Nvidia partnership and Iren outlook

2026-07-05 · 157 min · Tags: IREN

TLDR

The speakers argued that IREN’s Nvidia, Microsoft and Golden State Warriors relationships support its transition from Bitcoin miner to full-stack AI infrastructure operator, despite concerns about sponsorship spending, delayed commercial announcements and weak share performance. They expect Horizon One delivery, additional contracts, open-source AI adoption and a possible Nvidia cloud designation to improve execution credibility and drive a substantial valuation rerating, while acknowledging that many conclusions remain speculative.

Speakers

Notable quotes

Transcript

₿itcoin ₿utcher: Good evening. It's 9 o'clock or 3 minutes after. We'll jump into this. Hope everyone's doing well. Had a good Sunday here. My wife had her baby shower for our child. He's on the way, expected in September. So that took up most of my day here, but I was able to catch up on a few notes that I'd written down and chat with some people throughout the day and we'll just jump into it. Pretty eventful week. I'd like to, I titled this agreement versus alignment because this was one of the, this was one of those weeks that as much conviction as you have, your focus can stray away a little bit and there's a lot of negative naysayers that are looking for iron bowls to capitulate right now. And I don't think it's time yet. There were a few things this week that I was pretty vocal about that I don't like. But if you understand the distinction between agreement and alignment, I think by the end of tonight, you'll feel a little better about what you own. So just as a reminder, in life, whether it's personal relationship or you're in an organization or you're on a team, you're playing a sport, there's a lot of conflicting personalities and ideas. And the goal of the exercise is not for everyone to 100% of the time agree on every single detail of the plan. The point of it is for everyone to be rowing in the same direction, regardless of individual differences. And this week is a very good reminder of that. Earlier in the week, and I'm just addressing these upfront, the goal of tonight is not to dwell on the negative, but again, to look forward. Yeah, price action sucked, but so did everything in tech. So like, is that really Iron specific, most likely not. And then specifically, we addressed the tweet earlier in the week and we kind of barked up a few trees to address the newest member of the Iron family or deal hasn't closed yet. But let's assume that Mirantis gets acquired and that transaction closes in the next month here, maybe two months. But yeah, I didn't I don't agree. They're God particles in here. There's plenty of market data saying that GPUs last longer than two or three years. And I quite frankly don't know what the CTO was speaking of. But what I would say is that was communicated and that was swiftly addressed and that went to senior management. So I like I didn't like that. It wasn't. I didn't like that it occurred, but it was corrected proper, quickly and swiftly. So that's all you can hope for. And then certainly the elephant in the room this week. I don't intend on making this an NBA podcast by any stretch, but certainly now that the. you know, Iron has signed a $50 million a year sponsorship with the Golden State Warriors, and I believe their sister brand, I think it's the Valkyries, and there's spend commitments probably for both organizations. And I didn't like the move. I didn't like the optics at first, but I'm also still here. I didn't sell, and I think there were a lot of people who You don't have to like it, but to automatically sell or get rid of your position over a $50 million sponsorship deal without having more facts and more information on it seems a little shortsighted in my view. But yeah, I didn't, no one cared. I shouldn't say no one cared for it. Some people applaud it. I was able to get a little more context from some people that I think the argument is it's not a direct to consumer brand play and that it's more you're trying to have a greater presence in San Francisco and you're trying to attract world class talent to build your engineering arm. And on top of that, some of your future customers are enterprises and AI labs that reside in that area. And it's a way to open a lot of doors, not to mention the senior leadership of the Warriors organization, their connections, and their ability to open some doors for iron as they're still shaking this stigma of a Bitcoin miner in transition. And so I understand it philosophically and why you would do that. And no one dismisses the power of the Warriors brand, especially if in free agency, if they're able to sign like a LeBron James and pair it with Steph Curry, like two of the most popular athletes in the United States, if not around the world. So you don't have to sell me on why seeing iron on their jersey matters. But I my initial concern with it. But I again, let me preface this. It's an initial concern. Like I can't make a final judgment on it because it's too early. And that's what I would urge people be a little more open minded to it. You don't have to like it initially, but to say that you can conclude that it was good or bad, I think is way too premature. But my concern with it was our messaging to the market. And this is nothing different than I've messaged to IR before is like, I think the branding of iron right now would be a little better and there would be less of this perception of us being a transitioning Bitcoin miner if they advertise their commercial achievement a little more or simply made more commercial achievements. And it's kind of that chicken and the egg. And then the flip side of it, too, is this was a unique opportunity that you had to seize the moment and it, the prior, I forgot the name of the company, it's a Japanese company, but I think they had the partnership for nine years. And my understanding is this opened up fairly quickly and they struck fairly quickly. So it shows that if management wants something, they can get it done. And I just, I think what we want as a shareholder base And I would argue even some of the institutional holders want to better understand what exactly the plan is going forward, and that remains to be seen. But I think tonight, the best way to go about this is to maybe reverse engineer it. I'm going to pull up the post here, but I had responded to something from Chino Alleman had a nice post that Franz had featured, and I read in more detail, and I want to share that in the nest. If anyone hasn't seen it yet, I'd recommend it. And we can briefly talk about that, but just kind of at a high level, I think where things are going, we already knew we had the partnership with NVIDIA and They have this new DSX architecture that helps. This is all kind of a refresher, but you have this architecture plan where you lay out buildings in a particular manner, data centers with the end goal, and then use their hardware, which is DGX, this platform. And the ultimate goal is to maximize how many tokens or AI outputs you can per megawatt. So you have this factory, and if a token equates to intelligence, which equates to revenue, you're trying to squeeze as much juice out of that lemon as possible, and how do you do so? So my takeaway from all of this was, it appears to me that since February or January, when Irene most likely was having some problems with not necessarily delivery on the GPUs for Prince George, but as the high bandwidth memory shortage became more clear, this is just my thesis. I can't confirm it, but I'd like to think it's an educated view. I think where things started to break down is I think Dan and Will realized that it was time to partner with NVIDIA in order to secure compute for the future. And by doing so, NVIDIA had their own checklist where it was like, okay, we'll give you GPUs, but we want this option to buy equity in you and we expect certain requirements. And I believe Mirantis was one of those. And now it appears, given the relationship of some of the Warriors personnel with NVIDIA, that it may or may have not been related to this, or they might have pursued it on their own and got the blessing of those members of the Warriors organization because they were, have NVIDIA ties. So all speculation, but just my way of trying to figure out kind of the origins of where we've been recently and why we haven't seen as much as we necessarily would like. So working under that assumption, I think it's fair to say that, and I coined it as a, I had re-shared Chino's post and I said that we are a pawn being iron in a bigger game of the hyperscalers versus Nvidia. in some of the large enterprise labs as well as Tesla. And let me give you a few examples of that just for the sake of conversation. And I see, I think Franz was still dropping off his son at school and he'll come up later. And hopefully if Marbles is in here or Chino's in here, if they want to come up and present, I would, oh, let me see what Franz just replied, excuse me. So Franz replied, Mirantis was not part of the investment rights. They were part of the managed service deal with Nvidia. Yes, I agree those are separate Franz, but I think my point is that Mirantis was kind of a nudge by Nvidia to iron to Lock down that managed service seal, which later would follow with an equity investment, but. Thanks for the clarification. But what I pulled up earlier, I did just a quick Gemini search because I'm not as far in the weeds as everyone in NVIDIA. If there is, I was trying to reach out to Mark earlier. If Mark's in here, Mark's a longtime NVIDIA shareholder, if there's anyone else. But I simply put in a prompt into Gemini and just said, what products is NVIDIA developing to compete with Google, Amazon, Tesla, and OpenAI? And so Google and Amazon, we've talked about it before, have their custom ASICs, their TPUs and Trainium respectively, which kind of threaten, I wouldn't say threatens the supremacy, but it does take some market share from Nvidia GPUs. So in response to that, there have been equity investments in a lot of neo clouds, including hopefully iron in the future, but most notably Core Reeve and Navia. So that's kind of one front that Nvidia is protecting itself. The next piece is Tesla that there's a program called Thor that Nvidia released, and I believe they have a strategic partnership with Uber on autonomous vehicles. So that's kind of another segment. And then most recently, this past week, Broadcom and OpenAI created a relationship for OpenAI to have its own ASIC called Jalapeno. And in response to that, not directly, but previously, specifically as it relates to OpenAI and Anthropic, who have roughly 85% to 90% of the market share right now in enterprises, NVIDIA in the background has its own open source model called Nemotron. So these are all kind of the dynamics at play. I wish someone had the courage to come up here with more of a tech background to verify or add to what I'm saying, but we're going to just run with this for now. But in the greater scheme of things, like if you're NVIDIA, and a lot of your customers are beginning to try and take parts of your business, it's natural to respond with competing with them in their own businesses. And then we dial things back to, I don't know if it was a month ago or a few weeks ago with GTC. And I think the main takeaway I took from that, I'm not as technically savvy as many of the talented people in here, but I could tell you that The messaging, which I think Jensen's as good as it gets, he was starting to prepare people mentally to reframe what they see as Nvidia. The narrative was that Nvidia is not going to be able to sell as many chips. And then he comes to GTC and says, we're not a chip company, we're an infrastructure company. And that relates to the DSX architecture that we're talking about, the DGX hardware. And now with these open source models or attempts at autonomous, like they're going completely full stack. But the whole in their plan is the actual data centers and the infrastructure. So as I re-underwrite my investment in IRIN and a lot of friends and family who ask what I think of it, I think it comes down to there's a trust element in Dan and Will, but it's also, do you really think Jensen is going to put his name on a company like Iron if they couldn't deliver? Or he, at a minimum, isn't going to do everything he can with their resources to help Iron succeed? I don't believe that. And I also don't believe that the Warriors would put on their jersey the name of a company that will disgrace them reputationally. It just doesn't make sense. Like $50 million a year is a lot of money and no one dismisses that and they paid the market rate. It's the highest rate in the world, but I don't think the ownership group or the warriors. is gonna, I don't think that changes their lifestyle. Like these are guys that are already worth hundreds of millions of dollars. And by the time they get their allocation of that, does that really change things for them? I don't think so. So all that's to say, and then think about an alternative brand. in the marketplace in basketball, what's the name of formerly known as the Staples Center? Staples Center, they sold their naming rights to crypto.com. That just right now, I mean, the check's getting paid and from what I know, crypto.com is still in business, but is that really a brand that makes the Lakers look better right now? I mean, it'll survive regardless, but I just, my sense is the guys leading the Warriors are a lot more savvy as it relates to that, especially with tech and are essentially giving their endorsement to Iron, which right now doesn't seem meaningful. But I think long term, as the company shows more proof of their progress, which hopefully in the next few weeks is delivery of Horizon, that will This can jumpstart some momentum into the second-half of the year. And I think it's there was never going to be a perfect time for them to partner with them. But if things are going to start to come together now, then maybe at least people are have seen the name and they'll recognize what iron is and they won't necessarily dig as deep on this as the surface level and be questioning. some of the frustrations we've had as a community as it relates to signing more commercial deals or delivering Horizon a quarter later than we had hoped or like all that's kind of in the past. And now we're kind of looking forward. So that's kind of my opening monologue, if you will. I managed to talk 20 minutes here. I hope someone's ready to come up and Let's talk about this further. I think Xcap, he's not available tonight because he's probably putting his kids to sleep. But another good, I can find the post, but the All In podcast this past week was very informative and I respect his opinion a lot. And he said, you got to watch this. So I actually watched the full two hours on it. And I thought another interesting angle this week was There were a few congressional. I believe they were congressional races in New York and kind of this divide that's splitting the country now is the. It appears on the surface that the Democratic Party or the Democratic Socialists are. On one side of the spectrum, and this is an oversimplification. I know there's people that have views on both sides of the aisle that can be for or against data centers. But it does appear that 80/20 that most Republicans are probably pro data center and most Democrats right now would be... So they spoke about that topic and that was kind of an interesting angle that we were kind of... talking about privately and then more importantly, kind of where the industry is going and some of the just things, quite frankly, that I don't necessarily understand them at first glimpse, but I can at least acknowledge that I don't understand them and I can research them more. But there was this idea of distributed inference that you could almost attach they were talking about, I think, Tesla power packs, but like people almost running GPUs locally at their home. And I just that idea sounded kind of crazy and out of this world to me. And then then what was out of this world is them talking about how the big question is the DCs in space, which we've talked about in the past, but those guys are of the opinion that it's more realistic than not and kind of how SpaceX plays into that and when that happens. So a lot to think about, but also really like, I think for me, it just was kind of eye opening and things that I don't understand as well as I think I do and stuff that I need to look into more closely. As far as the upcoming week, I think last week, With markets, certainly, I think tonight there was a futures are up a little bit with the agreement from the US and Iran to delay strikes and revisit where I'm kind of numb to it at this point. It feels like the same playbook where they both talk **** during the week. They may or may not attack each other on a Friday, and then by Sunday or Monday morning, they make up and say that they're going to try again. It just kind of feels like one big shadow, you know, show where they want to talk tough to maintain their political power, yet financially they're sensitive to markets and whether they're acting on that's a whole other conversation. So that happened this weekend, but that feels like the 100th time that it's happened. So you're kind of like, okay, like, great. Same Same thing as usual. And then upcoming though, I think the big thing I saw, maybe he's willing to come up, but Phil was in here earlier and kind of has more of an institutional understanding, but the end of Q2 being the 30th is on Tuesday. So that's the end of, you know, we're halfway through the year, which time's flying. But I think towards the end of the quarter, there was a little rotation out of the higher beta names that had run and some profit taking. And be interesting to see how that if next Wednesday or Thursday with Q3 and fresh P&Ls for some of those guys, if they reenter, how they go about that, especially after the drop that we saw this past week. So those are some of the things I'm looking at and It's all I really got right now. And but as far as iron goes, we had talked most notably about, excuse me, July 19th being kind of ideally the handoff date of Horizon One and other things that we can look forward to are, I think, just looking And when Franz comes later, he can give us an update on some of the satellite imagery that he's been taking in for the various sites, whether it's Childress, McKenzie, or Sweetwater. But my money is on the next deal being Horizon five and six, which once one is handed over and the cadence of two, three, and four is agreed upon that execution risk is removed in theory, and then they can sign another deal. And it kind of makes sense when they have 2000 people, 2000 people working on site that are, you know, they can kind of transition from number one to two to three to four and then five and six. And I think my theory is that with Jensen having so much input in the flagship Sweetwater project that it might not come as soon as we want. I hope I'm wrong. I hope it's tomorrow, assuming it's the right deal. But realistically, if it's going to change the world, it's going to change how compute is conducted. I mean, I mentally I need to rewire how I think of this is I constantly am thinking when I do these, every time I model out a pro forma of where I think revenue's gonna be, you'll look at what I write out and it's GPU dollar per hour, but the reality is like we're going to a world where Iron's gonna have, I don't think they have the capability yet. We were talking about this in OnlyFriends. With Mirantis, that's one layer of the puzzle, but Mike Hillier also said that there were a few startups that this is another prediction. I think as Iron is going to try and compete against the hypers, they're going to have to add additional layers of software to get. caught up and I don't, I think they'll have in house solutions that they'll try and build. But my sense is also that there's going to be more M&A down the road. And one of them forgot the name of the startup off the top of my head. But the idea that if this company that they would acquire can help iron better with their token allocation, like with respect to billing and being able to segregate customers while maintaining their data sovereignty. But kind of the commercial side of it is a missing piece that appears to be. And if I'm wrong, then someone come up and tell me, but that's my understanding of it and that we're going to transition in a world instead of looking at me saying they have X amount of GPUs and someone's going to rent it to them for per hour, I can see a world where iron is charging by the token and how we quantify that. I'm sure it'll be a little more understandable over time, but gets back to the original theme, circling back. If you have an AI token factory, how do you squeeze out as many tokens as possible? Because it ultimately turns into a form of revenue. I'm trying to, let me see. Sorry, I just got a DM from Mark. Mark Cleary, if you don't mind coming up, maybe we can continue the NVIDIA piece and I can ask you a few questions, but if Mark's in here. Mark, how are you tonight?

