← Blockspace: AI & Bitcoin
Blockspace: AI & Bitcoin
IREN’s Nvidia Partnership and 5 GW Plan, Hut 8’s $9.7B Hyperscaler Deal, Strategy May Sell BTC to Own the Short Sellers
01:11:43
Loading summary
Transcript131 lines
- [00:00]
A
Foreign. What is going on y'? All? Welcome back to Block Space Live. We are back after a quick reprieve. We were out Monday and Wednesday as Will and I were at the consensus conference in Miami. But we are coming back in force, Charlie, because today we are wheeling and dealing. We've got a number of AI deals to cover, including Iron's path to 5 gigawatts and a new partnership with Nvidia. We've got a segment on Hut 8's most recent contract with an undisclosed hyperscaler. We'll also be covering some of the quarterly earnings from both of those companies. Then we have Khan Farahani of Luxor on to talk about their April hash rate look Back series. We will then have Francis Corvino of Lygos on to talk about what else but Michael Saylor saying that he's going to sell his Bitcoin. And also some reactions from Coffeezilla on the stretch market dynamics. And then we will close the show with a few more quarterly earnings, one from Terra Wolf, another from Core Scientific including a non materialized Hyperscaler deal which a lot of investors were expecting that sent its stock price south. And then we will close on Cypher Digital announcing a new Hyperscaler tenant as well. It's third AI deal so far, Charlie
- [01:21]
B
Box Space goes live Monday, Wednesday, Friday at noon Eastern featuring quick hits on the latest in bitcoin, bitcoin mining, AI and emerging tech. Make sure to like and subscribe. If you're on YouTube, hit the bell to get notified when we go live and this turns into a podcast. Shortly after it we wrap up the live broadcast. You can find the podcast streaming wherever podcasts are found, Spotify, Apple, etc. If you like this, you'll love our newsletter comes out every Friday. This morning I wrote about Michael Saylor selling bitcoin. We're going to hit this one like three times today. This show is brought to you by CleanSpark. Nasdaq listed ticker CLSK. More on them later on the show. Let's kick it off Colin with.
- [02:09]
A
The hash rate index update. Tom and we'll be a little quick on this because Kahn is going to cover a lot of this data for us here in about 15 minutes, but kind of more of the same. Network hash rate is piddling along at 976exahashes. We had a downward difficulty, a negative difficulty adjustment of 2.3% recently. We were about halfway through this current epoch and it's looking like we're going to have a 1% positive adjustment. So Kind of more of the same. Bitcoin's hash rate and difficulty keep see sawing back and forth between crests and troughs. Part of that has to do with the fact that bitcoin has been surging recently. We're at 80k at the time of recording. It touched 82k this week, but it hasn't really done too much for, for bitcoin's hash price. If we look at the three month, we're definitely up. We're down from these lows right here, but we're still kind of chopping along at $38 pet hash per day. Lot better than what we've seen. You know, we're up about $10 from the all time low, but still pretty thin margins for a lot of the miners out there still operating.
- [03:14]
B
Yeah, not a lot to say. This difficulty adjustment, really nothing to say. Bitcoin prices up a bit. That's really all.
- [03:22]
A
I mean, that's all she wrote. Folks, we're hoping for some bullish comments from Khan when he comes on, but we'll save those for when we get to them. We'll go ahead and dive right into some of the news items we have got. Like I said at the beginning of the show, we have a lot of deals and earnings to plow through.
- [03:39]
B
I apologize if the listeners wanted to hear about bitcoin or AI. You're only going to get data center deals. So just wrap in. It's just stocks, stocks, stocks, tickers, tickers, tickers, AI deals, hyperscalers.
- [03:54]
A
That's the show today, 100% and this. It's almost like it rains and then it pours. We had a bit of a drought in terms of deal flow for a while over the first quarter. But now as these companies have reported their quarterly earnings, it looks like they were saving all of their ammo for then. And we have some pretty exciting developments. So we'll start with IRON here. Iron. As most of y' all know, one of the front runners in the AI pivots among bitcoin miners. It recently reached an agreement to partner with Nvidia for and has an eye for 5 gigawatts of AI infrastructure build out. So the Nvidia deal is really interesting. The agreement will be serviced by air cooled Blackwell platforms from Nvidia, deployed for roughly approximately 60 megawatts of iron's existing data center capacity at Childress, Texas with a total contract value of approximately 3.4 billion. The deal also comes with a really interesting equity warrant. We've seen this, we saw this also with Terra Wolf with Google and the deal that it backstopped with Fluid Stack for one of Terra Wolf's first AI deals. But this warrant gives Nvidia the opportunity to purchase as much as 30 million shares of iron at a price of $70 per share, roughly 2.1 billion in investment if fully exercised. What's interesting about this, this is basically like a long dated call option. Right now it's about 14% out of the money. With iron at roughly 60%, the price popped after this deal and then it sold off. It just briefly touched $70. So it's out of the money right now. But if Iron can really expand to its ARR target, which it's targeting, $3.7 billion in annual run rate revenue by the end of 2026 with these AI deals, it's hard to imagine that that $70 price target for Nvidia would materialize and that they wouldn't be in the money on this offering. One last thing to note before I toss it to you, Charlie, for second takes and before I cover some of the other aspects of Iron's Q1. Iron said that it also has also entered into a purchase agreement for more than 50,000 Nvidia B300 GPUs and it expects to expand its total fleet to 150,000 GPUs by the end of 2026, with deployment phases through the second half of the year at its air cooled facilities in McKinsey, British Columbia and Childress, Texas. Super interesting because as Dan Roberts pointed out, I didn't get the chance to run the numbers on this. This would make them one of the largest cloud providers in the world.
- [06:29]
B
This is Iron coming for coresci. For cores. I mean it's, I mean I remember the Mike, the Mike, what's his face tweet a year or so ago saying that Iron is going to price up to basically Core's market cap. This is. It indicates that that's. It could head there. You know a couple things. I watched the whole Iron earnings call. You know, there's only a few counterparties which could be, you know, on the other side of this deal. Iron conspicuously did, you know, left that out. But Dan Roberts said, quote, you know who it is, and he said additionally it wasn't an anthropic equivalent company, but I think there's only a few companies out there who can service as our counterparty. So pretty, you know, pretty interesting take there as it's. I would say it's like, you know, limited to a handful of counterparties. So looks like Ivan caught one of the big fish.
- [07:32]
C
Yeah.
- [07:32]
A
And just to highlight the megawatts they have in the pipeline. Currently there's 480megawatts of AI capacity on track for. For utilization in 2026 with an additional 1210 megawatts in build for 2027. And now with their current pipeline capacity, they are looking at 5 gigawatts for 2028 onward, which is pretty amazing. A few more things to note. They also had some acquisitions last quarter that I think really flew under the radar. The first one to target is this nostrum deal for €165 which makes that to about 194 million USD at current exchange rates, which will add 490 megawatts in Spain, which really interesting. And a gigawatt plus development pipeline in that site. You know, we saw Mara be the first to move into the European market with their Xion acquisition. We'll see what they make of that in terms of actually developing AI and HPC data centers. But now Iron saying we're going to be moving into the European market as well so far in terms of at scale AI development because there are other companies like Hive that have smaller operations in Northern Europe. For instance these. These are the largest deals that we've seen from public miners on the continent. And it's interesting considering Europe is really kind of falling behind the US in this regard. Go ahead, Charlie. You look like you want to.
