Nano Cap Podcast

Dark Horse Technologies CEO Joel Block outlines pivot from Bitcoin mining to AI infrastructure via powered shells.

27 min Listen to the original Episode page
Bitcoin MiningAI InfrastructurePower DevelopmentTennessee Valley AuthorityDark Horse TechnologiesCorporate MergersHigh-Performance Computing

Summary

Joel Block, CEO of Dark Horse Technologies (DRK), discusses the company's evolution from US Bitcoin and HUT 8 to its current focus on AI high-performance computing (HPC) infrastructure. He argues that pivoting from Bitcoin mining to AI HPC requires significantly higher capital expenditure for cooling and uptime, rather than simple equipment swaps. The company targets sub-50-megawatt sites in the Tennessee Valley Authority (TVA) grid, partnering with local co-op utilities to leverage underutilized power infrastructure. Block describes a development model where Dark Horse secures power and land, builds 'powered shells,' and hands them to tenants who own the GPUs. The immediate priority is securing a reliable first tenant, while maintaining optionality for future uses beyond AI due to national security considerations.

Episode sections

  1. 0:01 – 3:41

    Introduction and Speaker Background

    Host Will Cherlin introduces guest Joel Block, CEO of Dark Horse Technologies (DRK). Block discusses his non-linear career path, starting in fixed income at Credit Suisse, moving to college counseling, and then into crypto and power infrastructure.

  2. 3:41 – 5:33

    Entry into Crypto and US Bitcoin

    Block recounts hearing about Bitcoin between 2012-2014 with skepticism. He became an early investor in US Bitcoin, founded by Asher Astrojanute (CEO of HUT 8), and joined the company in 2021, focusing on the power and energy aspects of mining.

  3. 5:33 – 9:54

    Corporate Evolution to Dark Horse Technologies

    Block details the 2023 merger of US Bitcoin and HUT 8. After a career reset, he joined Cathedral Bitcoin (TSXV) in 2025, which merged with Sphere 3D in June 2026, leading to a rebrand as Dark Horse Technologies, combining legacy mining assets with power/land development.

  4. 9:54 – 14:44

    AI HPC vs. Bitcoin Mining Economics

    Block argues that pivoting from Bitcoin mining to AI HPC is not simple equipment swapping. While AI revenue per kilowatt is higher, the capital expenditure (CapEx) for cooling, uptime, and tenant requirements is orders of magnitude greater, requiring rigorous risk-adjusted capital allocation.

  5. 14:44 – 18:27

    Community Engagement and TVA Market Strategy

    The company focuses on sub-50-megawatt sites in the Tennessee Valley Authority (TVA) grid, partnering with local co-op utilities in communities that have lost manufacturing jobs. This approach leverages underutilized power infrastructure and provides economic benefits to local stakeholders.

  6. 18:27 – 20:45

    Development Model: Powered Shells and Operations

    Dark Horse’s thesis is to secure power and land, then build 'powered shells' (potentially modular/prefab) and hand them to tenants who own the GPUs. The company plans to outsource day-to-day operations to trusted third-party providers rather than building a massive internal ops team.

  7. 20:45 – 23:18

    National Security and Optionality

    Block views power as a scarce resource and emphasizes maintaining optionality for future uses beyond AI, citing national security implications. He notes that while centralized governments might seem efficient for AI, the ecosystem is complex, requiring flexible infrastructure strategies.

  8. 23:18 – 27:08

    Current Goals and Investor Relations

    The immediate priority is securing the first tenant, focusing on their financial health and reliability. The medium-term vision is building a repeatable platform to bring power online in underserved areas quickly. Block highlights the challenge of raising capital in the current credit environment.

Assets discussed

AssetKindStanceContext
BitcoinBTCcryptoneutralstart hearing about Bitcoin? And were you right away, the second you started hearing about it, you're like, okay, I'm a sovereign individual.

“Stance” is how the speakers talked about the asset, not a recommendation.

Claims worth verifying

Joel Block heard about Bitcoin between 2012 and 2014 and viewed it with skepticism initially.