Mark Cleary: Hey, I'm good. I I joined late. I'm sorry I missed. your opening statements and just trying to wrap my head around what the real focus was here tonight. What did you have in mind?

₿itcoin ₿utcher: Well, see, Mark, I think you have a good perspective because you've held NVIDIA for a good amount of time. And this is more Cliff Notes, so apologies to everyone else who's been on. But I am of the belief that Iron is now You did a great job a few weeks ago or last month whenever we announced the partnership of detailing the ramifications of DSX and DGX as it relates to a token factory. But I think after this week with the sponsorship with the Warriors and some of the relationships with Nvidia and the Warriors, and my growing suspicion is that Iron is a lot more Which it's kind of hard to believe we didn't, I mean, at least for me, it's like they committed to five gigawatts. Like how can it not be obvious at that time? But I just, I see some of these subsequent moves such as Awaken. I think Iron has a new chief marketing, or well, the chief marketing officer was from Awaken, but there was also a new hire that Franz can speak to later. A woman, I forgot her name, but some sort of commercial development. And then, but where we were going with it, Mark, and maybe you can kind of riff on this with me a little, but Google and Amazon have TPUs and Trainium. Tesla has its own self-driving and OpenAI is now comes out with Jalapeno in partnership with Broadcom. And these are all examples of businesses that are you know, companies that have previously been and will continue to be customers of Nvidia, but also kind of are chewing away at some of Nvidia's market share. And then I said that when Jensen was at GTC, like I think his way of trying to reframe the conversation was pivoting from this mindset that Nvidia sold chips and now that they sell infrastructure. So my question to you is, do you agree with that assessment and can you add to it like where you see this playing? Because I see this more as NVIDIA going full stack and instead of trying to make up for lost time and acquire and go through all of the regulatory hurdles, they have the perfect company to partner with and iron that has 5.8 gigawatts announced and probably has at least another five. And in a lot of ways, I was looking at their financials this past quarter, like they had 50 million or excuse me, 50 billion in operating cash flow and they're free. Their their cash balance went up, I think 10 billion for the quarter. So I think Jensen likes having that margin profile and the best way for him to achieve his strategic goals of competing against the hypers, or at least protecting himself from the hypers, is by doing this, almost this off balance financing via utilization through strategic partnership with Iron and eventually an equity interest. But essentially, Iron's doing a lot of the, or at least I believe they're going to be doing a lot of the heavy lifting and it'll be lower margin for Denson, but at the same time, they'll have a competitive moat that protects those lower margins. So just any thoughts you have related to what I've said, and I'm sure everyone would like to hear someone else's opinion besides me.

Mark Cleary: Yeah, I think, you know, I'm kind of, I'm in the same place, you know, I've been right along. The pivot took longer than I expected, and I thought Nvidia and Iron were going to partner sooner than they did. But sometimes I'm overly anxious for things to happen too. And sometimes I'm even impatient when it comes to, you know, what's next, what's next, what's next. And things take time, especially in the physical world, right? You know, Nvidia is dominating, you know, in the, of course, AI space. But things move a little slower with supporting what's going on with AI, and that's the physical infrastructure piece. And of course, that is where Iron comes into play. You know, just from a simplistic standpoint, we know DSX is ultimately the architecture that NVIDIA has laid out. And this is Jensen's vision. Their Nemotron 3 is the operating layer. Iron is going to be the operator. Then you've got the hyperscalers, of course. But where this is ultimately going is this is where Iron's going to come into play in supporting Nvidia's focus on capturing as much of the enterprise and sovereign markets as possible, because that is what is still the biggest piece of the pie that they can target and segment. And that's where it's going. And that's and that was the biggest deal behind Jensen highlighting the new breakout of revenue for the for all the reports going forward. And remember this last report from Nvidia which was their Q1 fiscal year 27. I hope that doesn't throw everybody off because everybody's fiscal years as a company are differing in some ways. But their hyperscale revenue was 37.9 billion this past quarter. And the other bucket that they introduced of where they're targeting and segmenting ACIE, AI cloud and industrial enterprise sovereign was 37.4 billion. So it was, a half $1,000,000 less and growing almost three times as fast quarter over quarter, year over year as the hyperscale spend. So this has kind of been something in somewhat slow motion, but is really starting to pick up speed now where, we're at a big departure from where Nvidia's revenue has been coming from. And Iron is going to be a part of this revenue, if you will, this revenue renaissance, if you will. Iron is going to be supporting this vision that Jensen has had. They're going to be supporting it through the enterprise deals that Nvidia is going to be continually adding, because even though that's going to be a lot of on-premises purchases by, you know, these 250,000 other, you know, paying customers, They're going to be buying smaller scale hardware, but Iron's factories, the AI factories that they provide NVIDIA with, they're going to be the supporting piece to that because on-prem is not going to be sufficient for what they want to do. But it also, at the same time, Iron is the one that can provide the privacy aspect of it all. as well. And that is probably one of the biggest deals and probably one of the most important bullet points for this Nvidia Iron partnership. You know, it's different. You know, Google Cloud, right, and Amazon Web Services, you know, the hypers are all, they're all kind of colluding with all different other kinds of businesses and things. And the enterprise sovereign and industrial segments, the majority of them and the vast majority of them are going to want privacy. They're going to want to be able to manage their own data and not have it be co-mingled. So that's really the biggest piece. I don't know if, I don't know if I quite answered really what you were looking for there, but.