- [08:49]
B
Yeah, I mean, look, I'll jump on. I'll kick the dead horse. That Europe really did shoot the golden goose and has managed to somehow manage to snatch defeat out of the jaws of victory. As it comes to energy and abundance, however, is the Iberian Peninsula the new West Texas? I don't know. I mean, look, I would. It's way more fun to travel to south of France or you know, you know, Andalusian Spain than it is or Basque country, that is to West Texas basically. Do you get to choose between Mars or Paradise? And I would much rather go visit an AIHPC mine where I can also get tapas 24. 7.
- [09:31]
A
That'd be nice. You might have to deal with some rolling brownouts. I don't know.
- [09:35]
B
Could be.
- [09:36]
A
We actually covered that on the pod a while back. But again that's.
- [09:40]
B
That's just Europe, man.
- [09:42]
A
It's just the European lifestyle, guys.
- [09:43]
B
Six.
- [09:44]
A
Six hour work days. 30 hour work weeks. Anyway, a few more notes before we hop on to hut 8. Another really interesting acquisition from Iron. They. They purchased this company called Mirantis which is a independent. They're a Founding independent software vendor and a partner of the Nvidia AI Cloud Ready initiative. So I think that's kind of in tandem with this Nvidia contract. So there are pre existing relationships with Nvidia. I would imagine Iron wants to tap those specifically for networking their clusters. So Mantis will or Mirantis will give IRON the software capabilities to to make sure these clusters are acting in ways that are most optimal for their customers. And the acquisition was specifically listed by IRN to strengthen how COMPUTE is deployed, managed and operated for their customers. And it's directly going to support the delivery of the Nvidia AI cloud contract as well. That 60 megawatt rollout that we talked about, that was for 625 million. Interesting, because I think the takeaway from this is Airin just basically as I understand it, purchased the software layer of their business. So biggest question for me, when I've talked to a lot of these operators when they decide to either go to the PowerShell model where they're just building the infrastructure and managing the power versus doing what Aaron's doing, building that and also running the GPU clouds, my question was how are you going to have the expertise to make sure you're running that cloud properly? And it seems like Mirantis will be part of the answer there for Iron. Last bit to note, big quarter for their AI revenue. They pulled in 33.6 million from AI services versus 3.13 points. Excuse me. Versus 3.6 million in Q1 of 2025 and 111.2 million from Bitcoin mining versus 141.2 million in Q1 2025. So the AI revenue has now totally offset what they've lost in terms of Bitcoin mining revenue over the last year. And you'll see that as we go through these other earnings, AI revenue really revved up last quarter for a lot of these companies. All right, unless you have something to say, Charlie, I'm going to jump into HUD 8.
- [11:50]
B
Let's do HUD 8.
- [11:52]
A
Alrighty. So HUD 8 also coming out with a massive deal, a $9.8 billion lease with an undisclosed tenant. Quoting directly here from BlockSpace, Hut 8 said Wednesday that it signed a 15 year lease for 352 megawatts of IT capacity at its Beacon Point AI Datacenter campus in Newchess County, Texas. With a base term contract value of 9.8 billion, first quarter revenue rose 226% year over year to 71 million. So another massive deal. Again, the Tenant was not disclosed for this. So we have no idea who it is. But needless to say, 15 years at 9.8 billion is a nice deal for Hut 8 to add to their business. The total contracted capacity now that Hut 8 has for AI data center IT capacity is 597 megawatts across two campuses with an aggregate base term contract value of 16.8 billion. Of course, the other campus is Riverbend and Louisiana in Hut 8's fluid stack deal and that's for 245 megawatts at $7 billion valued over 15 year. And Hut 8 is targeting Q2 2027 delivery for that site.
- [13:04]
B
I have a hilarious correction my last take which I, which I misattributed to Dan Roberts. It was actually Asher Ganute on this call. But because we had Hut 8 and I were next to each other, I misspoke. So let me redo it again. When asked, when Ash was asked who the counterparty is, he said there's only a few handful of companies out there who it could be. But you know who it is. It's not an anthropic, it's not an open AI. So yeah, congrats to.
- [13:32]
A
But he said do you know who it is that it could be a lot. Is it another Fluid Stack deal? Is it Google?
- [13:41]
B
Oh shoot. I would have to pull the transcript. The caller actually named four or five companies of which I believe Fluidstack was on there. And, and, but again, go go check for yourself because I just already demonstrated that I, it was a little misspoke a bit but you know, so we actually have it pretty narrowed down because it wasn't in the four or five companies that the caller who Asher responded to, he said it's none of those, but it's one of the others. It may. I get the sense that it might be like one of those big like Amazon types or who like you know, doesn't sign a lot of the, you know, these, these deals, hasn't signed a lot of these deals with these, these miners now a HBC factories. So it could be like you know, of the remaining handful of folks who haven't come out and said anything. So eyes on that deal.
- [14:32]
A
Yeah, I'm really interested to see who they announced for this lease. A few more points to, to hone in on before we close this. The lease, as with other Hut 8 leases, includes a 3% annual base rent escalator. So they're, they're trying to get ahead of inflation in terms of their operating costs there and also there are three five year renewal options that could bring the total contracted value to 25.1 billion if exercised. And Beacon Point, just to put a bow on, this is the first phase of a one planned 1 gigawatt campus. Hut 8 said it has an interconnection agreement for 1 gigawatts currently and expects initial energization in Q1 2027. Expects initial data hall delivery in Q3 2027. And one last thing to note, looking at Hut 8's quarterly earnings for Q1, Hut 8 reported 71 million, up from 21.8 million a year earlier. The company posted a net loss of 253.1 million, which is mostly related to 295.7 million in unrealized losses in Bitcoin. Obviously, Bitcoin got hammered in Q1 and Hut 8 held about 1.3 billion in cash in Bitcoin. As of quarter end, they reported 66 million from ASIC compute, AI, cloud compute and traditional cloud solutions. Vast majority of that is bitcoin mining. I'm not totally certain. I don't think they have any build AI revenue currently. I could be wrong about that. And 3.7 million from power generation and managed services and 1.3 million from colocation services. So overall, for Both Iron and Hut 2, massive deals, very strong quarters. And again, we will have more on some of the other miners after our guest segments.
- [16:11]
B
Speaking of guests, it's time to bring them on. Conferani, let's do this. Welcome to Blockspace Live Con.
- [16:20]
D
Thanks so much. Charlie and Colin, great to see you both. Thanks for having me back on the block.
- [16:25]
A
Always, man. Always a pleasure. Mr. Ferrani. So another month, another Luxor look back. We will get the Luxor look back summary up here on the screen in just a second. But to set the stage, Khan, I just want to ask, you've been doing these look backs for a while. You've been knee deep in bitcoin mining data at Luxor for almost two years now. So you've run the numbers a lot. What was the biggest thing that surprised you in this most recent lookback series?
- [16:54]
D
Yeah, good question. I think there's a lot to dive into for April, but the big picture message was we saw an overall relief in mining economics throughout the month of April, but that didn't necessarily translate to network difficulty in hash rate. So there's much more to the story than what the surface level, bitcoin price and hash price improvement tells us. So, yeah, we'll get into some of these details right now.
