All right, Harry, thank you so much for joining me tonight. I'm really excited to dig in a little bit further with you, talk a little bit about the data center, specifically in Texas. That's been a hot topic over probably the last month or so. There's been some movements with the PUCT and then also with ERCOT, and we're seeing some potential windows of opportunity. We'll get into that in a little bit. And then I did some digging. I happened to notice you guys had a really awesome website called Sealy Forward where you explained just how CleanSpark is being diligent with the resources. Definitely want to touch on some of that as well. And then I did listen a little bit on one of the YouTube videos from one of the Sandersville city town hall meetings, or city meetings, and I heard from one of the representatives there. They did a great job explaining just how CleanSpark is going to be standing up the data center there and went through a lot of those same exact concerns that Governor Abbott in Texas also brought up. So, without further ado, figured we'll jump right in it. So first off, how are you doing tonight?
Good. Good to see you, Ernie. Always a pleasure to sit down and talk about what the latest and greatest is.
Yeah, absolutely. First time doing this virtually. I've had the pleasure of getting to know you a little bit over—it's been a little over a year now, about a year and a half or so. And really excited. You guys have really kind of changed course, if you will, from just a year and a half ago when I first met you two Bitcoin conferences ago. And this past Bitcoin conference, you guys had already announced your plans to kind of move into the HPC data center side, getting into the colocation. Of course, you already have a specific site there in Sandersville that has a contract signed. And before we get into some of these details, I just wanted to make sure that I did point out that nothing that we're talking about on this call is financial advice. All of the forward-looking information and guidance, you can find that in CleanSpark's latest 10-Q. There's an entire section there, so make sure if you're an investor, take a look at that. Tonight I just want to talk about the education, if you will, for the state of Texas and also how CleanSpark is actually positioning themselves.
Yeah, perfect. That's usually my job on the earnings call, but I'm glad that you got to take it on this time.
Yeah. So, for anyone who doesn't know, you guys have one site in Texas in the city of Sealy, and it's in Austin County. And then you have another site which is really broken into two phases in Brazoria County. Geographically speaking, they're both in Texas, but Texas is such a big state. Brazoria County, if I'm not mistaken, is kind of southeast, almost near the coast somewhere near there. And Sealy is kind of tucked in a little bit further northwest of there, if I'm recalling that correctly. So, with those two sites, instead of me talking, maybe you can kind of just let the listeners know what the current status is for Sealy, for Brazoria Phase 1, Phase 2 throughout this ERCOT process. I know that you've given a little bit of clarity, but if you don't mind just giving a little bit more again.
Yeah, certainly. So, I think that I'm going to even take us a step further back and just talk about what we did in order to blaze a trail for ourselves in the AI and HPC data center lane alongside what's historically been a more Bitcoin mining-focused lane for us. And so if you take a look back to October of last year, we really only had Sandersville in our portfolio that was sitting in that 250-megawatt or greater type of contracted capacity profile. And so one of the things that was an important early strategic priority was to add capacity to our contracted portfolio that could support some of these larger, some of these hyperscale-style type of campus opportunities. And so we very rapidly, in October of '25, acquired the Sealy, Texas campus. That came with a signed electric service agreement and facility extension agreement with our utility, CenterPoint, there to grow into 285 megawatts of gross capacity at that location. And then we followed that up in January and February with the acquisition of the similar type of campus assets required to grow up to 600 megawatts of gross capacity in Brazoria.
And so what's important in our decision-making around why those two is that we have a strong viewpoint around East Texas, around the greater Houston area—not because we love the city of Houston, though we do, but because we viewed the transmission landscape within Texas as something where there were going to be certain areas that had a smoother path to energization and others where there's a more competitive environment for the transmission capacity. And East Texas and the greater Houston area we viewed as strategically relevant from a transmission perspective. We're going to come back to that in a minute because that viewpoint turned out to be more accurate than we thought it was. And so that was some of our thinking there.
What we love about both sites is that they were already heavily progressed from an ERCOT review standpoint. The Sealy site had already been through the review process in 2025. That doesn't make it immune to the batching process that obviously is the topic of a lot of what we're talking about, but the study work on the transmission both upstream and downstream from that site all came back all systems go.
And I'm going to talk for the moment just about the pre-August 3rd Governor Abbott directive universe because I think it's really relevant about how we kind of got here. And so two things happened when the calendar turned from '25 to '26. The first is that ERCOT started to engage external engineering firms in order to do their load engineering on a go-forward basis. And so they saw the writing on the wall that they were resource constrained. They needed to have more engineering depth in order to review the application queue that was rapidly, rapidly growing.