₿itcoin ₿utcher: It's good context that I can play off of. There was a good, I'll find the posts. For anyone on here who isn't an iron bull, I has research was saying that he's here to observe the general level of cope and delusion, which we appreciate you for joining us. You're free to come up. But I was, I say that tongue in cheek, but there's actually a good I responded to this earlier. I'm trying to find it, but I didn't have a as great of an appreciation for. There's a guy called the fundamentalist that I'll share this in the nest here. So this guy, the fundamentalist and Mark, you know, I'm muting just because it kind of echoes in the background. That would help. Thank you. I think that post should be in there now. So it's kind of the same logic. If you read this post, yeah, he's a Nabius shareholder. He actually owns Iron as well. But at a fundamental level, we talked about Nemotron from Nvidia and given Nvidia's involvement, I would think that that would be where Iron's heads at going forward is to favor the open source model of a potential investor and the gatekeeper of GPUs. But this post points out that what I thought was what I couldn't appreciate earlier is your frontier models, your your open AI, you know, so ChatGPT or Claude from Anthropic. Those are pretty user friendly in the sense that you can just kind of get to work on it and use tokens where if you're using these open source models, there's two things market spoken to the data sovereignty. So taking a step back, these open models, all of their parameters are visible to the public. And that's the difference between a closed model and an open source model. So. Essentially, sophisticated enterprises can take these open source models and then they can do what's called fork, meaning they make their own version of it and they add on their proprietary information and then they can protect that for themselves. And ultimately, I think that's the goal of this exercise is the data sovereignty piece, which Mark was touching on, whether it's for actual sovereigns like governments, but also companies trying to maintain their intellectual property. So that's one piece of it. But I think what's telling is that this guy and his article is speaking to the need for the orchestration layers and then the billing layers and all these software pieces. If Iron's going to go in that direction, those are kind of the potential gaps in their business offering right now that they're trying to attend to. So as they pursue these enterprises, these M&A deals, or having engineers on, you know, on their payroll, all with the goal of carrying out this product so that your sovereign or your enterprise can protect their data. But I see we've had Franz come up and I will leave the floor to him and see if he has any comments or response to what I spoke to or Mark spoke to, or just since he's joining us now, if he had anything else you want to touch on, welcome Franz.

Frans Bakker: Hey Butcher, thanks for having me on. I only listen to, I mean, I only caught a little bit here and there when I was driving. So just continue and I'll chime in when I feel like I have something.

₿itcoin ₿utcher: Okay. Yeah, so the open source model piece, I don't know if Marbles is in here. He kind of gave the counter argument to that from a, and this was in Only Friends, but his point was, In theory, it sounds good, but these open source models are still, if you're practicing in tech, which he does, his point was the majority of it is still Claude or Codex, which I think is offered by OpenAI. You know, those frontier models still lead the way. But circling back to that all-in podcast, it was kind of telling to me that how much this element of the race that's in play with China and how the it's it's ironic that the Chinese are endorsing open source when in a lot of ways they're a censored society, but that's for another day. But what they do when they're training their model, they're using all of these tokens and asking Clod or OpenAI as they're trying to train the model, they're essentially asking questions and taking the answers and then putting them into their open source model. So they're essentially like just taking, and I'm oversimplifying it, but there's an element of truth to it where they're essentially using the IP of OpenAI and anthropic to kind of get caught up and then bring these lower cost models to speed because they didn't have to. They were just taking time to distill information from existing models and not necessarily building them. So that was kind of something that kind of new on the surface, but just having Sachs talk about it and Chamath and Gavin speak to it was another element in this and kind of where Iron, I think the big takeaway, though, from all of that was regardless of whether it's NVIDIA's open source model or Anthropic's frontier model, they still need infrastructure. And that's not me making that conclusion. That's just simply was said on that podcast. So that should, in my mind, make you a little more bullish on what you own, where in the past the line was, Iron's chip agnostic. Well, maybe they're chip favored to NVIDIA now because of their partnership, but model wise, they might have to add a few layers of software to better accommodate the open source models, but over time it shouldn't be shouldn't be a problem for them to regardless run a frontier model or an open source model once they're up to speed with their software. So those looks like Franz had his hand up. Go for it, Franz.

Frans Bakker: Yeah, so if I understood it correctly, it's the example in the podcast was Z. And the idea behind this Chinese open source model is that they take the Yes, you can call it the IP from Claude and other frontier models and back feed that into reinforcement learning and post-training. So I guess Marble's point is that at the moment these open source models are not strong enough or on par enough, but some of these benchmarks where they put Z against were like, you know, just marginally worse than, for example, the current GPT or the current Claude model. And it's not to compare, obviously, to Fable or anything like that. But I think the the the trend is becoming clear that, you know, maybe you you get like 10 to 25% better performance in a certain niches with frontier models. But overall, the open source is going to grow harder because of the much cheaper way to access these tokens. I mean, the tokens will be cheaper on a relative basis to the frontier model. So, I mean, I'm just trying to explain it in a way that I understood it. So I think that the idea behind this is going to, it's going to promote mass adoption because if open source models become just as good as your average GPT that people are used to, at some point when the, you know, the economics are going to matter for the end users, this is going to help spread adoption, in my opinion, when open source is going to play a bigger role in this. And there was another post shared in the only France chat today. I don't know if I can find it quickly. I'll do it when I have a bit of a downtime. But there is the growth of open source compared to frontier models is also a higher percentage month over month as far as they could measure that. So I think ultimately it will all come come down to infrastructure and GPUs. So this is a bull case, obviously. I mean, it's promoting a bull case for all neo-clouds, for all GPU as a service providers, and to a degree also to the co-location providers, because ultimately they will need power. But I think for us as iron investors, I think it's music to my ears at least when I heard this. It was a post from Warren Pies. I will share it in the nest in a bit. But yeah, I mean, once you start to follow these things like the All In Podcast and these people that are actually covering more about a topic and the profitability of, the token providers. it starts to become so clear that there is a massive mispricing going on currently where some neo clouds are valued at a much higher multiple and because they have signed backlogs well into the end of this decade. And there are companies that haven't done that yet, by choice and are getting punished for it, even though the economics are improving on the background for both megawatts as well as GPU hours. So you tell me, is this an asymmetric investment opportunity or not? I mean, obviously this is not investment advice or financial advice, but I mean, come on. It just, if you see the post I made yesterday where Gavin Baker is bullish on Sarah Brass, He's talking about Iron's end of 2027 ARR, the exit ARR of Iron in 2027, basically. I mean, of course he's talking about Cerebras, but it's the same, it's basically the same $9 billion amount, right? And then it's 600 megawatts for Cerebras and it's 730 for Iron. I mean, the resemblance is uncanny. And then He's super bullish on it because the market cap is just 40 billion. And now we're looking at iron and it's 17 billion. I mean, this is just one example of massive mispricing. And, you know, I don't want to get too much into the why. I think we can all make up some reason and then the stock will probably have to, you know, suffer again. But it is just Currently, there is a big disparity between these companies, and it's not just, the Nibius and Ireland, it's basically all of them in Ireland. So, I think we, there are reasons for this, but I guess we're going to find out if there will be an addition to our backlogs at some point later this year. So, that's just my two cents on this. adoption going into profitability theory that, you know, was presented in the all in part.

₿itcoin ₿utcher: Yeah. One thing I would add to that, there's the concept of margin of safety. And there's a few people, most notably Mike Hillier, who's talked about iron trading close to its asset value right now. I don't even know where we're at. Let's say 17 billion, but plus or minus, but below 20 billion, I'm pretty sure of as of Friday, I was honestly trying to not look at it. But it pulled up from last year, BlackRock's Global Infrastructure Partners and MGX, which is an Abu Dhabi backed tech fund. They purchased 600 megawatts of data center, and I think it was an additional four to five gigawatt portfolio for $40 billion. So it's like, those are two examples of companies that resemble iron pretty closely, whether it's their, in different respects, whether it's physical infrastructure or ability to compute. And, you know, understandably with iron, everyone sees the last quarter number and just thinks that 3.7 is not going to happen. Or if I say at the end of next year or God particle, if we say by the end of next year, it's 12 to 14 billion ARR that we're full of **** because they haven't seen it yet. And that's understandable, but we know what's coming and that rerate's going to be violent. And that, you know, so the one gentleman in here who says he's here for the cope, I'm I'm happy to oblige if that if that's cope to you. But as far as I'm concerned, like everything we hear in the background, whether they choose the communicator or not, yeah, it's a point of frustration. But they come to next quarter and August and Horizons delivered and that's $1.9 billion a year in annual recurring revenue. And then Prince George is fully installed and that's the 2.4. And then they just have to complete building of H2 through four and then contract Mackenzie and the Childers piece, you know, that 130 megawatts related to the 50,000 GPUs like they have a lot on their plate. But if they show some momentum here, I think people it will The sub-50 is not going to last, in my opinion. I think this is more a macro piece, right? Macro being the culprit at the moment. I mean, they've clearly underperformed amongst peers for the year. No one can argue that. But if they bring it and execute these next few weeks or the prior quarter and we hear about it for a change, then the market will reward them as they should. But yeah, just trading below 20 billion right now, knowing what we know, I don't like giving financial advice either. But this isn't us telling you when it's gone on a crazy few hundred percent run that this thing is going to the moon. It's more like this is beaten down and we see value and we're staying in and we're even acquiring more. what people decide to do on their own time. That's their business. That's their business. But I would say with the open source piece that I also find intriguing is that I talked about the models and all of those, the distillation techniques and all of that, but just, I can't help but just think the companies that are gonna be created. And Dan always talks about, you know, if there was a Dan bingo board, Jevons Paradox would be on there. But if open source makes all of these tokens cheaper, and Franz alluded to the increased adoption, but I think that's actually good for, in some ways, it's good for the frontier models because then their requests can actually focus on tasks that matter And they're not coming back to you because someone wanted to make a photo of, you know, Dan with a shirt saying when deal on it, like there's not going to be tokens are going to actually be more efficiently allocated if these cheaper open source models are able to kind of take care of the less important tasks and then that will give. more, you know, with the constrained compute and token allocation instead of seeing on Claude, hey, come back later, they're actually going to be able to. So it'll actually provide some fluidity to the system, I believe, which is ultimately good for everyone and show growth over time. But it gets back to it's as simple as they have to plug this stuff in and I know they're going to shoot rockets into space at some point and all that, but for now, and of course, Chamath is talking his own book when he happens to own two gigawatts in New Mexico, but it was referring to these terrestrial data centers, if you're in the hundreds of megawatts or gigawatt as Hope Diamond. So if that's true for his property, then at a minimum, that means he values iron a lot. And that means he's jealous of iron and suspiciously will not mention them on his podcast, even though we've sponsored, I should say iron has sponsored the All-In Summit. So just a few other thoughts related to that. But Franz, have you, can you bring everyone up to speed on any developments this week? with any of the properties, anything new. We had a member of Only Friends was driving from, I believe, through Texas to Utah to see family and actually drove by the Childers site. And I think even seeing that video, like I wouldn't mind seeing it in Texas at some point, but when you see the drive-by versus a snapshot from the sky. It's kind of crazy what they're building and what they're achieving. And they don't necessarily communicate that to the market, but eventually they will. And I really hope if anyone is listening on here, I saw Lincoln was in here earlier and he does some consultation for IR's former IR director. Like if they're, if they don't see the opportunity to announce the handover of Horizon One to Microsoft, then I don't understand what they're trying to do. Then it just feels like gross negligence from management and that they're trying to be evasive like that feels like the most obvious photo op. Cut the ribbon, reduce your executional risk perception and show the market that you said you were going to build and you build something and you're on track for end of year. But But related to that, Franz, any other property updates that maybe your average listener might not have seen if they're not a member of your group that you're willing to share?