Unidentified speaker (speaker_1) ·3:41

How to verify: Cross-reference with Joel Block's LinkedIn profile or public bios for career timeline; verify via SEC filings if he discloses crypto investment history in Dark Horse's S-1/F-1. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

Asher Astrojanute is the CEO of HUT 8 and founded US Bitcoin.

Unidentified speaker (speaker_1) ·4:21

How to verify: Check HUT 8's official website or investor relations page for executive team; verify US Bitcoin founding via corporate registry or news archives. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

US Bitcoin and HUT 8 merged at the end of 2023.

Unidentified speaker (speaker_1) ·5:59

How to verify: Search SEC EDGAR for HUT 8's merger proxy or press releases from late 2023 confirming the transaction. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

Cathedral Bitcoin was listed on the Toronto Stock Exchange Venture Exchange (TSXV).

Unidentified speaker (speaker_1) ·6:39

How to verify: Check TSXV historical listings or SEC filings for Cathedral Bitcoin's public status and ticker symbol. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

The merger between Cathedral Bitcoin and Sphere 3D closed on June 1st of this year (2026).

Unidentified speaker (speaker_1) ·6:59

How to verify: Verify via SEC EDGAR for Sphere 3D or Cathedral Bitcoin's merger completion filings dated June 2026. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

Dark Horse Technologies is currently focused on sub-50-megawatt sites.

Unidentified speaker (speaker_1) ·8:57

How to verify: Review Dark Horse Technologies' (DRK) press releases, website, or SEC filings for project specifications and capacity targets. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

The Tennessee Valley Authority (TVA) sells power to approximately 153 co-op utilities.

Unidentified speaker (speaker_1) ·15:04

How to verify: Check TVA's official website or annual reports for the number of member co-ops; verify against TVA's service territory data. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Modern data centers with closed-loop circuits use less water than local golf courses or pools after initial pressurization.

Unidentified speaker (speaker_1) ·17:47

How to verify: Cite industry reports from data center cooling technology providers (e.g., cooling tower manufacturers) or EPA water usage studies comparing commercial data centers to municipal water uses. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Dark Horse Technologies plans to build 'powered shells' and have tenants own the GPUs.

Unidentified speaker (speaker_1) ·19:05

How to verify: Review Dark Horse Technologies' (DRK) business model disclosures in SEC filings (S-1/F-1) or official press releases regarding their development strategy. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

Anthropic released an S1 document recently.

Will Cherlin ·25:43

How to verify: Search SEC EDGAR for Anthropic's S-1 filing; verify the document type and release date. For recent events, verify 2026-08-19 through 2026-11-19. Use another period only when the speaker explicitly names it.

Primary source

Do your own research

  • Verify the June 2026 merger details between Cathedral Bitcoin and Sphere 3D via SEC filings or press releases.
  • Check the current financial health and reliability of Dark Horse Technologies' first tenant through public filings or company disclosures.
  • Review the specific CapEx requirements for AI HPC cooling and uptime versus Bitcoin mining infrastructure from industry reports.
  • Confirm the sub-50-megawatt site locations and partnerships with local co-op utilities in the TVA grid.
  • Assess the 'powered shells' development model and outsourcing arrangements for day-to-day operations.
  • Investigate the national security implications and optionality strategies mentioned for power infrastructure beyond AI.
  • Examine the current credit environment challenges for raising capital as described by the CEO.
  • Verify the historical merger timeline of US Bitcoin and HUT 8 in 2023 through corporate records.
Full transcriptexpand

0:01Will Cherlin: Welcome to the NanoCap Podcast. I'm your host, Will Cherlin, and I'm excited to have you join me on this journey through the world of nanocap companies. For the past decade, I've talked to thousands of business owners across various industries. It's been an enlightening experience revealing the vast tapestries of entrepreneurship and innovation that thrives beneath the surface of mainstream media coverage. Each week I will interview different leaders of nanocap publicly traded companies to learn about their history, what they do now, and their future. As a reminder, none of this information is financial advice and it is all for entertainment purposes only. Let's start this week's episode.

0:38Unidentified speaker (UNKNOWN): Hi

0:38Will Cherlin: everyone, this week we have Joel Block, CEO of Dark Horse Technologies, Stock Ticker DRK. Dark Horse Technology puts overlooked power to work by developing AI infrastructure to earn more from every megawatt in partnership with towns that host it. Joel, thanks for coming on the podcast.