The second was the introduction of Senate Bill 6 into law that has some of the low-voltage ride-through concepts, some of the redundant gen concepts, as well as some of the mandate that took the mantle on in order to create a more stringent application process for their large load interconnections in order to get some of the flippers or speculators out of the queue and allocate time and resources to developers who had a higher bar of credibility that were going to be not only applying for power, but also building the data center sites or the industrial sites that was going to ultimately consume that power, rather than somebody who has a plot of dirt and an application and now they think they're going to make nine figures on that.
So that was sort of Q1. Q2 is when the batching process was really solidified and, in a detailed way, specified for how that was going to relate to the applicants that were waiting for approval on the system. And so you've heard, and your audience I'm sure has heard, a lot about this Batch 0 concept. And the way that it works is basically ERCOT designated a segment of their loads to be included in what is called Batch 0, which is the first batch that's going to go through a revamped application process.
Some of the changes in the new process are a fee associated with the application, a cluster study rather than a single-point study. And so what does that mean? It means: what are all the other loads on the system doing? And we need to study them together, because if you isolate a load and study it, and then you isolate another load and study it, there may be cross-dependency that doesn't get surfaced by studying it in that more narrow way.
And so Batch 0 was set to have their designations between what's called base load or studied load. Base load means you get everything on the schedule you think you're going to get it on. Studied load means you might get everything on the schedule you think you're going to get it on, or there might be some changes that come back once the cluster groupings have been studied as a unit.
And so what's really important, and taking this back to CleanSpark's assets, is that our Sealy site was designated in the Batch 0 base load category pending August 7th at the time final approval. Brazoria Phase 1 was in the same designated status, and Brazoria Phase 2, we were waiting to hear back whether or not we were going to be categorized as base load or studied load.
Woke up on August 3rd, and the governor issued a new directive. And that directive was to say that he had additional criteria that he wanted the PUCT to require of ERCOT. In order to set these approvals, there needed to be an additional information-gathering exercise that all of the large loads needed to meet, and ultimately the results of that information gathering would be one of the decision-making criteria associated with final approvals.
PUCT received a letter from ERCOT called a good cause letter that happened the following week. And within that letter, there was a lot of interesting nuggets and nuance in there, but basically it said, "Okay, well, we're going to have two more PUCT meetings, one on August 14th, one on August 20th, and we're going to keep kind of working through this."
I told you earlier we were going to come back to this transmission concept, and on August 14th is where it reared its head, which is that there's really actually two concepts that are being evaluated right now. One of them is the proposal for a 765 transmission corridor update, where new, very high voltage transmission lines are going to be required in order to move power from different places around West Texas and the Permian predominantly, but also South Texas as well. And that plan was receiving some very similar public pushback around eminent domain and land rights. And so it turned out there were actually two issues that sat underneath the hood on all of this, because we had a lot of transmission corridor certainty. And so one of the things that's important is that our sites in the greater Houston area are overwhelmingly not dependent on the 765 upgrade plan.
Now, the whole 765 go/no-go and approval process is going to play a role within the larger cluster study, so there may be some timeline components, but from a physical perspective and an engineering perspective, our sites are not sitting underneath the 765 approval process as a dependency. And so that was a big de-risking choice that we made much earlier, many, many months earlier in the process, came to be validated now more recently.
That takes us to August 20th, and on August 20th we had the most recent disclosure. So, we had the good cause letter before the 14th. Then we had the meeting on the 14th, they talk about the 765s. Then on August 20th, really what was discussed was the governor's directive and ERCOT's request for some exceptions from PUCT to rework their process to be able to in whatever way possible meet the energization timelines that are governed by a lot of the power contracts that are already signed and out there, including ourselves, as well as a more detailed scope on what the RFI is going to look like and what the timeline in the intervening months is going to look like for the remainder of this year through Q1 of next year. I'll pause there. That was a lot. Thank you for humoring a monologue.
Yeah. No, absolutely. So, when you were talking a little bit about the RFI and everything like that, I know that there's an important date coming up next week. There's going to be a subset of potential builds that are going to get some more clarity. I don't believe that the specific companies have been named yet, and from what I understand just from some research I did, that information is going to be given privately to each of the companies. So then it's going to be up to the companies to decide if they want to disclose that they've been selected into this first category, if you will. And then there's another date later on, I believe much later in the fall, where there'll be another—is it December?