Frans Bakker: At first, I want to also give a little response to the guy who said, I'm just here for the cope. Just looked at his profile and he's a big believer of AI, especially open ways and open source adoption and distribution. So I think that this is stock related because I furthermore saw his average on micro on his $50. Sure, buddy, great for you. But this leads me to believe he's price narrative driven. So everything we're saying in here, he will, you know, compare to the stock price performance and then say it's cope. So I think that this is a very valuable lesson. I'm just going to tell everyone here who didn't know that yet, but you should actually know that. There's a lot of people that claim a lot of things and say they are an expert or they know that they're right because they have an average on a certain stock that has performed very well. That's not being an authority on anything. I mean, This guy even himself said better lucky than good. So, the things we talk about here are for 90% or more part of the fundamentals. And the stock is, in my opinion, and I can just speak for myself, it's a whole different story. So just be careful who you're following on this platform. So just that was my two cents on that. Yeah, so The sites. I can talk a little bit about Shoulders. Yes, the drive-by video is really good. It's very impressive to see the scale of this operation. I have a picture of myself in May of last year when I was at the Horizon grounds and it was just ground. That was everything. There was no foundational works or any kind of structure up yet. And to see this drive-by now where all these generator buildings and massive liquid cooling plants and dry cooler installations. It's just, it blows your mind when you see it. We do see a lot of it, of course, in the satellite imagery, but it doesn't give you the same dimension as the, you know, these drive-by videos where you can actually get a grasp of the magnitude of this operation. So, yeah, In my opinion, we're still on track for a delivery of the first 50 megawatts cluster called Horizon one somewhere in in somewhere at or before earnings. So I do have a more specified date that I don't want to distribute for the fact that it's going to come back at me if I'm wrong. I am leaning earlier than earnings. So I'll just leave it at that. But yeah, things are on track. And as it relates to that burnt down generator, if you recall, that's been replaced. And the liquid cooling plant is operational. The chiller trucks are no more. They're gone. They're not even just decoupled. They're actually left the site, as far as I can tell. And, Horizon One is looking really good. I mean, if you look at the images from the satellites, you can see that there is just no more construction going on. that's it's a bit of a circumstantial fact, but it's true that if there are no cranes and trucks and, whatever kind of materials between the rows, then you start to understand that it's because they're no longer needed and they need to clear this space for the operational phase, which is, you know, they're right around that time now where they're starting to hand over these buildings from construction to operations. And yeah, so this looking really good. I'm very happy that they were able to figure it out with the with the liquid cooling plant, the Horizon One cooling plant is different from the other three. And it was a bit of a yeah, I guess you could say that the horizon two, three and four liquid cooling plant are the preferred one. So there must have been a reason, you know, why it took a bit longer to get this first one operational, but they did it. So I guess there's that main construction is difficult. They are pioneering to a certain degree. doing a 200 down to 140 back to a 140 and 200 kilowatt rack density is kind of challenging if you are, doing something that has never been done before. And to that point, there was someone who, it's another Mark, not Mark Cleary, but Mark Satogues or whatever his name is, He was basically saying, Iron may be regretting the Microsoft deal at this point in time because of the low GPU hour rates and, the challenging build outs. I think that's a very wrong way to look at it. You know, it is very easy to just compare GPU hour economics of 2026 and 2025. Of course, you're going to get a higher price for a GB 300 if you contract it today. But, you know, given everything we know about this partnership between Iron and Microsoft, I think we should be very happy that we are able to get the opportunity to build a 200 a flexible 140 to 200 kilowatt rack density, fully liquid cooled data center for a triple-A hyperscaler tenant. So, you know, these data centers, like I talked about before, they can be financed with assets, asset financing. So, you know, the quality of the data center is going to matter for their, you know, ability to borrow money on top of it, and the tenant will obviously also help with the rate of the financing. So there are many more secondary ways to leverage this thing, and you're going to see that the IRR is going to be significantly different when all these puzzle pieces come together. So just looking at the economics of the GPU hour for the Microsoft U, I think it's just, it's uninformed at this point to talk about it like that. That's why I have very little patience with people like Jim Chanos who are just zooming in on a single thing. And there are also other people that were iron bulls or are iron bulls that are looking at it completely wrong. It's not an isolated event. It's not an isolated economic situation. Of course, it is to a certain extent, but that's not the point. All the people that are saying, why don't we do one co-location deal, we have 5 gigawatts, are the same people that are saying the Microsoft deal is bad. You know, I can just throw that back at you. Why would you care so much about 200 megawatts of our 5 gigawatts? This is an R&D project. That's what it is. It is a partnership enablement program. It's an R&D project. It's pioneering. It's gathering the smartest construction and research people from all over the country into West Texas. This is something that is, you know, a floor for anything that comes in 2027. So we should be zooming in on this 2027 end of year megawatts amount. That's what we want to get to. That's the same thing that Gavin Baker does, and he's a lot smarter than me. So if he's bullish on Cerebras, we should be, you know, following Iron, reaching their goal. And I think Microsoft will be in part an enabler of that. And you're going to see that if they will deliver Horizon One, at or before earnings in August. You're going to see that signing Horizon 5 and 6 will become a lot better in terms of economics. Iren can talk about uptime promises. They can, you know, there are a lot more things that are happening. So economics are going to improve. If you read the post I shared in the nest, there is profitability all over the stack. And I think if you're bullish on AI in general, or if you're extra bullish on open weight and open source, I think dismissing Iron at this point in time is extremely silly because, you know, the whole notion that Iron cannot get a hyperscaler, I think is a very, yeah, it's a bad take because Let's take Meta, for example. Meta has been shopping at all the Neoclouds, and they've signed deals left and right. They're trying to get all the capacity that they can, because they know that every capacity they get, they will be able to monetize at their own internal margin level. And every megawatt that they lock in is not going to a competitor. So there's been a bunch of companies like Nebius and now Crusoe who are signing away massive amounts of megawatts or maybe even gigawatts in the case of Crusoe. I think Iron is rather not, you know, signing away that much capacity to another hyperscaler who is outside of the NVIDIA ecosystem to a degree. If all the theories that have been laid out here, and I didn't hear the 1st 40 minutes for the most part, but I'm pretty sure that Mark and Butcher have been talking about that, you know, where would you be right now if you would have signed away all your 2027 capacity to Meta? Will you really be able to get all the GPUs from Nvidia, who is starting to see that the neoclass need to be their, friend, their partner. I think this is just, I think Iron chose not to get involved with Meta up to now. That's not to say they never will, but I think it was by choice of Iron, not by a lack of capabilities or whatever. And I also think that the economics were probably not sufficient enough. I think it would be, it's better to wonder why would Crusoe go with Meta and why would Meta go with Crusoe? I think that's a big head scratcher to me because it's generally known that they didn't exactly perform very well in Abilene with Stargate. So And then we know now that they, were tossed out by, I think it was, wasn't it Microsoft who, walked away from them in Wyoming? Or it was Google.

₿itcoin ₿utcher: I think it was Google.

Frans Bakker: All right. So it was Google. And still Meta decides that they want to go with this company that completely relies on Lancium to provide power. And I've shown you one example of their campus in Childress. It's just wildly behind on schedule. So, this is the group that we don't want to be associated with, the Crusoes and the Metas of this world who are pitching tents left and right and bringing in random gas turbines that don't work. this is just not our style, if I can say our style. I'm much more happy to be aligned and, associated with Aaron's pick of the litter and including Microsoft, Nvidia, Fireworks, AI, and some of these other, you know, providers such as Together AI and Base Ten. That's how you pronounce it, right? And so to top it off, I just want to give you guys a very big fat leak or spoiler or whatever that we found out. Aaron, I don't know if you already talked about this, Butcher, the thing that was discovered on Discord. Did you already talk about this?

₿itcoin ₿utcher: Probably not. So why don't you bring it?

Frans Bakker: Yeah, we have reason to believe that Iron's going to announce that they have become a NVIDIA exemplar cloud status somewhere in the coming weeks. So we think that Iron has achieved this. And I remember that when Nibius got it, was a big deal. So this obviously not investment advice, but, you know, I just hope that the market will finally start to realize that iron is not under performing because they are incompetent. It's just they started a little bit a bit later. They chose a different path. They have different partners and of obviously they're going, you know, bottom up instead of top down. So I am very much looking forward to this. It's it's a bit of a validation. in terms of their software capabilities, I would say. I know we don't want to get into the whole debate again where people say, oh, I thought software wasn't important. It's a bit more nuanced than that, but I would really like semi-analysis to give us a shout out when we finally have the exemplar status. So I am looking forward to that one. So I'll just pause here. I'm going to get a coffee, so I'll be back in two minutes. Butcher, you take it.

₿itcoin ₿utcher: Yeah, while he's getting his coffee, to echo his point earlier, if there's still people who are concerned about the economics of the Microsoft deal, then you're not looking at the full picture. It's 300 megawatts of 5800 announced. 5% of their pipeline. What you're not considering is they just went to boot camp and just learned how to put up a liquid cooled data center to the highest specifications, arguably in the world with the world renowned company with Microsoft. And they did have problems and needed chiller trucks and they probably screwed up because they started Horizon before they signed Microsoft, but All of that appears to be coming to closure and in the past and I for one, it's the same reason why when we want, we ask for when deal, like it's the same reason if we haven't. My suspicion is that closer to the delivery that they have been contracting and most of it would be contracted or a good amount of it would be contracted by earnings, but It's the same reason I ***** and complain about 130 megawatts or even 30 megawatts at Canal Flats that's scheduled for last year. To obsess over the margin for such a small part of your portfolio, the reason you would sign sooner if you have the ability to, and assuming that you can build it, yeah, you're not going to optimize the economics potentially. but you de-risk your asset base. And most importantly to me is you get the revenue on your income statement sooner. And that's the biggest thing holding Iron back right now. And whether it's the service example or now, if they're able to get this exemplar cloud status that puts them on a leveler playing field, maybe not necessarily the same software offerings as Navius or Coreweave, but at a minimum, Nvidia, like acknowledging that they actually are an AI company and not this rhetoric that we're still a Bitcoin miner that's pivoting and might try this AI thing. It's just gonna... take on a life of its own when people actually take a closer look at this and get beyond the superficial narrative nonsense that is pervasive right now, especially when price action is leading the narrative, which is, that's why we host these. It couldn't be further from the truth of what's under the hood. And we try and come on here and we have our own frustrations with it and we're not immune to it, but I've never seen Franz personally question the future of this. And I can't say that in the short term, I've been annoyed with some of the decisions, but that hasn't changed my outlook for the next 24 to 36 months on this. But that's the point. Like I don't, how am I supposed to say I know five, 10 years from now that iron's going to be great? Like I just, I feel good about the plan. right here right now. And yeah, we'd like to see it happen a little faster and communicated a little differently, but that doesn't change what they're doing in the background. So, but Mark, anything that Franz and I touched on, anything you want to add or any new topics that you want to talk about? Otherwise, I might invite Jeffrey up here who we were referencing his work on the drive-by video.