0:53Unidentified speaker (UNKNOWN): Well

0:54Unidentified speaker (speaker_1): appreciate you having me today. Thank you.

0:55Will Cherlin: So you start off at Credit Suisse, kind of a traditional finance background, working your way up with that. And then you make a big pivot to helping like students around the world get into universities. What what how did that just come across your desk? What made you want to do the big pivot? And I'm kind of curious more about what what made made you want to go in that direction.

1:16Unidentified speaker (speaker_1): Yeah, no, it's it's a good question. It's one of because it's not necessarily a linear traditional path in a career. I think I think it starts with who I am as a person. I have a passion and love for business. And I've always been a bit industry agnostic or industry opportunistic where there were was opportunities that may lie. And I think what really fuels me as I've kind of grown through my career and building different businesses is the ability to work with teams and put them together and maybe get the best out of teams or out of individuals within a team that they may not see within themselves. And so that can apply across a lot of different industries and business. And so while I I come from a kind of traditional finance background on Wall Street and started my career in fixed income and at cards Carde suisse, I always kind of knew in the back of my mind I'd be a bit more entrepreneurial. I'd probably get into some smaller size businesses, help grow them, and help solve a bunch of challenges. And so that kind of led to an opportunity to lead and grow a sector that was fairly fragmented at the time. The college counseling business in the United States had just under 10,000 college counselors throughout the country. And a lot of these people were in local communities and maybe worked with five or 10 students. They were friends of family and those people. pieces. And the the way I viewed it is I thought that there was an opportunity to put a group of counselors together, working across a country or even across the globe,

2:36and kind of share that mind share or wallet share, as you thought, through thought leadership in the space. And I think a lot of those things translated into what I'm doing now as I've pivoted over the last five years into more of power and energy businesses, first in the crypto space. and now as we're transitioning our sites over to AIHPC. Again, you're kind of using a lot of first principles of putting smart people around you, the right types of teams that can grow and an industry opportunity that makes sense. No,

3:02Unidentified speaker (speaker_2): it makes me laugh about the uh fragmentation. I know growing up in Minnesota, I talked to college counselors, I didn't even know that colleges existed outside the Midwest at the time, since it's so fragmented and they kind of have their expertise locality. localities, and stuff. So no, that's really interesting. So you're working at college-wise, you have a finance background. When do you first start hearing about Bitcoin? And were you right away, the second you started hearing about it, you're like, okay, I'm a sovereign individual. This is perfect for me. Or were you thinking, hey, I'm not really sure about this? And then as it kind of is slowly but surely gained gamed adoption, you were like, okay, they actually might be. onto something.

3:39Will Cherlin: Tell me about you kind of first hearing about Bitcoin.

3:41Unidentified speaker (speaker_1): I probably heard about Bitcoin originally um somewhere between twenty twelve and twenty fourteen. But I I probably viewed it like a lot of other people with a lot of skepticism and and kind of from a traditional finance background, could this new currency come about? Could this new ecosystem come about? And then where would that look like on a global scale? Um, I had an opportunity actually in the education business to partner with a young entrepreneur that was based in China at the time. And he was actually my arm in that region to work with students. And we were doing a lot of online counseling, my group of counselors, primarily in the United States, with students that were overseas in Asia, Middle East, and Europe. That person ended up moving back to the United States and trying some different entrepreneurial pursuits. His name is Asher Astro Janute, Janu, and he's currently the CEO of HUT A. 8. Before HUT, he uh founded a Bitcoin mining company called US Bitcoin. Uh him and Mike Co. is partner then and today. And I was actually an early investor in US Bitcoin. So that was actually my first kind of direct investment in the broader crypto space and honestly through the lens of more of the power and energy that they were securing and utilizing. I ended up going to work with them and partnering with them in 2021. So there was a linkage between the education world and my current kind of industry, but um maybe not a again linear route to get there.