Yeah, so the next two dates that are really relevant: the first is August 31st, and that date is going to demark some of the conditional approvals that are out there. These are not final approvals. So, for sake of illustration there, there may be three projects that are all Batch 0 base load, and they all may receive that information. One of them may receive it at the end of August, one of them may receive it in December, and one of them may receive it in April of 2027, and all get the same treatment and all move forward exactly as expected, but it may take that kind of long range of time in order to gain certainty around different projects.
So the other thing that we didn't talk about that was in the original good cause letter from ERCOT to PUCT is this concept of QSA. And so there's really two approvals that you need in order for a project to move to final certainty within ERCOT. First is the large load interconnection approval, and the second is the QSA, which you can think of as like a final technical review, like: did you build your substation to spec? Are the wires clear? Those types of details are checked in that final process.
And so typically, under sort of a normal operating or ordinary course—now granted, there is no ordinary course because the first time we've ever done the batch process with ERCOT, anyone's ever done it, so everything is kind of first of its kind, which, you know, when you take something off the spreadsheet and put it into the field, things are a little different. Typically, you would receive large load interconnection approval first, and you would receive your QSA closer up to your energization date.
What ERCOT is attempting to do in order to maintain energization timeline integrity is ask for the ability to approve on a QSA basis ahead of an approved interconnection request. And you may have heard that there are two QSA dates that are being contemplated in this good cause exception. The first is the August 1st QSA batch, which mentions six projects being included in that. The second is the November 1st QSA batch, and that is indicated to be 17 projects. And the way to think about those is those would be correlated with projects that would energize in Q1 of '27 or Q2 of '27.
Our Sealy asset is scheduled for Q2 of '27. So the thought process—and again, you're exactly right, nobody has been told who is in the 17, who is in the six—but one concept is that we very well may be within the 17 as it relates to our Sealy asset because of the energization timeline and some tea leaf reading. But again, it's all a black box until it isn't. And so we're working very hard in conjunction with our utility, in conjunction with the PUCT and ERCOT, and the governor's office and the other large operators within the state. We're all working together in order to make sure that we're building projects that meet the requirements that everybody's asking for, because ultimately that's the goal: to build things the way that people are going to feel good about them being built.
I was just going to say, one thing that immediately popped in my mind when you guys are doing your due diligence and preparing for the commercial side with your specific tenants, you've done such a great job of ensuring that there's no risk on delivery dates. And I immediately thought after this specific update came out from the governor, I thought, "Oh man, what about those companies that already signed the commercial agreements that had the specific deadlines for energization?" That's got to be pretty nerve-wracking because there's probably some significant penalties, potential breaches of contract if they're not able to do that.
Obviously, you guys have your first contract in Sandersville, Georgia. Everything seems to be very smooth there, and because you were close, in my opinion, just from understanding you're using the same tenant and the exclusivity was signed, that the commercializations could have been ongoing. This is just speculation, but with that being said, you guys hadn't pulled the trigger, if you will. So in my opinion, I think it's actually a good thing that you guys have now the opportunity to get certainty on your energization before moving forward on the commercialization. At least that's how I'm looking at it and reading. I don't know if that's how it is between the different data center landlords, if you will, and the tenants. I'm not sure if there's an understanding there, or if a lot of the negotiations in Texas and other states have went on pause for now.
I think you mentioned specifically what happens if you've signed a commercial agreement with a delivery requirement date and now this kind of rears its head. I think that, one, we're very risk-averse in the way that we do business, and I think that that's a big benefit frankly, just wanting to be relatively conservative in those ways, especially around timeline risk. But I also know that there's a lot of great operators out there who've signed great deals, they're working with great counterparties. And what's really unique about what we're seeing, especially in the data center development space right now, is that this is a pretty collaborative dynamic between landlord and tenant at this stage where, because there is such scarcity and overwhelming demand, the goal is to create positive outcomes.
And so, while I don't think that we would have felt comfortable commercializing ahead of final certainty on these sites, the others that did, I don't think they're terribly impaired in those kinds of ways. And so the demand for compute is so great that if you've got to do a little bit of workout in the meantime, I think they'll be in a good position to do that.
Yeah, I was exactly going to talk about that. A lot of these hyperscalers, they're in a significant need for this energy. So when a macro event like this happens across multiple states, multiple cities and things like that, I feel like there's going to be somewhat empathy and somewhat understanding. And regardless, it's not like this is a one-company type of situation here. This is happening all over. But to your point, the fact that you guys are more risk-averse, I still like that as a shareholder. I'm not looking for what's going to happen at the end of this week; I'm excited for what you guys have put out in your presentation material and where you guys are heading.