- [17:23]
A
Yeah, so let's hop right into it. What stands out to you immediately, like in terms of what data points do you think are most salient for people to understand? The month that we just exited with Bitcoin mining, definitely.
- [17:35]
D
So two things stood out throughout this month when it comes to monthly averages for mining metrics. Overall, we saw a slight rebound and improvement in mining economics.
- [17:48]
E
Right.
- [17:49]
D
Hash price up over the month around 8.5%. Bitcoin denominated hash price up 2.7%. Average bitcoin price up by 5.6%. We did see from month start to month, month end, it rose around 10% and average difficulty down around 2.7% throughout the month. Now, at the same time, when it comes to Wall street and Tradfi, we saw a strong month there as well. We saw that bitcoin price recovered around 10%. At the same time, the S&P 500 returned around 10% and the Nasdaq Composite also returned 14%. These are some of the strongest months that these indices have seen since November of 2020. What's really interesting though is a lot of this upwards action was driven specifically by the tech sector. We saw a wide range of earning calls come in throughout the month of April and a lot of them surprised to the upside, specifically in terms of forward guidance when it comes to capital expenditure for AI infrastructure. This reflected specifically on the semiconductor space. We saw that the ICE semiconductor index gained around 36% over the month compared to Bitcoin at around 10%, the S&P at around 10%.
- [19:14]
A
And Nasdaq don't compare it to. We're tired of this, man. Don't, don't be, don't be comparing Bitcoin's returns to all these highfalutin AI stocks. I'm just joking, but I just. When you sent me this chart, I was like, oh man, another quarter. Another underperformance from Bitcoin over the last year.
- [19:28]
D
Right, Exactly. And what's cool to see is that obviously some of the publicly traded mining companies caught this bid as well. Our Bitcoin mining stock index, Hash Rate Index's Bitcoin Mining stock index was up 27% over the month. And the reason is because some of the constituents in this index are directly exposed. They're transitioning towards AI infrastructure companies. So they were able to catch some of this bid. Now that's all the good stuff. There's a lot of green to talk about in there. But when it comes to network parameters, specifically network difficulty and hash rate, we continue to see a decline in average network difficulty. And in fact at this stage we are now four difficulty epochs in to being below the one zeta hash threshold for network hash rate. The equivalent in terms of difficulty for one Zeta hash is 139.70 trillion. And we've now been below that level for coming up to 50 days.
- [20:32]
A
So are you telling me, Khan, that we have officially exited the one zeta hash era, at least for now? I mean.
- [20:38]
D
Yes, yes, we are firmly below the 1 zeta hash level at this stage for four consecutive difficulty. So that I think is the most hash.
- [20:53]
A
That's one of the more wild parts about this for me is just we've covered it on the last time you were on talking about the lookback series, but just how absolutely volatile Bitcoin's difficulty has been. And we're at this point where really hash rate is, if not plateauing, actually falling off the map. And you had a few notes about this in the summary you sent me specifically about a quote here. Marginal hash rate came offline at $30. Hash price has not returned even with hash price now near $39. I know it's hard to say, but is this, do you think this is just miners quitting for good, saying it's just not worth it anymore? Do you think some of this might be miners moving equipment to lower cost jurisdictions shuffling hash rate around? I know it's impossible, but if you could try to look into your crystal ball for us. What do you get a sense of why that hash rate isn't coming back online even if mining economics have improved?
- [21:47]
D
Great question, and we don't know the answer, but thankfully there are a few data points which can give us a roundabout indication as to what's happening. And over time we will find out, so we will have an answer. The most immediate data point that I would look at is the most recent difficulty adjustment which took place on the 2nd of May. This was a downward day difficulty adjustment of 2.3%. So what this immediately tells us is rather than seeing marginal hash rate come back online throughout the month as hash price improved, we didn't see this in March. We did see this, right? We had the whipsaw in difficulty. There was a massive drop in difficulty and then a very quick rebound essentially back up to where difficulty was. The reason was because as soon as hash price popped in March, marginal hash rate responded very quickly. Within 24 to 48 hours we saw that hash rate back block times went back down and difficulty responded by adjusting back up. We did not see this on the 2nd of May in fact, we saw difficulty went down. This immediately tells us that marginal hash rate is not moving back in and it points towards a continuation of structural decommissioning. Now, you know.
- [23:04]
A
Oh, sorry, no, finish your thought. I'll tack this on at the end.
- [23:08]
D
So that's one point. But there are other data points that we can look into in the near future as we turn into Q3. Once we get to the end of June, beginning of July, we'll be able to see in our next quarterly global hash rate heat map whether there have been any moves in hash rate and whether trends have changed. If the US or North America loses some hash rate and we see other jurisdictions in emerging markets, low cost power regions gain some hash rate, that will also give us some indication as to whether there has been a structural decommission and relocation towards other locations. And the final point I'll make is the hash rate forward market does indeed give us signals around this stuff. What we saw throughout the month of April, throughout this month, is the way that future hash rate contracts traded. We can back out implied expectations around hash rate and difficulty growth. And what we saw was that expectations for difficulty fell throughout the month of April by around 4% on average. The forward market is also opining that hash rate and difficulty growth should be slowing down and going negative over the near future, spanning the next six months.
- [24:25]
A
Yeah. So basically hash rate traders are looking at the current state of the network saying there's not really that much room for growth. All the megawatts in the US at least the bulk of them, are going to be going towards AI. So we expect forward hash rate to be worth much more than current spot hash rate. That makes a lot of sense to me and I want to zoom in on that. But I just have one more question. I know there's like a lot of noise obviously with the AI pivots in terms of what's going on with Bitcoin's hash rate in the ASIC market. There's the argument that some of this is marginal hash rate that can't come online anymore or which is not worth the effort because maybe the margins aren't even fat enough at 39 or dollars per PETA hash per day. But how much of it do you think is also these Bitcoin miners decommissioning for AI? I mean, we were looking at the Q1 numbers for iron and we're going to look at the others for core. Scientific is I think the other big anterowolf in terms of AI revenue. The big gainers for Q1 and they've really ramped up their AI revenue and it showed that they really started scaling down bitcoin mining. So how do you factor in that noise? Do you think that's playing a significant portion of the current hash rate decline we're seeing?
- [25:34]
D
That's a good question. Yes, I do believe that that has an influence and a role to play and the reason is because we track the proportion of network hash rate that publicly traded mining companies represent of the overall global hash rate that has gone up over the past year or so and I think it peaked at around 40% of global network hash rate since then over the past quarter especially it has declined somewhat. But that points to the fact that this has a significant influence as compute revenues. General purpose compute revenues are more attractive for these data center companies. They are now transitioning their hardware which inadvertently creates a lot of opportunity for the rest of the mining network. As ASIC market dynamics are now changing, secondary markets are flush with new supply prices are down. So this gives smaller, more modular mining operators a lot more opportunity in finance flexibility to now re engage in the mining market as some of the larger players are transitioning away towards general purpose compute. So yes, I do think that we're seeing this clearly in the data.