4:59Will Cherlin: Yeah, well, and you got the nominative determinism going with the last name

5:03Unidentified speaker (speaker_2): block

5:04Unidentified speaker (UNKNOWN): and

5:04Will Cherlin: everything,

5:04Unidentified speaker (speaker_2): getting

5:05Will Cherlin: into the blockchain space. So yeah, how can you resist? Um, so you're working with them, you're creating these different uh Bitcoin mining, basically rig. rigs, you're mining tons of Bitcoin. Again, how does it come across your desk that you're like, hey, actually, I think I could run my own Bitcoin mining company. There's financing behind it. People are excited about it. Kind of tell me more about that and how it first started turning into the first version of Dark Horse technologies.

5:33Unidentified speaker (speaker_1): Yeah, so I think there was a natural progression i in that. And I I think the key to to realize is the thesis we had at the early days at US Bitcoin was let's grab as much access to power and land and work with communities and use that power and electricity for Bitcoin mining on a near term basis and maybe medium and long term. But there wasn't a necessarily like predisposed end outcome in terms of what we were going to do with those assets. Fast forward in 2023, I helped run the deal and led the merch. merger with HUD 8. So US Bitcoin and HUD 8 combined forces at the end of 2023. And I got a unique opportunity for me in my career to take some time off. And I think that was a nice reset. I got to spend a lot of time with my family. And after working in that industry, which was quite volatile and exciting at the same time, kind of having a reset at the right time, I think, in your kind of career trajectory is helpful. It also allowed me to kind of look at the landscape and look at different opportunities. And so when I came back into the broader industry kind of summer of 2025, last summer, um, it was first with a company called Cathedral Bitcoin. We were listed up in uh Canada on

6:39the Toronto Stock Exchange, their TSXV, the venture exchange up there. Um, and we quickly realized that finding a pathway to the US capital markets and the US exchange were were necessary. And so we put the merger in place with Sphere 3D. That closed about four and a half months ago, just under five months ago, June 1st of this year. And then quickly thereafter, we went through a full rebrand and name change. And so that's what's kind of morphed into what is dark horse technologies today. It was a combination of these two prior companies that were legacy Bitcoin mining primarily, one on the power and land side, one on owning the mining. rigs and using those to generate Bitcoin and converting into what is dark horse technologies today.

7:21Will Cherlin: So, yeah, leaning back to your college-wise days of you taking this totally um kind of fractured market that has these kind of weird geographic things to it. And then now you go and figure out that power is extremely fractured and where power is changed. cheap and available is not necessarily um where you always need it to be. When when did you kind of dawn upon that that hey, there is a use for a lot of power, but the only way you're gonna be able to use all this power is if you start teaming up with communities. Because I feel like the data centers of today are starting to just learn this lesson that you might have learned years ago.

7:57Unidentified speaker (speaker_1): Yeah, no, I I think that's right. And if I look back like six months or a year ago or even 18 months ago. um, there there were there were data centers that were kind of getting announced. And a lot of them were being done maybe, you know, middle of the night type things or in rural areas. And maybe community engagement was not where it needs to be. And you fast forward it to where we are today. And myself and I think a lot of our peers are really understanding there needs to be a community benefit to really work in these areas. It's one thing if you're extremely remote and there's no population nearby, and you'll get that in some parts, let's say of Texas or other areas, then you need to worry about grid and grid strain and making sure there's not too much congestion on the grid. But in areas that are, let's say, a little more suburban or even in rural, but outside of urban areas or suburban areas, you need to make sure that you're engaging with the stakeholders early. And so what we've done, it's it's funny that you connected those dots, prior industry. unrelated at all, but it was a fragmented industry where I saw opportunities to scale. What we've been going after right now is sub 50 megawatt sites that are somewhat fragmented, maybe the afterthought or underlooked sites compared to some others that are looking at much hyperscale and figuring out opportunities to engage with the community and put a structure The and platform in place to be able to scale and turn these into positive economic activities. No,

9:15Will Cherlin: yeah, I see really no difference between you finding some small rant random community like South Bend, Indiana, and teaming up with their university to help them get students and that kind of thing. And then you teaming up with a different small community to help them do another thing that's crucial to kind of humankind, like education and now power. So tell me if I'm wrong here, but this just seems as I I've talked to a few companies that have pivoted from crypto into more of an AI data center plate. Is the economics as simple as these AI companies and hyperscalers and neo clouds are now willing to pay more per kilowatt than we just make at Bitcoin's current price?