Mark Cleary: I know you guys touched on a lot. I like, I love hearing the, I love hearing the happiness in Fran's voice. So, you know, some weeks it's a little, it's a little down, you know, and it's, the growing pains of the day-to-day, you know, sluggish movement of how things work. But things take time. You guys have been doing this a long time. And, sometimes, listen, Iron is a much, much bigger company now than when the two of you started this, right? Of course, absolutely. It was even almost more of a niche company back when you guys started talking. You know, this is, this company is a completely different company now, completely different. And a lot of times what happens is when, you know, you're not a, you know, you're not a, small, small, small cap. I mean, a super small cap anymore, like 1 to 3 billion. You know, the bigger company gets, the farther and farther away investors are going to be or are going to feel they're going to be away from the day-to-day kind of stuff. I mean, none of us work for iron, you know, so I mean, we can't possibly know what's going on, you know, and there are those day-to-day frustrations, which I've learned to block out a lot better as I've as I've aged. And, you know, this whole this whole advertising deal with, you know, the Warriors and I mean, I didn't even give it any thought because everyone was really concerned. You don't have to be concerned. Dan knows the money's coming. You know, it's not a mystery to these guys. They know the money. They know what money they have coming. They haven't disclosed it yet, but they know the money's coming. No one in their right mind spends money when they don't know they have money coming. And again, they already know they have money coming from obviously the Microsoft deal, which I love and I've never questioned. And I don't think it's being, I don't think we are undersold to get this Microsoft deal. It's a wonderful deal. And Microsoft and Nvidia are the partners I want. I don't want AWS. I don't want Google. I've never been fans of those guys and I probably never will be. And I think, I think Iron has the best two partners. I think there's going to be another Microsoft deal in the next 6 to 12 months just to be, you know, I like to play things conservatively because, but I know they're coming. You know, that's going to be another deal. Nvidia is growing really close with Anthropic and some of that stuff gets downplayed, but there's a lot of huge things going on behind the scenes. It's going to end up being a, you know, Microsoft, Nvidia, Anthropic, Iron World here in the next 18 to 24 months, you know, at a larger scale. It just is. And if you haven't figured that out yet, that's OK. Do these do these do things do take time. The other piece I wanted to mention is, you know, BE Partners was brought in by Nvidia. to ensure that the virtual digital twin DSX architecture was facilitated without question. So Iron has bolstered their partnerships with the right companies and the right people at the right time and building out at this pace and taking on contracts as they go is going to prove to be the most economic rewarding way to do it. And that's something I thought two, three years ago before they even officially got together. But I'm very excited about Microsoft and Nvidia and Iron because I think they're going to be doing a lot together. And then, of course, there's going to be a lot of there's going to be a lot of revenue coming from, the industrial, the enterprise, and the sovereign. Probably the two, probably the more from the two former than the last latter. But nevertheless, we have a lot to be excited about.

₿itcoin ₿utcher: Totally agree. Let's bring Jeffrey up. Jeffrey, how are you?

Frans Bakker: Hey, I am. I just want to share more sentiment than data, but I'm actually in the middle of sitting in Moab right now eating a sandwich. I was driving from Texas to Utah for a trip and we passed through Childress and so I took a little detour off the country road and drove to the Childress site. I posted a video for people to look at and It was very calming to me to see that site. I think there was nothing short of like 700 worker trucks parked outside the site. I mean, it was like total chaos there in terms of people and the movement and the structure is enormous.

₿itcoin ₿utcher: And I haven't appreciated like how big these structures are.

Frans Bakker: I mean, the physical height of them, they're like, Much higher than I thought they were in the expanse and the I thought that in my own perception that some of these structures were more slapped together.

Speaker 4: I mean it is.

Frans Bakker: These things look like at least what they're building on that side looks like it's built to last the generation.

Speaker 5: And when I looked at that.

Frans Bakker: I could only think of like what also Chamath was saying on the OLN podcast about how terrestrial gigawatt data set, I know that site's not necessarily gigawatts, it's going to be closer to like 850 in that range, but he said that terrestrial data sites are going to end up being diamonds, regardless of what happens with kind of data sites moving to space. And the numbers that they're saying that they're going to value in terms of the cost and the.

Speaker 4: Build out for data.

Frans Bakker: Like a one gigawatt data center, they were talking about being roughly about 60 billion all in. I don't know if you can correct me if I'm saying something incorrectly, but that's what one of the speakers was saying, that it was like 25 billion and just GPUs. And then the build out was like another 25 billion and all these other things. And then that's where, like, I feel like very calm about how the market cap of the company is really disproportionately low in relationship to what they're building, what they have, and whatever is going on right now in the market with the stock price, it'll all sort itself out in time. I think you just have to sort of take a deep breath sit back and just if you can buy more shares.

₿itcoin ₿utcher: Well said, the piece, Jeffrey, I think Gavin was speaking to 35 and chip costs and then 25 for the labor and all the other piece and like how that will compete against like the 35 will be in place. And then it's this kind of like break even analysis between the reusable rockets that they can launch into space eventually and that whole dynamic. But that just to me, I'm sure there's some people who feel like it's sooner, but I can't see that happening for at least another five, if not 10 years. And that's a very high level assessment based on what I've read. But I mean, if anyone can do it, it's Elon, but I just, I think there's a reason why they're still signing capacity right now. You know, not everyone's just given up on planet Earth just yet. So remains to be seen. I think one other thing from that conversation that I found very telling, and if you have any thoughts on it, The 30 or 40% I think was the number that he quoted was the high band HBM costs. So if HBM is that expensive now and it becomes cost prohibitive, I think you're gonna see fewer neo clouds. It's gonna be, since it is more cap intensive, we're taking the brunt of the hit right now, but if firing does scale up to size, it's going to be a really hard industry to enter. And on the flip side of that is if the HBM's that expensive, I do think you're going to see the GPU life extend beyond four or five years. And we've already seen evidence of it in the marketplace, but even going forward, if HBM's short for, I think, through 2028 is what I read most recently, then that's easily an extra two or three years on what they're buying right now. And they'll be able to kind of be able to build those a little longer. And even the training pieces can be repurposed for inference. So I'm long term very bullish as well for a lot of the reasons you've said.

Frans Bakker: Yeah. And I think I think the kind of the talking about how the costs are-- some people are saying, oh, it's going to get cheaper to build data centers. But then they were sort of correcting that and say, it's going to get more inspensive. The capital costs are going to get more inspensive because it's almost like an inflationary rise in the build-out costs over time. And I think they did spend a little time talking about how NVIDIA is going to try to enter the space. And do you think it's far fetched to believe that maybe Iron will be the main site to provide Nydia their entry into that space? Because it would sort of explain the quiet and the lack of pumping of Iron by Nvidia, because they don't want to sort of broadcast their.

Speaker 6: Move about what they're doing necessarily.

Frans Bakker: If they're trying to enter, you know, that kind of that space competing with the other other models.

Speaker 6: Do you have any idea?

₿itcoin ₿utcher: Yeah, we were talking about that earlier. Mark had brought up the open source model that NVIDIA is starting to develop. And I do see specifically at Sweetwater having two gigawatts that's contiguous and connected and I think, and maybe I'll ask this question out loud, whether it's Franz or Mark or even Jeffrey, if you have a opinion on it. I think what I'm struggling with right now is there's gonna be a world before Rubin and after Rubin. And I think in that after Rubin world, I had talked about it earlier, but monetizing tokens as opposed to, I think the neo clouds, Or maybe it's just from an iron perspective because we've only had a bare metal deal, but how the layout of the site, the DSX architecture, like, I don't know what a deal looks like going forward. I don't know if it's a matter of committing critical IT to enterprises or sovereigns as much as like, I view it almost as like a, commitment of tokens and allocation of tokens. If the factory produces X amount of tokens, then they're buying a share in that output with maybe the residual being able to be sold at the spot market. I don't know what that looks like, and I'd be interested to hear if Mark or Franz had an opinion on that, but I think that's where we're headed and I'm kind of trying to conceptualize it better myself. But I, I think with the Mirantis with their, you know, set up Kubernetes and the orchestration layer to organize these workflows, there might be some sort of acquisition to better commercialize it for billing purposes as well as segregating and scheduling like unique enterprises or sovereigns that are within the factory. But I think the whole point of the factory is being and using this open source is like you're going to have a bunch of entities with the same end goal that want to maintain their autonomy and their intellectual property. But I'm not sure they're crazy about spending a ton of capex on building their own site and or renting a space where if they can just plug into this token factory and commit to a certain amount of output that, you know, they have a budget at the beginning of the year and then they're able to do whatever with their models. Like that's how I think of it. But I also come from a non-technical background so people can poke holes into that. But that's at least a question I want to ask to the guys on stage. Anyone, maybe Franz, if you want to lead off or Mark, if you had a thought on it.

Mark Cleary: I myself, no, I don't have a technical background in it. What I've read from there, though, is that the optimization is going to be so skilled, so skillful, that they will be able to just automatically respond instantaneously to the different types of demand that they're going to be faced with or presented with. So a Vera Rubin factory, no matter how many megawatts it is, and then that's the thing too, the concentration of GPUs and the amount of token output that these factories are going to be able to produce, especially as they build 50 megawatts at a time and time goes on. And you also have to remember that the rack changes are coming too, the voltage changes are coming. So 800 volt Kyber racks, those are coming and those will likely be installed in the medium-term, later-term stages of the partnership between NVIDIA and Iron, particularly at Sweetwater. But the Sweetwater factories are going to be multipurpose. They will be a factory and a provider that is unlike any other data center on this planet. And that's that's where you can get lost in the weeds a little bit. But you're going to be able to pack so much production into a 200 megawatt data center that'll be able to perform like a five or 600 watt data center. And that's where you really can drive profitability. But it is it'll be instant. I mean, it'll be instantaneous automatic switching. There'll be engineers available on site to monitor big picture issues, if you will. But the level of automation that this factory is going to have, it's just going to be beyond anything we've seen. And that's when it's really going to come to fruition to everybody that, holy moly, this is the real deal. This isn't a chip. This isn't, this is a living system. And it's not, oh, this inference chip is better and blah, blah, blah, blah, blah. Besides, you know, Nvidia just took 9% more share of the inference market as of the last numbers that I looked at last week. They went from 64 to 73%. Nvidia is bringing the goods. Everybody else is an imposter. They really are. And there's so much news every day that we get saturated with. It's just, well, now this and now that. Everybody's changing their opinion every half. It's not even a full day anymore. People change their opinions three times during the one day. You have to stick with the king. You have to stick with the godfather. He developed all this. He created all this. This is, everyone else is just like, oh, how can we compete with Nvidia? How can we compete with Nvidia? Oh, we're going to make our own chips. Oh, we're going to do this. We're going to do that. It's nonsense. Yeah. It's nonsense because the total addressable market is going to be vastly, vastly larger than what we see today. This is a 50-year process. You remember what it was like when you got on the internet the first time with the dial-up sound? Okay, this is where we still are. There aren't any AI factories in progress right now. There's nothing going, there's nothing like what we're going to have in Sweetwater especially. You know, the GB300s in Horizon, that's going to be that's going to be something special. And I think a lot of people in here know that it is it's going to be it's going to be really special. But we're we're just we're just so early. I mean, if this is going to go beyond my lifetime, I'm in my early 50s. I don't I mean, no one else has to tell me how old they are. But this is going to go beyond my lifetime. And it's it's incredible. But I just don't think we understand the total addressable addressable market here. You can bring in more and more and more players. You can have people making chips, new chips, this chip, that chip until until the end of time. But the total addressable market is going to keep going because there is not going to we're not going to meet the demand for decades. Decades.

₿itcoin ₿utcher: Franz or Jeffrey, anything to add? Otherwise, let's go to, we'll go to Sean, but I'll give the Franz or Jeff a chance.