- [26:47]
A
So Khan, as a closing question I want to go back to the hash rate derivatives market and you mentioned in the post that the regime flipped where USD outperformed BTC contracts. This to me seems like a logical, you know, a logical result of the fact that difficulty has been whipsawing everywhere. Is that your take on it? And also what other, what other trends can we unpack from this other than like you said, forward hash rate expectations are much higher than spot currently definitely.
- [27:21]
D
So if you take a look at our look back, we provide two time frames for rolling hedge performance and coincidentally the month of April 2026 is a nice window to look at because we're halfway across now towards the next having and our time frames align actually. So on the one hand we have a trailing twelve month time frame looking at results from May 2025 to April 2026 which is what you're seeing right there. And then we have the since having time frame which aligns nicely because that's from May 2024 to April 2026. So on the one hand we have a one year time frame and now we have the two year time frame as well. There's two key things that we can take out from these pictures. The first is regardless of contract denomination or hedge horizon, hedging against hash price has made sense over the past year. And over the past two years, whether it's US denominated or BTC denominated, whether it's one month or five months, we see all across the board that hedging hash rate has made sense. So that's a clear takeaway for mining operators from this picture. The second is there's a difference in the regimes that we see between the two time frames. So if we look over the past two years, since the 2024 halving, Bitcoin denominated hedging strategies stand out. And the reason is because we've seen secular growth in difficulty from the 2024 halving onwards up until Q4 of 2025. We hit the one Zeta hash milestone in September and then bitcoin price peaked in October of 2025. But then the story changed. Since then we've seen an unexpected decline in average network difficulty. Difficulty has gone back down. We're now firmly below the one zeta hash milestone and at the same time bitcoin price action has become uncertain and volatile. These results show directly in those two pictures. The one year time frame USD denominated hedging stands out. The two year time frame bitcoin denominated hedging stands out. So we've moved from a difficulty growth regime into a bitcoin price uncertainty regime now. So that's the second key takeaway for astute sophisticated mining operations that continuously monitor the forward market. These are clear signals for dynamic decision making. At some point the vibe shifted from growth and difficulty to unknown bitcoin price action. And miners that followed these signals and acted on them would have stood to make significant gains versus spot mining. So I think those are the two main stories to take out from this.
- [30:16]
A
Well Con, thank you for those takeaways man. Hopefully when we have you on here next time hash price will be above 50 of Brett/ per day. With all of these miners coming offline, we're holding our. We're crossing our fingers out there for all y' all fingers. Yeah, Con, thank you. Thank you so much for joining man. Have a great weekend.
- [30:33]
D
Take care guys. See you.
- [30:36]
B
Always love Con. We have our next guest, Francis Corvino from Ligos. But before that a word from our sponsor, CleanSpark.
- [30:45]
A
We are CleanSpark, America's Bitcoin miner. A publicly traded company with the largest
- [30:52]
E
operating hash rate powered entirely by by
- [30:56]
A
self operated infrastructure across four states. This is our proof of work and
- [31:04]
E
we are setting the standard for what's next.
- [31:07]
A
Learn more about the Intersection of energy and bitcoin@cleanspark.com.
- [31:16]
B
all right, a brief break. We're not going to talk about mining stocks. We're going to get. We're talking about the other stock that matters, MicroStrategy. So let's bring on Francis Corvino. Francis, welcome back to the show.
- [31:28]
E
Hey, guys, I never realized your Clean Spark commercial was Harry.
- [31:33]
B
Yeah, like, who else has a voice like that?
- [31:36]
A
Yeah, right.
- [31:38]
E
It's memorable, for sure.
- [31:40]
B
Exactly.
- [31:41]
A
Yeah.
- [31:41]
B
So, you know, this week, Saylor said he might sell some bitcoin, but. And we're gonna.
- [31:48]
A
We.
- [31:48]
B
We can talk about that. But we were jamming before the show here, and I think that it'd be super fun to talk about the latest Coffeezilla and Jeff Walton video where they go toe to toe on, you know, is. Is Stretch a Ponzi? Is it this? Is it that? And maybe there is actually a nuanced middle ground. Francis, you said that you have some good critiques of some dumb things Coffeezilla has said. How do you.
- [32:24]
E
I. I think that it. It was a really interesting example here. Right. Because when you leave the sphere of, like, influencer meme coins and you end up chatting with people who have, you know, real professional experience, the fact that you're an influencer starts to kind of show itself. I would also say, you know, if you are doing a bunch of, like, niche videos on areas of the economy, areas of life that, you know, there aren't a lot of people with deep expertise in, and you produce videos and you show those videos to 6 million people, and there's no one there really to critique you, you can start getting away with a lot. So I think that the clear, indecisive nature in which he lost this debate really just at times making stuff up really kind of draws a lot of ire on, like, the rest of the work that he has produced historically. Like, I look at it now, and I'm like, huh? This guy was, like, clearly talking out of his ass on an interview that I watched where I know some of the background on it. And now what's to say about what else you kind of made up? So, yeah, I kind of compiled, like, a top five things that he said, which I found to be the most. The most awesome and hilarious.
- [33:36]
A
What tops the list, like, in terms of talking out his ass, what things stand out? Specifically?
- [33:41]
E
My favorite is, without a doubt,
- [33:45]
B
a
- [33:46]
E
bond is guaranteed to pay back. And then he sort of pauses, and he looks at Jeff Walton, and Walton looks back at him, and he's like, sort of, well, I mean, first of all, just like elementary truth, right? A bond is a way in which you are pricing risk. So technically speaking, stretch is equity, but at the end of the day, it's sort of performing like a bond, right? And there is no such thing as a bad bond. There is only a mispriced bond. If you look at stretch and you say, okay, there is a greater than 13% chance that this thing is going to go default, go maximum loss defense, maximum loss event per year, then it's a bad bond, right? Like, you have a higher percentage chance that you are going to lose greater than 13%, then you are going to retain it. If it's vice versa, then it's a good bond. Jeff tries to describe this a couple times. He's trying to say this is really just a way to quantify the risk associated with this instrument. Coffeezilla, like, not, not willing to listen to it. Right after that, he sort of jumps into another sort of, I would say, faux pas, right, where he says, okay, well, these stretch coupons are paying, you know, what, like, you know, 12, 13%. So at 12 to 13%, you're going to need bitcoin to appreciate by greater than 12 to 13% per year in order for this to sort of be like a profitable financial move. The reality is that is not the case at all. Because he's calculating that as if there is $1 of Bitcoin per dollar of sort of like stretch liability, right? That's not the case. The capital base is significantly larger than the percentage of debt. So bitcoin really only has to go up, I think like 2.35%, 2.3%. And for Strive, I believe it's like five and a half percent. But again, just like a mathematical issue, which leads to just a fundamentally different conclusion. He then takes that 13% number and he multiplies it by, I think, 70, right? And now what? We need $36 million Bitcoin in order for this to be a good trade. Look, I personally think that $36 million Bitcoin is, is a reasonable outcome here, but I'm not everyone. And the math is still wrong, right? Like, again, if you do 70 years of bad math, it doesn't make your math any better. It actually like, ironically kind of makes it worse. So that was another one where it's like, okay, like we're on our third prong now and I still got a couple more. But yeah, I mean, another thing that I think that sort of went untouched, right, Is, is there is area in Which I think maybe we could, we could consider like, hey, if I'm an investor in one of these different instruments that MicroStrategy has, there are some things that are really worth talking about. And I think that is the historic nature to which Michael Saylor has been able to have just no real covenants, no tripwires over this debt stack to. If you are a creditor and you see something happening at the company that you don't like, there's not a lot of ways for you to sort of like crumble the house of cards. So if you're an equity investor, that's great because you have a relatively responsible leverage which has not a lot of trip wires on it, which will force you to rapidly unwind that. And if you are a, if you are a debt sort of like debt type investor here, though that is worth looking into. You know, why are there not so many covenants on this particular. It's like historically low, low coverage here. There's definitely things worth pointing out, but those didn't support Coffeezilla's real sort of message, so he didn't bother to bring those up. Or he never really kind of researched it enough to figure.