9:54Unidentified speaker (speaker_1): I wish it was that simple. But then that's definitely part of the part of the thing. I I think what you need to look at is, you know, as as the CEO of a public company, I have a duty. to my shareholders and making sure that I can maximize return to to the shareholders on a risk adjusted basis. And that that's kind of how I view the lens is how how do I allocate capital? Where do we put capital and spend? And then what is the growth and opportunity we can get as a return for that? And so I I think, you know, on on one one end, yes, there the revenue numbers are much higher. You can charge more, um, whether it's on a computing hour, whether it's on an energy basis, however, you want to do the conversion on AI HPC than you can on Bitcoin mining. At the same time, though, the build-out cost and the infrastructure cost and the capex that you're going to spend to earn that dollar are much higher at orders of magnitude. And so part of part of our lens as we think through prudent capital management is you know, what is the opportunity spend and from a growth. capex standpoint? What is the return on a risk-adjusted basis? And how that compares to other opportunities that we see out there. The way we view the world right now is that the opportunities are vast and very strong in this industry, which is why we are pivoting from the AI E PC. Um, but I think as you look at companies and specifically companies that are pivoting from a Bitcoin mining or crypto background over,

11:14investors are right to ask questions and be skeptical and understand, okay, is it really as easy as you just turn, you know, turn the switch and plug in a different set of computers? And the short answer is no. It is much more challenging. It's a totally different type of data center you're going to be running. Everything from the cooling and how you operate these computers to the kind of r restrictions and and rules that your tenant is going gonna to put on you. in terms of the upkeep and those different requirements. There's a lot of different components that go into running, let's say, the tier two or tier three data centers versus what were the traditional kind of Bitcoin mining or crypto centers.

11:50Will Cherlin: Yeah, and I I I'm curious about from like a culture point of view, as you pivoted over, um, previously, in if you're doing Bitcoin mining, it's almost the only business you can be in where you almost don't have to have. customers. You can kind of just sell them on an open market. You're not beholden to anyone, but just the market itself, type of thing, where now you have like the biggest enterprise customers that have ever existed, type of thing in the history of humans. So tell me about kind of just going from like a sales team to on one end, you're just trying to sell almost the idea that, like, hey, we're gonna use your capital efficiently to create more capital in the open market versus. hey, we actually have a bunch of people who have connections or we have great salespeople who are actually going to get these deals done. Tell me about kind of that culture change.

12:34Unidentified speaker (speaker_1): It's a good point. And I think there's a you got to recognize that there's a number of different models out there. And so even on the Bitcoin mining side, um, there there was parts of our business and my legacy business I inherited and have grown a little bit where we actually did have some customers. We were hosting for third parties that own those Bitcoin mining machines. So in a true sense, if you were self-mining or prop mining, depending on the terminology, meaning you own your own ASICs and you were doing your own Bitcoin mining, in that scenario, you don't need necessarily a sales team or customer service or those components. Then you can move to hosting where you could have a customer. But the question is, do you have one customer at a site or do you have hundreds of customers if you've gone downstream or like a retail? And there's a couple of different models out there on that front. I will say the the kind of requirements and kind of the rules and regulations that tenants and customers will put on as you look to do the AIHPC oftentimes are much more stringent and strict. Um and they're less tolerant necessarily of downtimes or other commitments you need to make. So that's definitely something that you have to take into account. And I think even with that though, there's different models. Like there's these neo clouds that's that oftentimes will own machines, and some of them will have very few customers, and some of them will have a lot of customers. And they may sell things by the hour or by the week or by the month. And it may be more retail or maybe startup type companies versus enterprise. And so I think it's unfortunately it's not like

13:54a one size catch all in terms of the constraints. And I I do think it's important for a company to really analyze what are their core competencies and where can they exceed and and where maybe they struggle. So that to your point, if they are going to take on a task where they've got now a thousand customers overnight from zero or one or two or three, it's a totally different infrastructure from an organizational standpoint. You need different people within your company. You probably need different third parties that are gonna going help to and assist you along the way. And it very much is a different business model if you're looking at that direction. Yeah,

14:26Will Cherlin: and it goes. back to what you said of it's not like you just plu unplug one computer and

14:30Unidentified speaker (speaker_2): plug another computer in, and then you just keep printing cash. It's a whole, it's a whole nother battle behind that.

