisitions there and then in areas where there needs to be a ton of volume or lending capital, they're probably going to just go to firms and and say either sell to us or we're going to fund your competitors so well that we're going to put you out of business. So it's going to be interesting to track. I think that with a company like that they're going to make some interesting acquisitions. But the pure play, dapp play is very much struggling, just like we've been talking about over the past 15 or so months.
- [61:57]
Brian
Mike, you got anything?
- [61:59]
Michael
Do you want to go?
- [62:01]
Brian
I really don't have much to add. That was a good overview, Liam. I did want to just pull up what you sort of alluded to on the mica side. This was an article from CoinDesk. Europe's high regulatory bar could spark new crypto industry M and A wave. So exactly what you described. Like these smaller firms are not going to be able to get compliant in a way that allows their business to basically continue. There's likely to be a lot of M and A and I think it's natural to assume that Tether would be a player there. Yeah, I don't really have anything on the 21 stuff.
- [62:32]
Michael
So going backwards, the mica stuff is super relevant. We're seeing a lot of the United States. I think there's going to be either a lot of consolidation or just closing downs. We've already seen this in like crypto. Bitmex recently closed down. There's a few things happening there. Another firm's bitmart, I believe. I do tend to agree. I think there'll be some interesting angles on. I don't necessarily know the acquisitions. If you're buying zombies, what that will look like. Unless it's just an optics play.
- [62:59]
Brian
Right.
- [62:59]
Michael
Like some of these firms are just going to go away either way unless that capital helps them. But just going backwards on the strategy stuff, independent of Stretch and what they're doing, I think we've kind of made our positions known and if people are still buying those things instead of bitcoin, that's what they'd like to do. I think the consortium thing is interesting because there was notable folks embedded. So my understanding of this is there's $50 million pledge, so it's not fully like committed yet. Anchorage Arc, blackrock, Block, Blockstream, there's interesting firms there that you wouldn't expect. Like Block is an example. I think Steve Lee is a representative galaxy. Alex Thorne. There's some people that are real fundamental bitcoin believers that make me have a little pause in being super skeptical simply because I don't think there's certain people at these firms who we obviously talked about jack earlier that can be bought. And so there's an angle of along with Mike Schmidt running it, being involved dayto day in the coordination. But also he has a history in. In with. With good reputation in the bitcoin space. So with that said, I think it's just natural to be vigilant and watch this stuff because these consortiums, you know, we obviously people are familiar with like the block size wars and then with the mining coordination that they worked on back in the day. I think it was one, maybe 2020, but I do think it's. It'll be interesting. We've talked about it before to watch the quantum discussion and where, where it starts to accelerate and it doesn't have negative ramifications on the 21 stuff I think I was sharing with you guys before. Like a lot of people, you don't really feel it till you're in this space. Like a lot of warts get exposed when the price depreciates, right when it goes back, everything's everyone's get excited, everyone's frothy. If you go back to 25 and how hot dats were that you have like riches promised, you have opportunity set. And when the price cuts in half, a lot of that starts to come out. You know, whether it's misalignment, whether it's valuations. My understanding without any intimate knowledge is I think that there were some natural versions of like the direction of treasury companies. Naturally they have to like evolve because to your point, I don't think there's enough buyers. We'll see what happens hopefully when the price goes back till naturally this knowledge and understanding. You just want to hold that coin. I have to tweet it out later. But I do think it's going to be worth pointing out that I think a lot of these firms exist and haven't closed up even though we've seen some emperor selling off their stack. I think pivoting to AI, there was an article in Bloomberg about some of these dats pivoting to AI. There was Satsuma that closed up that I think a lot of these firms are naturally continuing with the story because of. Of legal and liability issues. It came out yesterday or two days ago that Alex Mashinsky and I think his COO or CFO are. They owe like $15 million in lawsuits because of their promising of the 18 yield. And it just reminded me of like the, the stretch and like promising these high yields. And so a lot of these firms are going to naturally have to continue with it. But I think that like where a lot of this comes down to is also on the valuation. It sounds like there was just like a valuation misalignment on like what would you pay for strike Vers? So yeah, I think, you know, good for. I think Jack is, you know, all things considered, you know, somebody that understands bitcoin wants it to succeed and that probably ends up better for bitcoiners. That strike is just independent and lets it go on its vision. Because the last part in all this is God knows where tether's going. I think people forgot with a lot of the whitewashing that happened with the Trump admin. It came out last week with Bloomberg's article. Liam, you called it on. Like you have all this capital that's come in. What are they going to do? They're the sole owner. There's just a lot of, of my understanding of, from the regulatory apparatus, tether won't be allowed under the Genius Act. It's just not going to come in. And so I think just a notion of probably being a little bit more careful on the tether side makes sense for, you know, somebody that wants to have autonomy.