- [37:15]
B
I get the impression that Coffee Zilla really doesn't what he's talking about. I mean, because I've had some exchanges online which are like. Well, Coffeezilla knows like when meme coin, you know, promoters clearly scam or rug something. But that's kind of the limit of his knowledge to understand, you know, covenants or have you say trip wires. To be able to pulled it requires a better understanding of traditional finance. Francis, what are. Because I want to get into this
- [37:43]
E
like I just want to say one more thing. Those people that like I work in, you know, a repo business business at the end of the day, like we take your bitcoin collateral and we sell it. There are covenants associated with some of the things that we do. But the complexity of those covenants are not, you know, such that I would consider myself a deep expert. There are people who are artists when it comes to covenants and Coffeezilla is just frankly not one of those people and hasn't even talked to any of those people in order to really put out a cohesive and interesting, interesting piece, I would say.
- [38:12]
A
Yeah. And sorry Charlie, just to interject with this really quickly, you know, it analyzing meme coins and like when someone's rugging or something is really kind of boils down to a combination of like blockchain analysis and almost psychology and social analysis. Because like you're looking at the way they conduct their social media, the way that they communicate with the community and you're also looking at what's happening with the funds on chain bitcoin. Treasury company is a capital markets play. And if you don't understand capital markets well and you're trying to argue against something like Stretch, there are plenty of, you know, things that you can criticize. I was just kind of surprised to see him go to bat so strongly for something that seems so clearly out of his wheelhouse. And just one more note on what Francis said. Going back to like the one to one bitcoin to, to, to USD liability here with Stretch. Charlie ran the numbers for today's newsletter. Annual dividend stretch is nearly, is just south of 1 billion. Strategy has 60 plus billion, 65 billion in Bitcoin. So in terms of like actually being able to furnish this debt and we'll get into this in a second in terms of sailors saying that they'll probably sell some bitcoin down the road, right? There's plenty of Runway. It's just whether or not the market likes him selling his bitcoin or not. So anyway, Charlie, I'll throw it back to you.
- [39:27]
B
Yeah, I am kind of curious, Francis. None of us are covenant artists here, but what would one of these covenants or tripwires or conditions might, what one, what would one of these like look like? What would an investor in Stretch or these financial products want that does not exist, do you think?
- [39:49]
E
I mean based off of my understanding, I would say one thing that they would really want to look for is like leverage ratios, right? Like what is the maximum amount of leverage that MicroStrategy can take before I can say, hey, you know, that's too much, I want my money back.
- [40:03]
A
Right.
- [40:04]
E
And it doesn't actually appear that MicroStrategy has clearly outlined what their leverage maximums are. They could hypothetically take a huge percentage or even all of their Bitcoin and just do an asset backed loan, double down and buy more bitcoin or buy, I don't know, ripple, right. Like they can do pretty much whatever you want. You don't have a lot of discretion that you, you can, you can push on the MicroStrategy team on what's happening with your capital and you really don't even have a lot of guardrails.
- [40:34]
B
I mean I, I will say Micro Sailor seems like the kind of person who would get very, very levered, but that's it.
- [40:42]
E
I don't know about that. Like I, I Think another thing I was talking about, and this is where we sort of turn into Coffeezilla and it gets a little bit more tabloidy. But like, if I was to, I, I've lived through like a couple of these, you know, credit crunch cycles within the like, bitcoin crypto industry as a whole. And one thing that, like, if I could have traded, you know, a macro indicator of like our industry based off of sort of like a human interaction, one of the things that you would see very quickly or when people started to take out significant amounts of leverage, whether it be sbf, whether it be Mashinsky, you started to see like a real degradation in their mental capacity and also their health. Right, right. Like SBF was a vegan who got fat. Mashinsky went from like, looking not necessarily healthy but not, not terrible to like, this guy looks like he, I know someone close to Machinsky who worked with him. That said, at the point that things were about to go south in Celsius, Mashinsky looked like he was already going to die. That's a real quote. That isn't really the case with Saylor. Right. Like, Saylor's still on the road. He's hitting the road, shows like he looks good. He doesn't have an untenable amount of debt. There's a lot of like, you know, hubbub around his stack. And I think he's a competitive guy who feels like, willed to respond and maybe even sort of like jump out of his skin. But like, I don't necessarily, I guess I'm ultimately, you know, a bull on microstrategy from the standpoint of like, I think Micro, MicroStrategy and Michael Saylor, generally solid, reputable investors. Yeah, I don't see the same activities.
- [42:16]
A
And going back, Sorry, Charlie, just to interject here, going back to the debt portion of it, I haven't looked at overall in terms of their convert footprint, but a lot of the convertible notes that they did to buy bitcoin was really in the early days. You saw a lot of that early on in the second to last bull market. 21, 20, 22. And most recently the majority of the buying power has come from equity issuance. As far as I can tell, a lot of, lot of dilution of, of, of strategy stock, common stock and also issuing these new preferreds. But Charlie, you seem to have a question. I, I kind of want to get into what Francis thinks.
- [42:56]
B
No, I, I, I don't have a question. Going to say that if, if Poly Market could set up some like a, like A physiognomy founder physiognomy index. We would, we could trade that. It. I'd be very excited. I, I did, yeah.
- [43:08]
E
Quickly, Colin. I mean, the, the converts are in the money though, right? So like, we aren't really in a period of time where we're like, those are additional equity issuance as of now. Yeah. At the end of the day, like, it would be dramatically and outrageously economically irrational.
- [43:25]
B
That's really good.
- [43:26]
A
I'm glad that you, I'm glad that you pointed that out. That's a really good point. Because when you look at all of that, you kind of come to this conclusion, like you said, where there's nothing really messy so far and, and strategy's debt structure. That would have alarm bells. But I think a lot of people. Let's just move to the big news item for a final few questions here. Francis. So on their earnings call, Strategy announced that or Michael Sailor said that they are going to, quote, probably sell some bitcoin. And he said specifically to send a message, which I thought was interesting. And later he said that that message was specifically. He kind of hinted that it was towards short sellers. It said, if you think that I have to continue diluting to pay back these dividends, I'm going to rip your wings off. I think was.
- [44:13]
B
That's a direct quote. That's direct quote.