14:36Will Cherlin: So I know you got a couple facilities in the workslash now in Tennessee and Kentucky. What makes those two states

14:43Unidentified speaker (speaker_2): so alluring?

14:44Unidentified speaker (speaker_1): Yeah, so the our portfolio today is solely in the power market or power grid of the TV. T VA, the Tennessee Valley Authority. And I'd say there's some different um structures that the T VA has in place that may be different than let's say ERCOT and Texas or some of the other grids throughout the country. Uh first, th they operate where they sell their power down to usually local co op utilities. And there's about a hundred and fifty three 53 of these co co-op op utilities. And oftentimes these

15:12Unidentified speaker (UNKNOWN): fragmentation utilities are fragmented. They're in a

15:15Unidentified speaker (speaker_1): local town or a local community or local area. And you really have an ability to kind of work with them at a very local level and make sure that they un you understand what their demands are and their constraints are. And vice versa, you can be a willing partner. And in a lot of these communities, there has been uh manufacturing, light industrial, commercial opportunities that have left these towns in and these communities. And you know. there was a big area in Kentucky where there was a lot of automobile manufacturing that has moved to other states and other regions. And so what you have there is you've got a fairly stable grid and there's already fairly stable infrastructure for energy and electricity to kind of work through it. But you don't necessarily have the commercial activity to replace Place those industries that have left. And so what we've been able to do is work with communities and say, you have power here. It's actually not even stranded. It's connected to the grid. It's just underutilized because you may have had some manufacturing or commute industries that have left. Let us come bring positive economic activity. Let us utilize the power there, become a good citizen, a ratepayer on your grid, which will allow you now to have more revenue in your local utility where you can do upgrades and different components that you may need.

16:24Will Cherlin: So, how much would you say the success. of this strategy is kind of based on the fragmentation almost being a good thing? That you get to talk to the stakeholders right away and you get to talk to kind of the decision maker immediately, opposed to talking to this large bureaucracy that has to go through all these different chain of commands until it finally gets down to the actual person and them actually getting maybe a rebate on their an electrical bill or

16:46Unidentified speaker (speaker_2): however you work with communities?

16:47Unidentified speaker (speaker_1): I would say it's balanced. I think on one hand you get first mover advantage, specifically if you're in a community or a set of communities, they already know you. And you know, the best referral always in business is word of mouth, right? And it's usually when the other side is making that call for you or at least making that recommendation. So we'll be able to leverage that. But my hope as we build this platform out is that we will be able to expand, maybe outside of the existing communities and other communities, or maybe even to different regions, but with a mindset. of we know what it is like to engage with local stakeholders and the local community. And we can almost bring a platform and a scalable model to other areas and other regions that will work at the local level to really engage and make sure these community benefits are realized. You're seeing a lot of backlash throughout the country on data centers. It's probably a big voting topic as we go to the polls next week in a lot of communities. A lot of it is coming from misinformation. I think people have created a narrative of facts that are just not facts. They're not true and it's not the case. There's been a lot around water and water usage. And then you look at the models with the modern day data centers that are on closed loop circuits, they're using less water than the local golf course is or oftentimes the local pool after the initial pressure pressurization. So I think a lot of it is getting education out there. And I think a lot of it is having like a willingness to engage with these communities and

18:07explain to them the benefit and not just, you know, explaining away the negatives that they may see.

18:12Will Cherlin: Yeah, for sure. It's the uh data centers did exist before 2023 and ChatGPT4 coming out.

18:19Unidentified speaker (speaker_2): If

18:19Unidentified speaker (UNKNOWN): the

18:19Unidentified speaker (speaker_2): people are curious.

18:21Will Cherlin: Um, so here's the the million-dollar million dollar question: what do you actually do? How far do you get into the kind of development process of you finding a site? You put up the wall, set up a substation, and let them plug in their GPUs? Do you start helping them finance the GPUs and then have sell it to, like you said, a thousand different customers? What how far do you go along in the development process before you kind of hand it off to someone else?