- [66:55]
Liam
Yeah, I think. Do we want to touch. Brian, do you have anything more or we should touch a little bit more on the bitcoin security thing?
- [67:03]
Brian
Yeah, I thought that's. Actually, I was going to go back to that because I, I did have just a point that I think there's a natural inclination for bitcoiners to be wary or cognizant of any type of consortium that may be even contemplating advocating for certain protocol changes. And so I think that's a natural and healthy skepticism.
- [67:21]
Liam
I would say.
- [67:22]
Brian
I am cautiously optimistic about this because of the names that are involved, the high integrity people that are involved at these particular firms as well as I think, think I've been on record as saying, like, I think the quantum stuff is generally at a high level, fairly overblown right now. That being said, there, there is still a perception that it is an issue. So something like this, at least near term, does do a little bit to like quell that perception, even if it's just on the surface and there's no actual protocol changes being advocated for by, by this consortium. I think just the fact that it exists is actually probably a net positive in the sense of, of institutional capital continuing to allocate and getting comfortable with that. At least there are large names, large stakeholders who are thinking about this stuff. So I would say cautiously optimistic. It's something to keep track of in terms of the transparency and the direction that this thing goes. But what you got, Liam?
- [68:22]
Liam
Yeah, no, I was just going to say the fact that it's half focused on funding developers and then the other half is just essentially being an intermediary to communicate any thoughts on protocol changes or research for bitcoin, Quantum, et cetera is actually the more interesting aspect to me. You go to BitDevs in New York City and see however many people there are compared to like a sailor AI generated tweet and there's just like a massive divergence in what people are actually focused on. With respect to, you know, advancements with the COIN protocol and tightening that up versus just, you know, who actually is getting a lot of the focus right now. And so I do think that having the education out there is going to be a net positive. And then longer term, I imagine that this, they just have differences in opinions over time of where they think their protocol changes should go, especially with respect to Quantum. And so this is probably a short term consortium just because they're going to have differences of opinion, which is probably net better over a long time period too. So it's not just all agreeing because I can see people like Block and Square are going to think independently, not just listen to whatever the consortium says. For sure.
- [69:39]
Brian
Okay, moving on. We have a little bit of time left. I'm going to do a couple quick hits on some deals. This was Augustus raised 180 million Series B at $1 billion valuation to build the Global Dollar Bank. And then Altman's world coin raised 52.5 million via crypto sales. And then Michael, I know you wanted to talk about this, so maybe if we want to skip through that stuff, we can just go to this. Travis Kalnick's business, his new venture, Atoms, which he's been working on for several years, but in stealth, I guess. They came out of stealth a few months ago and now are partnering with a 16Z. And so what do you got on this or any of the other things?
- [70:19]
Michael
Yeah, the Augustus I think is just interesting because we see Erebor and these other firms that are just going to become, you know, you're not going to know the difference between a stable coin bank and a regular bank. It's just going to be around money movement. Yeah. So the Adam stuff, I don't necessarily know if there was much in this blog. I didn't actually get a chance to read it that was different from, you know, his kind of coming out of stealth when it comes to collapsing and bringing together a lot of the different projects he's working on.
- [70:46]