So, kind of going past the Texas site, unless is there any other tollgates or milestones that investors should be looking out for for any of these sites?
No, I think that the two that are really relevant are the two we mentioned: one's August 31st, one of them's in December. Those are the times when I think we'll see material updates from the ERCOT side of the house. We're going to be forthcoming about those, but my guess is there's going to be continued significant dialogue around it. But from a process standpoint, those are the two dates that we're very focused on.
And we're quite well positioned, frankly, to respond to the RFI, to be able to meet the exacting requirements that the governor has requested. And frankly, I think that there's a bit of this that's sequencing, where I think ERCOT and PUCT were going to make a lot of the same requests, already have made a lot of the same requests that we've answered fulsomely. And so I think it's more so about reordering the process a little bit rather than a departure from the ultimate goal.
Nice. And in case no one paid attention to the Easter egg—sorry, I went on a little side tangent in my mind here—you guys are also investigating a significant increase in power capacity in the state of Georgia as well. So that's kind of exciting because obviously a lot of those activities are going to continue to go underway while Texas kind of sorts itself out.
But you mentioned CleanSpark is prepared and has already met all the criteria that the governor has laid out. I wanted to touch base on a couple of those items. I know you guys had an expo there near Sealy where that one site is proposed to be going up, and you guys had almost like an open town hall where you were able to bring some people in and show them some of the equipment. You had, I believe, your infrastructure partner was there as well. Don't necessarily want to go into how that specific event went, but more so the messaging. What is CleanSpark's stance on a couple of the large heavy-hitter items? I'm thinking about water consumption. I'm thinking about noise mitigation, light mitigation, and a few other items as well. Potentially there's an on-site generation aspect as well down the road. Maybe if we could start on the water and cooling. I know you guys are going to do a closed-loop system. For people who don't know, what is the difference in a traditional data center and a closed-loop system type data center that we're seeing a lot of companies moving to today?
Yeah, I'm thrilled you asked this just because I think it's one of the ones where there's some of the most concern and transparently we've got the best answers. So, if you look 20 years ago at the data center industry, the most technologically advanced cooling methodology was called evaporative cooling. And then you had big cooling towers. You would run water through the system, the chips would heat the water, and it would turn into steam and evaporate. And then you would put more water into the system and keep doing it. That used a ton of water. It was a heavy consumptive item, and frankly it was also very expensive to buy that much water. And so, like with many things, a free market delivers good solutions.
And so, initially as a cost-saving mechanism, the closed-loop direct liquid-to-chip technology became the new leading technology for data centers. And so that's what we're going to be employing in Sandersville. And it's going to be standard to all of the designs that we run, is not to use evaporative cooling. It's expensive, it's inefficient, it creates a big new risk to the project that frankly we don't need to take on because we've got great technology that does this really, really well and doesn't consume that natural resource. So, to put it into perspective, a 300-megawatt data center will use less water one time at fill-up than an Olympic swimming pool. One Olympic swimming pool one time, and the number is a lot less than that, but I'm being conservative.
It's crazy to see some of the comparisons that are out there. People compare golf course consumption, I've even seen some compar—
Yes, almond farming. I saw that the other day on FinTwit. So it's very interesting just the various industries that are just consuming so much more water and—
Orders of magnitude. Why are the almond farmers not being audited right now? Where is the large questions posing to them on how can they do a better job of reusing water? Anyway, I'm being facetious here, but yeah, when you look, when you compare the newer generation data centers to a lot of the other industries, it's minuscule on the amount of water consumption and excess water usage over time compared to so many others. So—
We also use water on an ongoing basis because there are bathrooms and kitchens and break rooms. Those are significant water usage where we use less than 10 households worth per day. So one Olympic swimming pool one time, 10 houses worth of water on an ongoing basis. And in many cases, we represent huge percentages of current tax receipts for places where we plan to build. And so the water issue just isn't borne out in the engineering or the data. It just isn't. And I know that it's an issue that folks feel really passionate about.
I love living in a world where conservation is a priority as a society. Data centers are great for that. We deliver tons of tax revenue and use very, very, very little natural resources. It's exactly the kind of thing you would want in order to keep a more rural, pastoral, or beautiful place the way that it is, because the alternative is thousands of new housing developments, rising property tax rates. I saw on Twitter or on X—I'm never going to get it right—they were looking at Loudoun County, which has the highest data center density of any county in the US, and their property tax rates have gone down 30% over the last 10 years. They have the best school system, they've got more dollars per kid, and lowering the property tax rates all over the place.