Mark Cleary: I'm good. Thanks.

Frans Bakker: Yeah, thanks. I have nothing. Go ahead.

₿itcoin ₿utcher: Sean, welcome.

Speaker 4: Hey, what's going on, guys? Yeah, so just like for full transparency from like how on, like what lens I'm looking at it from. I were an investor from like late, probably like mid-September of 2025. I first heard about it really late, you know, so.

₿itcoin ₿utcher: Hey, Sean, not to be * **** but everyone, we're familiar with your background at this point. We know you're a shareholder. What do you got tonight?

Speaker 4: Yeah, just get right to it then. So I guess like my concern is like, I'm not looking for investment advice, but it, It feels like I'm just like holding on to like, like there's a lot of opportunity cost into it, right? Like Nibius was like a $93 at the same time. I know Franz made a comment that he expects Nibius to be like under $200 in the near term. And I guess like my growing concerns are with some of the things that I mean, you guys even vocalize with the they come out on an earnings call. He looked frazzled, like you mentioned, and then he said, oh, we don't need to market or anything with, you know, the product that we have. And then they acquire this company. They create a marketing officer and now they're paying two and a half times the highest in history paid for like an NBA team and a WNBA team.

₿itcoin ₿utcher: I want to jump in, Sean. So the 20 million was the initial annual amount. Do you mind going on mute while I'm speaking? Alright, so the 20 million was, I believe, from 2017 for starts with an R and it's a Japanese company, but the Gemini research that I found was that it was 40 by the end of it in annual commitment spend. So it's really. 40 to 50 just for the sake of that stat and people can decide what they want. As far as opportunity costs, I mean, no one's gonna, the data supports what you're saying that iron is underperformed, but I just, I don't, I mean this as respectfully as possible, but if you're that bothered by the opportunity cost, then it might be time to trim like I and I'm not giving financial advice, but like if you answered your own question and you're chasing performance and something else, then I and that's the other distinction here. It's like, are you investing for a few years from now or are you trying to make as much money as possible this year, which there's nothing wrong with that.

Speaker 4: I mean, respectfully to that point, right? I mean, like, I'm not a family member of Iran, I think everyone should have an emotional separation between investing. I love Elon Musk, but I didn't buy into SpaceX because of the valuation. We're putting our hard-earned dollars in there not to be charitable and say, Hey, I know that these are great people. I don't do that. I think everyone puts their money into something, whether it's, you know, Bitcoin, because they see this like tremendous future. And Bitcoin didn't have like this alternative to it that, you know, could perform like cipher or anything else, right? Like all of the neocloud plays that I have, aside from that, has just exceeded it. And I guess like, that's where my like frustration is coming in.

₿itcoin ₿utcher: Are you willing to share what your allocation is with iron right now?

Speaker 4: Yeah, of course.

₿itcoin ₿utcher: So my allocation just percentage wise, we don't have to you don't have to show off how much money.

Speaker 4: Percentage wise, I want to say it's like 60% more than that, to be honest, because I have iron two x and I have iron and just regular shares like I went Like all in, in Iran before any other play. And then I just entered in, you know, in like my Roth IRA with like Cipher and, and then I finally just capitulated in buying Nevius, you know, just because at first it didn't make sense. You know, the everything about Iran made sense. And every time I sit on these earnings calls, their management team is just so incredibly arrogant and dismissive and not sharing information.

₿itcoin ₿utcher: Sean, just because we're I'm jumping in a little. I want to be respectful, but I also like you're selling yourself on trimming and they're like crazy that you're going to hear this from me and I don't give financial advice, but maybe you should like you might just like be a little happier about this and maybe should divert. But what I would say is like, if you look at the relative performance of everything this past week, year to date's another conversation and you've identified that and there's reasons behind it. And we're trying to make the bold case that things are turning around. What I would say though, is to like sell into a week where it's down 20 some percent, Like you're playing right into. The hands of the people that are trying to drive it down like it was a low volume week that it's easier to manipulate the share price, not to mention the Russell inclusion on Friday. So there's an incentive for them to drive shares down as a hedge fund and then sell it back to the index at the end of it. So like. Yeah, it's a big position for me. It's a big part of what I'm doing, but my allocation, like I do have other plays and I do think some might perform better than it short term. So I think you have to just like. From an outsider's perspective, like you have to really ask yourself what you're trying to achieve with iron, because if it's about a two or three-year time horizon, then I don't know what you're, I mean, I can understand it's scary along the way or whatever, but there's a lot of evidence to suggest that you're in a macro trend and that we've partnered with arguably the most powerful businessman in the world. Like I don't, it's hard for me to. feel bearish about that long term. Now, short term, if you have leverage or you have, which you said you had two X play, like, yeah, you might get lit the **** ** like that happens or options like coming from me, like a guy who played May earnings and lost and was able to salvage things. These are things I've shared with people. So I'm not saying it from a. place of being judgmental because I've been there and I've lived it what I would say is like just feels like you're over allocated to it and there's nothing like and I'm bullish on it and it sounds crazy to say that but like if you're so worried about the grass being greener maybe it's time for you to find out and just make that decision for yourself and then I think Franz wanted to add to it if you don't mind so Franz what do you want to say?

Frans Bakker: Yeah, Sean, if you would invest all the time you spend on these spaces in studying the company, you would feel a lot better because we are not going to hold your hand through this. is a rough world, Wall Street displaying ping pong with your life savings. Just get confirmation from studying the company or just cut your opportunity cost and move on.

Speaker 4: Yeah, no, I appreciate that, guys. I just want to take an opportunity to come on here because I know there's a lot of listeners that may be in my spot, maybe in a worse spot, maybe they bought at $4. And I see Twitter's filled with all these people that share, oh, I bought this. And it's super easy for people to be comfortable when they bought into a company when no one knew about it. If you're sitting on you know, 500% gains and it drops 20% from that, it doesn't really matter, right? You have this time. You know, you can be the Mike Alfred that can just pour all this money into it. And I'm just using my questions and my concerns and my scenarios as like a cautionary tale, right? Not more to just beg on a company, but for people to like, if they're looking to enter on either of these places, you know, like maybe my bad experiences of taking too large of a position or using leverage or something like that, they can use that because maybe they're afraid to talk. You know, like a lot of people don't like coming onto these spaces and saying, hey, you know, I really ****** **. Here's where I'm at. Here are my concerns. And there's an extra level of sensitivity. to these plays that maybe don't matter in five or 10 years. Right. But, you know, you guys have tons of followers and, you know, not everyone entered in really early on.

₿itcoin ₿utcher: And yeah, but Sean, you know what the difference is? People who the difference with you in particular, I'm not trying to make this personal, but what I would say is like, I feel like we've had this conversation at least four or five other times. with you in particular. So I'm not like, I try and invite you up. I want to be open-minded. If you have questions, if we play bear side versus bull side, like. But it just feels like to Franz's point, like instead of like staring at the screen and seeing the performance of other companies, like it just doesn't feel like at a deep level you've gone any deeper on what you own. And like, I can't do it for you. He can't do it for you. We try and provide a platform to assist it, but like. That's where what I struggle with. Like, if this was the first time you were coming up and you had this, like, it's like, yeah, it sucks and whatever. But like, I can, I mean, we do this almost every week, like out of 52 weeks a year, we probably past two years, we average at least 40 for sure, probably 45. And I can think of at least two or three other times where we've had a similar conversation. That's what I struggle with, if I'm being honest with you. And I think, Jeff, what do you want to say?

Frans Bakker: I was I was off for a little bit, but I just got back in and was hearing the tail end of that conversation. And I just wanted to say like that. I feel like, you know, like Mike Alfredo keeps talking about iron and, you know, he's telling everybody to buy it around like when it was like one or two bucks. I missed that. I entered more around like seven. And, you know, I did enter like terrible for around 90 cents and to have held that thing from 90 cents to where it is today. That was a very painful process, and it was kind of saddled with persistent kind of second guessing and doubts about whether there could really ever pull anything off and whether their leadership was any good and. I just heard so much negativity all along the way, all the way to where it is today. And I feel like we're with Iron, we're getting a second dollar entry point to Iron right now. I feel like if you hold this for the next five to seven years, the $40 today is going to feel like the dollar entry point. And I don't know if that's over-imagining it or over-believing it, but I really do feel like If you want to make 20, 30, 40, 50% on a stock, this is not the place to go right now. I think like if you want to just, or if you want to play the volatility of the stock and make money that way, I guess that's another thing to do. But I see this as more of a long-term investment. And so it's going to have periods where it corrects and it goes up a little bit higher.

Speaker 5: And then it's going to kind of.

Frans Bakker: Be volatile for a little while, and then it's going to correct and go higher again. I think, unfortunately, I think there's a lot of people in the space who are looking for a big bang in a short period of time. And there might be a little bang, but I think you'll be bummed if you, on that little bang, exit at that point. And some people may need to exit at that point for their own personal reasons. There's nothing wrong with that, but I think I'm not here for the short-term bangs. And I think a lot of a lot of the other people who are also sometimes more visible in this space are also not here for the short term.

Speaker 4: Bang, if that makes sense.

₿itcoin ₿utcher: So yeah, I'm trying to build something. Franz is trying to build something for his son. I'm trying to build something for my wife and my child on the way. Like this isn't about tomorrow. This is about five, 10 years from now. And You know, I'm sure Sean's a good guy and I don't want to come off as an ******* but like I just, I lose patience when I feel like you're having the same conversation with people. And by the way, like there are plenty of Sean's in my life who coming from, I saw AI guy in here. You know how unpopular it is to own Bitcoin right now? I had my hometown paper, the Detroit Free Press. The woman reached out to me just looking for someone who still owns Bitcoin because it's so popular to dunk on it right now. And I just like and I have a good relationship with the woman. Her name's Susan Tomper. And like, I just smile because I don't let it bother me. But like, I have one of the guys who manages half a billion dollars is a friend and he's like, even a broken clocks, you know, right twice a day. And this is a Bitcoin bear or whatever. But I remember the run up in 2021 making over seven figures and then entrusting my crypto being Ethereum and Bitcoin to BlockFi and having everything torn down by that *******. SPF and FTX, and then everyone just thought it was ******* over. And then I was still, every time I'd get paid, buying more Bitcoin when it was 15K. And this is the same situation right now with iron. It's a parallel situation. We can all hate Bitcoin. No, I don't want to mine it right now. I think it is hurting. our short term share price, which I know I just said I'm planning for the long term, but to the extent that we're going to weaponize our equity and use an ATM, I'd like to have a higher share price and not sell down here. So that's why I want to get rid of mining. But I've like, I've lived this and I know Franz has lived it. I know AI guys lived it. I know Jeff, from what you've said and Mark, it's like, You have to be 1 tough *** ** * ***** to stay in this game and just not let it bother you. And it can bother you, but as long as you stay in and whether what however you got to deal with it, whether you got to go work out, you got to go do something with a loved one, maybe have a friend disconnect with it on the weekend, whatever it is, but. Yeah, I just that mindset is such a losing mindset and I and I fell victim to it earlier this year and was chasing the home run swing for a Sweetwater announcement and almost blew everything I have up. So if anyone learns anything from this here, like I I've walked this road on both sides of it. I've had the patience when things weren't looking good and buying Bitcoin. you know, 15 K when it had been at 65 or 70 K or whatever, and then some ******* decides to blow up, blow it up with FTX. But I think AI guy wants to come up and comment, but I think he feels the same way as I do about this and we'll get another perspective. But I think I'm going

Speaker 5: to just throw in one thing.