- [44:16]
A
I actually, I had to ask Charlie in, in the, in when I was editing the article, I was like, you put this in quotes. Is that, did he really say this? Because it was, it was really kind of a no punches pulled moment. But here, here's our coverage of it here on Block Space. And this kind of raised a whole hullabaloo. You know, obviously we have this cute little meme in the article where it's like it went from never sell your bitcoin to buy more bitcoin than you sell. With all of that context. Francis, I'm just curious now that strategy's opened the door for selling in order, if they need to, in order to meet dividend obligations. What do you think about this move and how do you think the market will react to it?
- [44:52]
E
I mean, it like the, what Charlie said earlier, right. Was like microstrategy, Sailor maybe seems like the type of guy to take on excess leverage. Well, this is sort of a practice in deleveraging, right. And it shows a willingness and an openness to be financially responsible. So I think rationally, long term, the market should respond to this positively. Right. Like the idea of holding a line like Caroline Ellison tried To hold the line with CZ and Binance on the price of FTT and by trying to hold that arbitrary line, when that line got broken, the, you know, the viewpoint of the market was such that FTX was now broken. Right? Like it became a binary outcome. Saylor can and should at the top of the market or near, you know, 80k is not a bad bitcoin price, guys. I know it feels like a bear market, but like anyone who's telling me that Bitcoin at 80k is a bear market has like really just not been here very long. Like I remember they bought last year at 100 back in 2017, talking about like random privacy tech. Like if you told the two of us back in 2017 that we were going to be at Bitcoin 80, 80, whatever K and Bitcoin is a bear market, like it's just ridiculous, right? We're, we're in a bowl, like we're at 80k. Things are going pretty well for Saylor. Like he decides to take a little bit of leverage off the table just to show a willingness and, and openness and not make it a binary outcome of I'm a lever long and I have a liquidation price and when I get liquidated, I get liquidated and everything goes to hell. Right? If he just shows the market that he isn't a stubborn old man like you might think he is, and in fact has, you know, a level of scrutiny that he can put on his own company, take some leverage off the table. I, I think that takes away a lot of the argument that a short seller might be trying to make. Right? Like you can show everyone the ghosts of crypto's past. Like all these people that tried to hold this hard line and now you can continue to sort of paint Michael Saylor in that same light. Like he's unwilling to sell bitcoin. He'll never sell anything now. Look, he's selling some bitcoin, he's delevering a little bit. That automatically I think really weakens the story of the short seller trying to superimpose Micro Saylor on the ghost of spf.
- [47:02]
B
I love it. Can I have you react to this clip I've got of the Coffeezilla interview? Because I had it pulled up because it's only 10 seconds long because it's, it calls back to what you said earlier, which is like the coffee zilla really kind of out of his depth. Here's the clip on the bonds thing.
- [47:22]
C
They will repay. If you hold on to it to term, they will repay your Principal. Right. They're guaranteed. Well, guaranteed, you know, subject to some other things, but especially if you have Treasuries, they're going to repay it. Right. Whereas here, you don't have that same thing. And so I think. And Michael Saylor will even, you know, say, like, hey, he'll compare it to a bank account. Okay. It's not a bank account. Even in your own disclose, in Yalls own disclosures, which you put below tweets, y' all say, hey, this is not a bank account. It's not a money market fund.
- [47:55]
A
Right.
- [47:55]
C
But why is it being preferred to bonds and money markets? I think it's bank accounts. I think it's fair when it's not
- [48:02]
B
that I think it's fair to compare
- [48:03]
D
it against other credit instruments in the market.
- [48:05]
B
You could go look at other perpetual preferred equities that are issued by any other bank.
- [48:10]
A
I mean, they're almost saying two different things here. In a way, it's like Coffeezilla doesn't understand that the interest rate for Stretch is so high because investors understand that there's risk involved in it. Right. I mean, exactly.
- [48:24]
E
I mean, again, I want to impress this quote, which I learned as a kid. There is no such thing as a bad bond. There's only a. A mispriced bond. Right. Coffee Zilla is not making the argument that this is a mispriced bond so much as he's making the argument that they're using the wrong terminology, which he doesn't even seem to really understand, which is a really bizarre way to go about.
- [48:44]
A
And also saying, like, why aren't investors just putting this into a High Yield savings account or U.S. bonds? And it's like, because the High Yield Savings account. High Yield Savings got like, what, like, two.
- [48:54]
B
A gratuitous 3%. Yeah.
- [48:59]
E
Yeah.
- [48:59]
A
It's interesting. I think you're right.
- [49:02]
E
There isn't better debt to buy out there. Like, I'm not. I don't. I don't understand. He seems to have put everyone in the position of, like, we're defending, you know, the. The 12. 13. Right. Like, that's not really what we're doing. We're just sort of pointing out that I'm. I'm not sure you understand that this is a way to quantify risk. It's not, you know, it's not a grenade. Right.
- [49:23]
A
Well, for instance, that's probably good for a closing question here. You know, when we look at Stretch, you said the rate's 12 to 13 now.
- [49:30]
E
Yeah, I think. I think it's 13 now, okay, well,
- [49:33]
A
I mean that's crazy high. You know, when we, per Jeff's point, when we look at other interest bearing instruments, I mean, do you think that current rate is sustainable? How long until that return? What do you think it would take for investors to look at that return and say this is just not worth it to put my money.
- [49:49]
E
Well, I mean it's almost to the point where it's exactly sort of what we were saying. Again, there is no such thing as a bad bond. There is no such thing as bad risk. There's just mispriced risk. So the rate of the risk will continue to go up as the risk increases, as the market sees it. Right? There is just the only thing that is happening right now is the market is pricing that risk. There is, I think some outside opinions here, which is like there are still dollars which are unwilling to access Bitcoin specific risk markets, whether it be, you know, Bitcoin tang tangential credit markets and that is arbitrarily increasing the reflected risk. So like, but I don't necessarily think that that's the case so much in this capacity. Right. Like 13% is probably just a good way to reflect the risk associated with this instrument based off of its position in the, in the debt stack.
- [50:42]
B
So I guess final question, one word answer is stretch. Risk priced appropriately.
- [50:52]
E
I'll leave that up to the viewer. At the end of the day, I'm not a financial analysis here. Like I, I'm just reacting to a influencer is far out of his fear to the point that even a, even a layman can look at it and be like, wow, this was, this was a, an epic embarrassment and an opportunity for a really incredible slap back.
- [51:13]
A
I guess Francis is ever the professional man. He doesn't want to say yes and then have his pants sued by someone who lost their mortgage on stretch because they didn't know how to manage the risk properly with that for sure.
- [51:27]
E
What are you trying to do to me, Colin?
- [51:31]
A
Hey, it was you giving a really diplomatic answer, I think. Well, that's what markets are beautiful for. Eventually, like you said, Francis, the market will tell us whether or not that risk was appropriate or not. But we've taken up enough of your time. Francis, thanks so much for joining, man. Fun segment on this super interesting stuff with Stretch and really appreciate the take by take takedown of Coffeezilla. So have a great weekend, man.
- [51:52]
D
Bye.
- [51:52]
B
Rock on. And now a word from our sponsor, Luxor.
- [52:04]
A
This episode is brought to you by Luxor's Commander Bitcoin Miner Management Software built for enterprise operations. Commander gives real time fleet monitoring bulk remote commands across your entire fleet. And Intelligent Miner. It's an automated profitability engine that runs every five minutes to adjust your power settings to live hash rate markets from Luxor and energy markets. Ercot backtests show 10% more profitability with intelligent mining versus binary mining. If you want to try it out. Commander Pro is $100 per megawatt or a 25 basis pool fee adder. Roughly half the cost of competitors. But you can try it for 60 days with a free trial. Go to Luxor Tech forward slash Commander to get started.