Brian
I. Liam, you.
- [70:47]
Michael
You had called it last week on Signal versus Noise that I didn't get a chance at that time because it had just come out to listen to those pods. I'd highly suggest anybody that does. He did. He did a like 30 minute TVPN episode and then one with a 16Z. This guy, man, like, I just genuinely, I have a personal affinity. I don't have a lot for a lot of entrepreneurs, but I think like he might go down as like the greatest entrepreneur of our time right now. It's Below Bezos. I think Bezos is just. And it's funny because he learned Bezos is just kind of like, you know, what Bezos was able to achieve. But the thing that I keep coming back to is. And this you, I don't know who else would do this. But like he went in stealth for eight years after the ousting of, of Uber. And like there's two parts of his story that are interesting. One is he talks about it publicly, that he's not not aiming for any kind of redemption or whatever, but you know, for certain he's gunning to go take back Uber. I said it like back in the day to certain people and this was before any of this stuff. They're just knowing what he was working on. But there was. But the other side of it is that he also. And then this is how you can only be like that great, is that you just truly love the game. And so you can only go eight years to go build a business with nobody watching you. If you just generally like that because people talk about it. I think it was on that podcast that you see this all the time. He was very wealthy when he left and most people don't want to go from the top floor penthouse down to the ground floor to restart. And there was this, I think it was in that pot as well, referencing going from leading, you know, 20,000 person all hands, et cetera, to going to a six person cloud kitchens that he acquired and you know, doing the same kind of rah rah. And so it led me going back down to listen to the famous book for them is super pumped. And it just talks about the whole story. And so I got about halfway through over the weekend and it brought a lot of nostalgia back around. If anybody was around from like 2010 to 2014. And just remembering the app stores and I was in Austin and they came in, came out, but they, I didn't. The things that I know they talk about is like growing up and there was a component of his mom was like a salesman and she was like a door to door, like newspaper salesman and she just got rejected a lot. And he learned a lot about, you know, attrition and being able to come back from it. But they reference him like a pit bull that got kicked a lot of times that he never like died or whatever. And then there was something else about like the redemption arc, but you just know that that's where he's going. And then obviously like the. That's all goes without saying. It's easily one of the most interesting things that he's working on because the whole world went to this like sas ephemeral world. And that's kind of where I was always interested. Like if I were Google and WeWork is like the notion and something I kind of miss on the bitcoin side is like, like there's just a real application that no matter how much you can get into technology, there's a physical world and physical constraints that need to exist. And he's always been interested in those hard problems. And so I just think the whole thing's fascinating and it's going to be interesting to see where it goes now
- [73:55]
Brian
that yeah, for sure. I listened to the pods over the weekend. Very, very fascinating stuff. But I've actually never read that book. So I may need to go back and check out the origin story for real. Liam, you got anything on this before we wrap?
- [74:09]
Michael
One? One quick thing. The, the thing on the origin that I kind of knew but I had forgotten was like he didn't obviously invent Uber. Like Uber was James Camp. I think it's James Camp or something Camp. And then you know, he had already sold a company or whatever Travis and so he was just helping out and then he finally got pulled in to join. But the other thing was. So that wasn't his even his idea. But the other thing was that like even peer to peer ride sharing as exists today wasn't his idea. It's like they were doing the black car. Even though they started the. The bounds of the. The lines of what could be allowed, they still were doing like black car. People that had licenses, it was really. And it was a firm before it was called this. And then it turned into Lyft that went in with cowboy and just put phones in everybody's deal for like the peer to peer. But the point of sharing that is that like, like it's just who wants it more at the end of the day and like that guy just wants it more. And that in the, you know, went back and forth. I was just on a pod talking about this. That one of the things we also forget and this is like the return of it and you can call it with Trump or not. There was an insane amount of cancel culture back in the day. And so you're always going to have like there's probably things he wish he would have done differently but you're always going to have. When people are like at that part of. Of the pack and competitive nature, they're always going to do certain things. And so like that's what it Takes to win. But I just thought it was interesting that like all these things weren't his idea, but he was able to out execute everyone else. And then they also made an insane amount. Like when you go in and learn about the amount of fraud, crazy levels of insights. I thought it was the most interesting, the fraud. They never talked about it publicly, but they had to go and hire all these specialists because especially in China and in the US they broke down one of the largest New York drug trading rings because they'd use the. They'd take the. The black market credit card numbers and then they would get the rides and the rides would be transporting drugs and like prostitutes. But then China took it another step further where they were able to completely take both sides of the marketplace where they would like set the ride up. And because they had all these, what is it called, incentives, right? They were burning for that cash to win that market. And so they were just giving like tens and hundreds of millions of dollars in China. So anyway, it's just, it's super fascinating when you go back to like how Silicon Valley was really built. And it just reminds me of like we work days and just like the insanity there across the world.
- [76:30]
Liam
Yeah, Not a ton there other than it's tough to kill somebody who's having fun and he seems like he's having a ton of fun and likes to do it. So yeah, I think that the story is fascinating and I'll check out that book as well. Seems. Seems pretty cool.
- [76:44]
Michael
Do you just come up with a title and stuff to kill somebody having fun? If that got the clicks, that'd be a good title, but I don't know.
- [76:50]
Brian
All right, boys, good rip. We made it through most of the list, but thanks as always for joining us. Like, subscribe Rate 5 stars. See you guys next week.
- [76:59]
Liam
Thanks, gents.
- [77:00]
Michael
Good stuff.
- [77:02]
Brian
Thanks for listening to this week's episode of the show. If you found the information valuable, please share the episode with a friend or leave a rating on your favorite podcast app. All the links we discussed in today's show will be in the show notes inside your podcast Apple app. Before we finish, a quick reminder that Onramp Media is for informational and entertainment purposes only and nothing should be construed as investment or legal advice. Regardless of where you are on your bitcoin journey, we'd love to hear from you. Visit onrampbitcoin. Com Contact to schedule a consultation with one of our private client advisors.