Frans Bakker: I think in some funny way that I think Bitcoin is a little bit easier to hold. Because I have all my Bitcoin in a hard wallet, and every now and then when I think about, oh, maybe I should sell it, I begin to think like, I don't think I remember how to move the Bitcoin off my hard wallet.

₿itcoin ₿utcher: It's like, I haven't done it in so long.

Frans Bakker: And so it creates this sort of inertia to doing things. And in some ways, there's a lot of talk about tokenizing stocks.

₿itcoin ₿utcher: And when you start tokenizing stocks?

Frans Bakker: It might make it like having a trading platform where it's so easy to like press a button and then move something. It really, it makes you much more susceptible to being governed by your emotions than by logical thought. Because it's, you know, you're watching like volatility over the week or volatility over the day. You have a little micro anxiety attack, you hit a button and then all of a sudden it's gone. You know, and it's, if you have your Bitcoin on chain, I mean, you can do that too. But when the Bitcoin's off chain, it really, that's made it, it's sort of out of sight, out of mind in a way. And I really like, again, just to say again, like when I drove by that site, that really gave me an enormous sense of comfort and warmth and confidence that, you know, When people start going like, this is a garbage company and like, it has poor leadership just because of the absence of PR and the volatile price action, you go see what they're building and the amount of intense kind of labor that's going into it at the current time in that town. And that's just one of their sites. You just realize like, I mean, these are people building something to stay, you know, I mean, they're building, They're building for a future. And they're clearly not building for kind of a short-term return and just trying to get something up that produces.

Speaker 6: Or meets someone's need for like the next two, three years.

Frans Bakker: I mean, they're really building for the generation. And yeah, anyway, I'm going to sign off after that.

Speaker 6: So thanks for letting me speak.

₿itcoin ₿utcher: Yeah, thanks for joining tonight. Let's get The man, the myth, the legend, Bitcoin AI guy. How are you?

Speaker 5: Yo, you guys mentioned my name a couple of times, so I thought I'd request to speak. First and foremost, Sean is a ***** *** *****. People like that don't make it. You know, I've seen thousands of Sean's, you know, over the past few, you know, years. And, you know, I posted a screenshot of a Sean. His name's not actually Sean, but his name is Puru. You know, many of you guys don't remember this guy, but he was just like, he was just like a Sean, right? So I'm going to read this tweet. This is from February 24th, sorry, February 24th, 2025, about a year ago. Tweet goes, Bitcoin miners were the wrong investment call this cycle, at least till date. Lots of opportunity cost lost, irreparable damage caused. I was part of it until June last year, but for good reason. I moved away to MSTR. Stuck with my old bags from 2023, still need to offload on any run. This is a guy that, you know, similar to Jeff, He bought Wolf at the bottom, right? That he sold his winners, his miners, you know, AKA the HPC AI equities that have gone up 10, 20, 30, 40, 50 X since the bottom. And he was there at the bottom. Like I was, you know, one of the few, it was me, Mike Alfred, pretty much, you know, bullposting it in 22, 23. And he was there too. It wasn't many of us. There was like a dozen, right? And we were just like the weirdos, right? That were bullish of these equities. And I thought we were all gonna make it because we were so early. I was like, okay, we brought the bottom of Bitcoin. And yes, it's volatile. Even back then, we were like, this is like three to six X as volatile as Bitcoin. And, you know, I've been following Bitcoin for 12, 13 years. So, you know, I know what volatility is. So this was, you know, fun for me because it was like reliving Bitcoin in the early days, right? And I've held them through like 90% drawdowns, right? Like you mentioned FTX and previous cycles. That was actually very rough. Like today, It's a joke. I remember I was in the space yesterday with Mike. He was like, how bad are you down? I was like, I'm up. What are you talking about being down? That's not how I look at things. We've come so far, right? And there's another 10x or more to come over the next three, four, five years. That's a big opportunity, right? That's why I'm excited.

Frans Bakker: It's Bitcoin AI. I think you're giving too much credit to Sean to compare him to this Puro guy. I think Puro at least is someone who posts and who shares things. The only thing we see from Sean is he pops up in these spaces. He wants to confront bulls on a live stage with only negative takes concerns week after week. Honestly, I think we should label him a troll. I think he's got, he's insincere. I don't think he comes here to get his concerns handled with. I just think he wants to confront bulls on a live stage to trigger, to get engagement. He literally just said that, you know, he wants to, you know, there's a lot of people listening. That's what he said. So, you know, what does the amount of listeners have to do with calming his concerns, right? It's a pure engagement play and I think it has a troll angle into it. So I kicked him off the stage. So just you guys know he didn't leave. I kicked him because. I have about 1% of the patience that Bitcoin Butcher has, and the only reason I have let him on before is because, I thought it would be a good opportunity to handle with legitimate concerns. But this guy is a troll. So whatever the Puru was, he's probably a bad investor, but Sean is just a troll. So it's a different story. I just wanted to say that before you, you know, go on. So go.

Speaker 5: Yeah, but still, like the volatile, the point is the volatility is not for everybody, right? There's going to be a lot of people that use the short term, medium term, like 2 years is I think a short term, right? Like, So people like to flex their year to date, like if you're not up, you know, 69% year to date, like you're a ******* loser, right? If you're not up 10x every year, you're a ******* loser. Like, who thinks like that in real life? Like, you know, if you talk to a boomer on the street, you tell me you're up 20% year to date, that dude's going to think you're Warren Buffett. Like he's going to be like, yo, are you Mike Alfred? Like, who are you? like you know posting 20% year to day like that dude's like you know uh long Blockbuster right so uh but we're we're definitely in a in a you know in our own little Niche right I don't want to say the word bubble well I just did uh but we're not in an AI bubble we're definitely in our you know financial Twitter ****** **** right and uh You know, we could, we could lose sight of things because we see things 10xing every year, right? Like I had two 10xs back-to-back year, you know, year over year. That was insane. Like I, there's no way I'm gonna repeat that like ever. So my, my position is just like, I'm just gonna hold the winners. And I, you know, I could have sold iron at 76. last year, but I didn't, right? Like the, you know, my thesis was, you know, Bitcoin was like, you know, like I'm a long-term Bitcoiner too, but I had a feeling that Bitcoin could go down to 70K and it went down to 59, right? 58. But I held iron despite the fact that I suspected Bitcoin would pull back significantly because of the AI upside the you know unfortunately like it would have been smarter to trim and and buy the bottom right but that's just not how real life works I was like okay there there's too much upside with Sweetwater and and all these catalysts uh you know the time to compute I know it's a beam right uh all these other things but I'm here for years, right? Like Butcher said, you know, I genuinely, I don't know if this is iron at a dollar again. I don't know about that, but I could see a 10X, right? Like, and I'm willing to endure, you know, multiple drawdowns. So how many 50% drawdowns have we had this year? last year right the year before that it sucks right uh but that's just what you have to go through to to you know to capture some of these gains uh and you know Mike Alford will tell you like I don't like he thinks I think that there's too many uh more 50% drawdown to come uh but I like to prepare myself like for the worst right so if we only have three more 50% drawdowns from here, like that, that's a win, right? But if you prepare yourself to say, okay, I'm in it for like, if I want another ten x, right? And I can withstand mentally, financially, like another ten more 50% drawdowns and several round trips, you know, you know, I can do that trade. I can do that trade off, right? And that's just how I'm wired, right? Like for the past few years, like I've been talking about these drawdowns pretty consistently. And it's actually, I've been optimistic, right? There's been more 50% drawdowns than I predicted. But I don't mean to sound bearish, but like you gotta be able to endure. But the benefit of AI, right, is the sustainability of these earnings. You're not going to see these, well, I believe that AI is not a bubble. Maybe I'm wrong, but because of the sustainable earnings, right, the access to credit, like prior to all this AI stuff, these Bitcoin miners, no one would lend a dollar to them, right? And they had cyclical revenues. But now, right, this time is different, that, you know, you could actually hold these things, right? Like it was very, 2022, 2023, it was a very contrarian idea to say, you know, Bitcoin mining stocks were investments and not trades, right? But now AIHPC, like people are saying like, you know, it's understood that AI is here to stay and these companies are actually, going to get some ROI from their activities, right? So, I'm rambling at this point, but yeah, dude.

Frans Bakker: You are, but I sympathize with your sentiments.

₿itcoin ₿utcher: You lose your mic, Cranz.

Frans Bakker: Every year, otherwise you're not cool. You know, my best returns on a single stock has been MetaPlanet. And you never hear me talk about MetaPlanet. I had a 36X on MetaPlanet from, what was it, 2024 to 2025? And I trimmed heavily. And I've shown screenshots to fellow MetaPlanet investors where I was trimming. And for some reason now, everyone remembers MetaPlanet as the big failure and how it's down I don't know, 80% or something year over year. But for me, it's been my best success on paper ever, right? But, I met these people from Metaplanet in Japan, in the US. I was pretty bullish on the whole idea behind it, you know. I still believe in Bitcoin, but, you know, I just, saw a better opportunity in terms of fundamentals, long-term investment. And that's why I don't really feel anything when Serenity says, you're still back holding iron to their 600-0000 ATM. You know, all I'm thinking about is, you know, I don't have to be ashamed of my year-to-date performance. This is not a one-year race. The race I'm running is a multi-year investment race, right? And just because I am not having the best returns ever, like the Micron investors or the Nibius investors or some of these other stocks having in 2025 and 2026, doesn't mean I've never had my own personal 10x or even 30x, right? This is just the market just, and especially Twitter, is just only out here to find the next hot thing. Hot money is everything on X. And I just think that that's not investing to me. So, you know, if you're in these spaces, you'll find out that We take our sweet time to explain this story to everyone. I mean, we've been running these spaces now for over a year. If you're not here to learn, then you will never be strong enough to endure these drawdowns. Ultimately, we still have the best ahead of us. And that's why I am so bullish on iron and I will keep telling the story. So that's just What I wanted to say, and you can continue to ramble if you want, or maybe Butcher has someone else.

Speaker 5: Dude, there's a point. There was a point that we were the serenities, right? Not too long ago. You know, I don't know if you guys remember, but the first Bitcoin equity that really rallied the cycle was Cleanspark. And then it was MSTR, right? And then Iron obviously had that crazy, you know, 2025, right? But CleanSpark was like the first stock that really did a 10X in a year. And I was like the loudest CleanSpark bull, right? Like there was a point where I was a serenity, like no joke. And I went from like, you know, 50 followers to tens of thousands. It was crazy, right? And it was just because I was bullish on Bitcoin when everyone was bearish. And more specifically, I had a thesis that operational leverage was more important than a Bitcoin huddle. And, you know, and I further, like, it wasn't just, you know, I was open I was open minded to to the to the AI upside right back then it was AI upside, because it was we didn't understand how, you know, the true demand was until, like, you know, more recently, 2024. Like, no one could really wrap their heads around $7 trillion, in AI CapEx spend. And now we just throw around trillions, you know, every day, like, like it's nothing. But, you know, it's it's crazy how fast this market moves. And yeah, Twitter's definitely focused on the hot money. So it's not sexy to talk about, you know, a five year ROI on a data center.