- [52:46]
B
And now we go back to stocks. We never left stocks.
- [52:51]
A
This is, we never left. This is just the, you know, OG block space enjoyers who remember the bitcoin stock show and that's basically what we're doing here today.
- [53:00]
B
So this is just, that's just, this is just now our main show because look, crypto markets are boring S and P all time high every other week. So yeah,
- [53:13]
A
Khan was coming on here like bitcoin lagged S, P and, and semiconductor and disease. I was like again, of course they did.
- [53:19]
B
Oh my God.
- [53:20]
A
Anyway, all right, we're moving on to quarterly earnings from Terra Wolf. Terra Wolf posted 34 million dollar in revenue as AI and HBC lease income offsets bitcoin mining. So the big takeaway from Terra Wolf's most recent quarterly earnings, they pulled in $21 million of AI in HBC lease revenue and 13.4 million BTC mining slash other while net, while its net loss widened to 427.6 million from 61.4 million. We'll get to that net loss here in a second actually mostly paper losses tied to its warrants with Google for its fluid stack lease. We'll get to that here in a second. But a few more things to cover on this Q1 now. The AI and HPC revenue for Terra Wolf is 62% of its total revenue. And so it has become as far as I can tell. And we'll look at core scientific next. They might be kind of neck and neck, but it seems to me like Terra Wolf now has the largest portion of its revenue coming from AI and HPC with its AI business lines. And the majority of that is coming from its core 42 rollout in Lake Mariner which is at 60 megawatts of HPC capacity as of the end of March 31, 2026. Now addressing the elephant in the room. $427.6 million loss. This was largely non cash based loss and 216 million of it comes from a again non cash on paper loss from Google's warrants on 73 million Wolf shares. So per its deal with Fluid Stack, Google is backstopping the revenue for that partnership that Terra Wolf has with Fluidstack. And, and with that deal came $73 million Wolf shares or warrants for $73 million Wolf shares with the warrants priced at a penny each. This is what you get for playing with the big boys. You're going to have to give a little more than what you get back in some cases. And because of this, you know, this is why we're seeing this net loss. Terra Wolf Tera Wolf stock has absolutely surged. It's up 109% year to date and over 700% in the past year. With that dramatic appreciation, outstanding warrant liabilities ballooned in value creating a large non cast accounting loss even though no cast actually left the building. So just wanted to get that out in front because I think people see those numbers and they're not really sure, you know, where that comes from. There's a lot of accounting games and a lot of, a lot of line items that get factored into something like that. A few other points From Terra Wolves Q1 Before I toss it to you, Charlie. They acquired their Hawesville, Kentucky site. We covered that on the pod a while back. It's 480megawatts of grid connected power. They say that they are nearing completion for their CB3 site in May 2026. CB4 and CB5 are on schedule for delivery and rent commencement in 2026 at Lake Mariner. And it's Abernathy JV with FluidStack targeting 170168 megawatts under a 25 year lease. One with a target delivery date of Q4 2026. So if we see, you know, the first phase of that fluid stack operation come online at the end of the year, you're going to see even more impressive AI and HPC revenue from terawolf relative to Bitcoin mining. With that taking a breath, I don't
- [56:38]
B
have, I don't have a ton of commentary on this. You know, Tera Wolf sitting at like 11 billion total market cap I believe, which puts it, I would say maybe in the mid, mid range for like the aihpc, like former bitcoin miner hyperscalers. So like up year to date 100 which indicates it like it flipped.
- [57:02]
A
It's flipped core scientific.
- [57:05]
B
Oh really?
- [57:06]
A
Interesting enough, yeah, yeah. So 11 billion is actually it's just shy of. Well, it's about half of iron, which is at roughly 20 billion. But to your point. Yeah. So I guess you could say, yeah, kind of firmly in the middle there because really I think that the companies to watch right now for the AI HPC things release the front runners. As far as I can tell, it's iron. It is hut 8. It is cipher, Terra Wolf and Corsi, you know.
- [57:35]
B
Yeah. So it's in, it's in the, it's a. If you have the frontier AI models and you have the frontier former bitcoin miners now turned AHPC factories, at what point do we stop calling them former bitcoin miners and just talk about.
- [57:50]
A
I think once you get where Terra Wolf is, if your revenue is 50% plus AIHPC, you're an AI HPC data center company now. Right. So it's not even, I don't even think fair to really call them bitcoin miners turned AI factories, at least in Terra Wolf's case. But I think we'll leave that there, Charlie, and go ahead and move right on to coresci here because kind of a much different take on their Q1. And I think part of this is due to the fact that there were investor expectations going into this call that Core Scientific might announce a deal, but they did not actually announce a deal. So core Scientific shares tumbled 9% as no new AI HBC deals materialize. This is coming straight to us from Block space. The firm announced 61Q1 2026 revenue of 115.24 million and an adjusted earnings per share of negative $1.06. But leading up to this earnings call, as I was saying earlier, Core Scientific actually Core Scientific announced that it was expanding its Oklahoma site in Muskogee by 440 megawatts. That comes after it acquired land from or acquired a site from a company adjacent to its Muscogee site. The announced expansion Muskogee is now eyeing 1.5 gross gigawatts, 1 gigawatt leasable after the 440 megawatt edition. Again, it purchased this from Polaris DS LLC in a site near the Muskogee site that is one of Core Scientific's primary sites for AI and HPC revamping. The development has already begun on an initial 82.5 megawatts in the building. And it is targeting first delivery in first half of 2027. So this Muskogee deal was announced before earnings. And I think a lot of people thought that Core Scientific was going to announce a tenant, but they didn't end up doing that. Now, that being said, there are still some silver linings. Core Scientific plans to scale its Pecos, Texas site to from 300 megawatts to 1.5 gigawatts. Interesting part about this is they have 300 megawatts currently at the Pecos site. They have 300 megawatts of additional capacity from their utility provider agreed to and signed for. And so that leaves the remaining 900 megawatts to be behind the meter because they did mention that they are developing a gas pipeline for behind the meter power here, but they didn't specifically state where the rest of that 1.5 gigawatts would come from. But all signs point to it being some sort of behind the meter off grid thing with the gas pipeline. We've seen this with other data center providers launching or, you know, rolling out these turbines as they look for as any amount of megawatt they can to power these sites.
- [60:54]
B
I think what you said is what it is. Oklahoma is actually currently halfway through potentially passing a bill which requires AI data centers to quote, generate their own power. Now that's kind of a reductionist. It really just means that they have to be behind the meter. That power plant that they're nearby that Polaris is next to is natural gas. It does land a pipeline. And like I see this as a major behind the meter deal and, and potentially power generation play. Would love to get someone from Polaris. I got some connections there as I am an Oklahoma native and I know that site well. I remember when they stood up 200 megawatts of, I would say, not very new ant boxes back in 2023. And so now they're doing big deals with CoreSci. Congrats to the Polaris team and a
- [62:01]
A
few more notes on coresight Charlie before we close out this segment. Going back to the AI revenue portion of things, Core Scientific's AI revenue was 77.5 million last quarter. That's up from 8.6 million just a year earlier. Bitcoin mining revenue fell to 30.1 million, down from 67.2 million for Q1 2025. So core scientific firmly within the same camp as Terra Wolf at this point. The bulk of their revenue is coming from AI and HPC deals. And I think at least when we're looking at the last few years, this was the big question mark of like, when were these miners going to turn these, when were these miners going to actually be able to turn these AI and HBC outfits on? And how would they actually manage the phasing out of their bitcoin mining and supplementing it with this. So far, Iron Terawulf and Corsi have done it very well and there have no been, there have not been any massive hiccups I think in their revenue profiles. And again Terra Wolf and Coresight the first ones to get past that 50% revenue from AI and HPC. A few more notes that I thought were really interesting from the earnings call. Core Scientific sold 2,385 Bitcoin for 208.3 million in order to fund Capex and other cash needs. Not surprising, many bitcoin miners are doing this. Bit Deer sold their entire stack. Mara said they're going to start selling. Cypher opened the door to start selling. I believe Bit Farms did as well. So no surprise there. They also closed a 3.3 billion project bond for one of their Core Weave sites or for the Core weave contract at 7.75% with approximately 2.9 billion in net proceeds after cost. They also, we, I think we already just. Oh, we didn't discuss this one. It's just like the quarterly calls this, this quarter were just filled with like and we did this and we did this and we did this and we did this and we're just now announcing it today. It's like every single miner decided to just throw all of the deals that they have on the currently at the wall. But they acquired, they acquired the Hunt County Texas site before the call and announced the Muskogee Polaris acquisition. That Hunt County Texas site is going to be a greenfield site, I believe together accounting for roughly 700 million of 2 billion in plan. 2026 CAPEX and sorry, last note, Charlie. And then I promise I'll shut up. A exclusivity agreement has expired for the Pecos and the Muskogee Muskogee sites. Now Adam Sullivan didn't say this exclusivity agreement was with Core Weave specifically. That would be the kind of logical reasoning I think if you're looking at who they had an exclusivity agreement for these two sites. But it might not be because Sullivan pointed out in the call that the additional capacity is quote, outside of our current contract, end quote. That could mean that it's not within the current core we've contract and core we've have to do another contract for that additional capacity. It could also mean that they're recording another hyperscaler. But anyway, that exclusivity period has lapsed and Adam Sullivan on the call said that they are engaged with currently three hyperscalers for those two sites. So like I said, deal flow kind of lulled in the first half of this year, but we're really starting to see it vamp back up in Q2 and as we approach Q3.
- [65:30]
B
Yeah, I am kind of surprised that we didn't see these deals announced the previous quarter because I mean everybody is clamoring for these. You've got to wonder if these deals were in the pipeline over the past 90 days to six months, what are the current conversations they're currently having? Because the trend is up and to the right. There is no decline in appetite for these megawatts. Anthropic just signed a deal with XAI to use their Colossus site. Like that's how compute constrained these people are. They have infinite money and finite gigawatts. I mean so I would imagine that the next quarter's calls are going to be what we just had, but bigger. So that's my prediction. How do I bet on this?
- [66:19]
A
So yeah, make some, make some poly market prediction markets.
- [66:23]
B
Make some poly market prediction markets. Or you can just I guess buy the stock in the regular market like a normal market participant. I have no idea. Okay, we got one more stock deal from Cypher. But before that, a word from our sponsor. Lygos.
- [66:44]
A
Hedge funds are getting liquidated. Is your Bitcoin safe? It's not just Bitcoin's price drying up. Big whales hedge funds and lending desk are reeling after the 10, 10 and 2, 5 liquidation events. Counterparty risk is rampant. So it's more important than ever to understand who actually controls your Bitcoin. And with Lygos Finance, that is always you. With Lygos you will not set yourself up to be the next FTX or Celsius victim. If you're working with another Bitcoin backed lending provider, please do yourself a favor. Work with Lygos. They are our preferred Bitcoin backed lender. At Block Space. They use Bitcoin native smart contracts to secure your stack and make sure you are always controlling your keys and no one else. With Lygos you always know where your Bitcoin is. There's no wrapping, no bridging, no rehypothecation. Get competitive rates for as low as 10% APR. Go to Lygos Finance to learn more. All right Charlie. To wrap up the marathon, no pun intended, they're called Marin. Now of quarterly earnings and AI deals from Bitcoin miners, we have Cypher providing their Q1 2026 business update which also includes an AI deal undisclosed tenant, much like HUD 8 and I wonder what, I'm curious what that's about. I'm wondering who's asking whom to stay, you know, under the radar here. I imagine their partners are saying please don't disclose this yet for a number of reasons, I would imagine. But anyway, the, the takeaway here is that Cypher currently has no AI and HPC revenue. They had $34.8 million in revenue from their Bitcoin mining arm in Q1, but they also announced a third AI deal. Again, no tenant was announced, but they said that they signed their third data center campus lease with an investment grade hyperscaler tenant, bringing their total portfolio of HPC capacity to 907 megawatts of operated and contracting across three signed leases with approximately 11.4 billion in contracted revenue across base terms of 10 to 15 years. Cypher also closed a $2 billion high yield bond at 6.125% to fully fund its black pearls construction through completion. Also close a $200 million revolving credit facility. And yeah, it seems like those are the high end items. And overall kind of thin in terms of the details we got on this new AI lease. But Cypher clearly confident enough to come out and announce it during their Q1 financials. And like we've just been saying, it seems like all of these miners waited until their financials to announce some of these massive deals. Like you said Charlie, it's just really curious to see what we'll get over the next couple quarters. But overall, strong quarter for a lot of these miners. You know, even you could say with. Even in Core Scientific's case, sure their stock fell after no deal was announced, but they're hauling in the most revenue from AI of any of these bitcoin miners, as you would expect considering they were the first bitcoin miner to announce one of these deals with Core Weave.
- [70:00]
B
Yeah, I mean is there a single miner who's like down after these quarterly, after this week of quarterly announcements, like maybe just the play is go long pre all the earnings reports and then close the week after. I don't know. You know, for all the stock traders out there, let me know in the comments why this is a bad idea. I think that about does it, Colin.
- [70:27]
A
Yeah, I think that's it. Slew of deals. Appreciate y' all slogging along with us. I have to say, just to cap it off, really exciting to see some of these deals start materializing in. I mean we went for a long time without having anything to talk about with the AI pivots from Bitcoin miners and it just seems like we were drinking from the fire hose this week. So hats off to all the teams. You know, I think it's easy to look at these deals and almost see them as obvious in hindsight, but massive amount of work goes into making one of these materialize. So I would love to be a fly on the wall when these deals are being struck, but maybe for another lifetime.
- [71:03]
B
Maybe we'll just pick back up where we left off next week. Otherwise, thank you all for listening to Blockspace Live. This turns into a podcast. Read our newsletter newsletter.blockspacemedia.com and make sure to like and subscribe. Subscribe Our socials are popping off right now. If you like mining stocks, my name Mining Equity News, our Twitter is the place to be. You can catch us announcing deals and major stock movements there, so make sure to follow us on Twitter, currently known as X. See you all next week.