₿itcoin ₿utcher: Let's go to Nick. Nick, long time no see.

Speaker 6: Yeah, what's up, guys? I was just kind of share my thoughts, you know, I guess you guys are in kind of the same position I am where you're just sitting on such massive embedded gains and the thesis over the last several years hasn't changed, right? I mean, AI data centers are going to continue to be built into a global infrastructure. And I don't think that there's really a better opportunity for a risk adjusted return than some of the names that we're in right now. And so I've kind of just moved them into a separate portfolio that I intend to hold for, I don't know, I'm not really sure, three, four, five years, six years, could be longer and kind of see how it goes, but I don't really intend to sell any of them. And so I know a lot of people maybe showed up late, but what I wanted to reiterate to everybody is like, as a retail investor, you don't really have an edge, right? I mean, the edge that you have is to group up with other smart people like the guys that are in this space and try to share your thoughts and try to gain some sort of analytical edge. But in terms of trading, it's very difficult to trade against institutions and volatility and order flows and so forth and so on. And so the edge that you have as a retail investor is the inability to be liquidated or margin called and the ability to hold for a undisclosed period of time. And Like we've all sat here and I remember when like I ran drew down to five, 6 bucks from, was it 15, 20 or whatever it was? I mean, we've taken some bad hits, but we're all still here. And over a number of years now, the thesis is held. We're sitting on incredible gains and really like, I know like Bitcoin and I guy and I don't know where you're at, Butcher, Franz. I mean, you guys are all big chillers now, you know, just China chilling. So it's definitely possible, but I don't even know. Like if I exited my IRAN position or my cipher position or my HUT position, like I'm not even really sure that there's a better risk adjusted return or a better thesis that that I know about right now that I would feel comfortable moving into. And the amount of diligence and work that it would take, I'm just not even sure that it's it's worth it because the AI infrastructure play is going to be perhaps one of the most lucrative ones from a risk-adjusted return basis for a long period of time. Now, that being said, I'm curious to hear your guys' thoughts. You know, I've been accumulating MetaPlanet, I've been accumulating MicroStrategy, and I started accumulating Bitcoin at 70K, and I accumulated kind of all the way down to 60. Who knows if it's going to go deeper, don't really know, starting to run out of cash anyways, but It certainly feels like we're getting close to capitulation across the board, right? Like there's a lot of negative sentiment. It's been just sideways asinine price chop action for quite a while. And, you know, do we go down to 45? Do we go down to 50? I don't know, but it feels like a good time to continue to accumulate. And like, you know, that might be where the opportunity or the biggest returns come from in the next say 12 to 24 months. The last thing that I'll say guys is like for you guys that have held the positions that are sitting on massive embedded gains, if you haven't been like assigned a account manager at your brokerage that you know you got his cell phone, you can get him on the phone. They got all sorts of crazy programs now for like pledged asset lines and different ways to extract cash without taking tax hits. And so you can get really creative when you're sitting on those gains. If you really believe in the company and you want to hold it for three to five years, there's no real reason other than a fundamental change in your own thesis and your own belief structure about the companies that you own. But there's no fundamental reason to sell the companies. There's no reason to sell them to raise cash, most certainly. So I don't know. That's just my thoughts, guys. Sorry, I've been absentee. I've been traveling. I've been working on, you know, some of my other portfolio companies that are actually companies that I own, you know, that I help operate and advise. So, but I'm still here, still chilling. I hope you guys are good.

₿itcoin ₿utcher: Let's go to Franz first. Franz, you had your hand up. Alright, well, Franz is figuring that out. Jeff, what do you got?

Frans Bakker: I just overheard the pledge asset line. Can I get a thumbs up if you can hear me? Because I cannot hear anyone on my desktop anymore.

₿itcoin ₿utcher: I've let Jeff start speaking, but yeah, for a minute we couldn't. Now we can hear anyone.

Speaker 5: Yeah, we can hear. Yeah, I hear you.

Frans Bakker: I was just going to say something briefly about pledge asset lines. So what's interesting, like I just discovered pledge asset lines recently. I'm not. big finance person. So just kind of... Yeah, I just wanted to, did you ever talk about the negative gamma exposure theory? I thought that would be an interesting one to talk about just briefly. It seems that I'm not 100% sure if this is completely correct, but I have heard a lot of stories about why the iron stock has such a high volatility and so much beta to macro. And we react so much more extreme to both the upside and the downside, but particularly the downside. So the theory is that Iron, because of the massive amount of convertible notes and the leveraged ETFs, is in the absence of news in a permanent negative gamma exposure regime. And this is what's why the price is reacting sometimes counter-intuitive or mostly downwards. And I think that's why I wanted to say this is because it is a perfect mismatch with Iron as a company. And I think that's why we're seeing such a big disparity between the, you know, the promising future of Iron as a company and extremely negative stock performance at the same time. Because like I said, in the absence of news, iron as a stock is mostly being drawn to the downside. I mean, I've given my own variations of describing this issue. I've called it hedge fund aids. I've called it like an hourglass that keeps ticking down. And it needs to be turned around in order to go up again. And that's the thing. It is basically true that without any news, the stock just will keep going down like a balloon that's deflating. And there is, so there is a real truth behind that because of the gamma exposure. And I just thought that was a very interesting theory. Of course, this is not investment advice, but I think for people that question themselves holding this stock when they are bullish on the company fundamentals, but bearish on the ticker, which I feel myself sometimes as well. Like I don't have that many hairs left on my head, but if I, the ones that I do have, I'm still trying to pull out sometimes when I see what's going on. I think it's really wise to start to learn how these convertible nodes arbitrage and, you know, the leveraged ETFs like the IRE and of this world are negatively promoting the volatility. And with the gamma exposure, it's just it's like a perfect mismatch with Iron, who is very quiet most of the time. We have a very inconsistent PR machine, a very inconsistent marketing team. I mean, there were people that said Dan Roberts had like 26 tweets in May and he had zero in June. I mean, I guess up to last week, I think I had one post about the new sponsorship. But that gives to show that Iron as a company in communication is not feeding the right signal to the stock to break out of negative gamma. So I mean, this is not the only reason, of course, and the absence of news is driven by fundamentals to a certain degree, but it's also the company's culture to not brag about things, to not talk about things that are too far in the future, that are not 120% confirmed, et cetera, et cetera. So The fact that Iron as a stock has a leveraged ETF is already a problem because the company is not really suited for that kind of exposure at this point in time because the fundamentals are, I mean, the earnings are not there yet and the company is not really that kind of company to feed news on a very consequential basis, very consistent basis. So I just thought that was something, I wanted to share because it really helps me to deal with this volatility, knowing that this is maybe a big reason behind the disparity between the stock and the company. So I'll just pause there.

₿itcoin ₿utcher: Jeff, you got cut off earlier by Franz. You want to finish your thought and then we'll go to AI guy and then I'm going to try and wrap this up. Yeah.

Frans Bakker: Yeah. Hey, I was just going to follow up, but I just want to say something. Follow up with a quick comment about the pledge asset line thing. I'm not a financial advisor, so like this is not financial advice, right? So but the pledge asset line is actually, does everybody know what it is? Sort of, like the other guy mentioned it really briefly. Do people know how that works?

₿itcoin ₿utcher: I'll describe your understanding of it and then it sounds like a good follow up.

Frans Bakker: I actually took all my assets and threw it into the pledge asset line. So what happens is all my stuff is with Charles Schwab and it's under a pledge asset line. And with that, they extended me a very generous line of credit, which bears at an interest rate of about 4.9%. What you don't want to do, if anybody does it, is you really don't want to, I think, go above 25%. So they'll give you more, but you're going to run the risk of maybe being margin called if you have a smaller amount and you're taking out a larger portion of the credit line they extend to you. So I would be very careful with that for anybody who looks into that and gets excited because you get a big line of credit. But it is actually, I talked to my tax accountant, And he said that every dollar of interest I pay on that, I can end up writing off against any other capital gains. So if you think about it, make up numbers, I'm going to just make up big numbers. You have 10 million, they give you 5 million in credit, you take out half of that, you take out two and a half million in credit, you pay interest on that two and a half million over the course of the year.

Speaker 6: At 4.9%.

Frans Bakker: At the end of the year, you could sell something if you're in gains with what you have that bears a capital gains cost equal to the amount you paid in interest, and you basically get the loan for free at the end of the year. It's a little bit of a **** **** to think of it that way, but I mean, basically, I think that's how a lot of people never part with their stuff.

₿itcoin ₿utcher: Good stuff. AI guy, you want to anything to add? Otherwise, I'm going to wrap this up.

Speaker 5: Yeah, yeah. So with about like the negative gamma, you know, it's all the max paid conversation, right? So it's kind of, you know, another meme. But on the on the on the convert hedging, right? I mean, that has been a red flag in this uh sector for years like people didn't want to touch a marathon uh you know for that reason I know you know it's in their bottom signal comparing Ireland to marathon uh but like Marathon got a lot of for having those converts uh years ago right so did uh MSTR and uh you know obviously you know it creates that extra volatility. But, you know, it's just part of the game. But I want to answer Nick's question about, you know, Bitcoin, right? I think, you know, I'm going to answer this in like from a trader's perspective and then from an investor's perspective. From a trader's perspective, like a very short term time frame, you know, months to within two years, I'm more excited about IRIN than MSTR or Bitcoin, right? I think if I can increase my buying power by waiting for IRIN to take a leg up and trimming some to opportunistically accumulate Bitcoin at the lows, I think it's a better play. than just buying Bitcoin or MSTR today. You might buy them at higher prices, Bitcoin and MSTR, but your buying power increases because iron could double, triple, or more in a very short time period. That's a trader's perspective. From an investor's perspective, this is an optimal time to buy Bitcoin, right? I mean, if you're looking at a, you know, five, 10 year time horizon, like you want to buy when people are apathetic, right? Like people don't care about Bitcoin. Hot money is into AI bottlenecks, right? You know, Michael Saylor is the bad guy, right? I had a tweet last year. I was like, there's going to be a point where it's going to be, there's going to be a meme. where you can see like Marathon and Riot do pretty well, right? And you can see Michael Saylor is a bad guy. Well, it's playing out like Marathon's outperforming Iron. Riot's doing pretty good on the AI pivot. And MSTR is a bad guy, right? Michael Saylor is a bad guy. You know, that screams Bitcoin's a buy for a long-term hoggler, right? So yeah, I'd be, Uh, if you're, if you're just trying to accumulate for a very long period of time, uh, yeah, buy Bitcoin.

₿itcoin ₿utcher: The only thing I would add, uh, AI guy, I, um, I made this comment in your comments, but I think Sailor, I was ******** about optics with Iron this week. I think Sailor has a optics problem right now and he's running a company that No one doubts he's a smart guy, but his basic business strategy is the, well, the long term compound annual growth rate or CAGR of Bitcoin has been, you know, it was 60% maybe. And, you know, his assumptions are that it's 30 long term or whatever. But