18:45Unidentified speaker (speaker_1): Yeah, no, it's a good question. And I'd say it's evolving. And I think if you don't even look at the larger players that have already made the switch, some of them are evolving. Some of them are now potentially, you know, vertically integrating even more and owning their own stack of GPUs. Our current thesis right now is that we are going to develop the power and land and create a powered shell. Now that shell may look different at a different site. It may not be a large building. It may be more modular a prefab, an approach that we really like because you can manufacture these units off off-site site where labor is and abundance. abundant in a factory, and then assemble them on site and use local trades in those communities to get that assembly. So I think at minimum, we're going to go to kind of the powered shell or co-location model, as they will say. And likely our tenants will be the ones owning and bringing the GPUs into us. Where it get where it gets interesting, and this happens in all these discussions, is who's going to then operate the data site. center.

19:36Will Cherlin: That's what I was going to ask next.

19:37Unidentified speaker (speaker_1): And and that's a model that we are prepared to navigate. We'll bring in trusted third parties to do it. Initially, we do not envision building a massive team internally to do on-the-ground operations, but there's a number of kind of global, national, and regional providers of data center operations. To your point, data centers have been around for a while. They didn't just spur up overnight. And it's about finding the right partner on that. So it's a service that we'll be able to offer, usually as a pass-through to the tenant in terms of the operations and day to day.

20:05Will Cherlin: So when I I said the intro of your company and I mentioned that dark horse technology puts overlooked power to work. How much of your strategy as you're kind of building out these shells and knowing that you're in these previously industrialized areas sometimes and American industrialization is as hot reindustrialization is as hot as ever. How much of your strategy is kind of keeping it open to hey, we don't know if the next Bitcoin is going to be the next AI is going to be the next everyone needs robot factories? Um, how how do you kind of keep that optionality of who knows what people are gonna need in the future? But we know that they're gonna need structures that have access to reliable great power sources.

20:45Unidentified speaker (speaker_1): I would take it to one step higher than your last comment, if I could. Um, you know, my thesis is power is a scarce resource on the earth and specifically in the US, and the right type of power in the right type of place. And so that that's kind of our first principle. Let's work with communities where there's excess power or opportunities to leverage power for potentially better economic use. Then the the question the next question is what are you doing with that power once you've secured it? Historically, a lot of uh people in our industry and other industries and everything. adjacent have been doing crypto mining or bitcoin mining and now they're seeing an opportunity to support this kind of AI ecosystem. And I think what's also interesting as you think through kind of power and also AI is you kind of get into national security and sovereignty and a lot of other issues that kind of spawn off this. Probably a a different segment a longer conversation that we could have. But I think you know as we sit here building power and data centers and infrastructure within the United States, it's something that we're um fortunate to be part of this industry. Um we take the opportunity um seriously, and but we think it's vast

21:51and large. And that's an area that we want to continue to grow and develop. And I think keeping that optionality to your point is always important as you think through businesses and long-term businesses, businesses. that are surviving for decades, if not longer, and not just kind of an overnight business, where you can keep optionality is where you where you'll have the best opportunity. And so for us right now, securing power, securing land, putting that power to its best economic use in the short and medium term, and then navigating kind of in the longer term where we may And they want to use it is as part of our thesis. No,

22:23Will Cherlin: yeah, you but you mentioned national security. We don't have to go off the rails, but it does always shock me that when you talk about all these things you need for day AI data centers, where you need, say, you want water, you want land, you want access to tons of power, you need um smart people with math skills. That the

22:39Unidentified speaker (speaker_2): country of Russia has no frontier models.

22:42Will Cherlin: It

22:42Unidentified speaker (speaker_2): seems like they would just be killing this thing with a centralized government, they could just go crazy with it, but instead,

22:47Will Cherlin: no.

22:48Unidentified speaker (speaker_2): So it it's obviously a little

22:49Unidentified speaker (UNKNOWN): bit more

22:49Unidentified speaker (speaker_2): complicated than that. But it's uh it's interesting that

22:52Will Cherlin: these

22:52Unidentified speaker (speaker_2): different um

22:53Will Cherlin: pieces

22:54Unidentified speaker (speaker_2): to the puzzle are getting rearranged constantly and trying to make sure they stay flexible to be there for all of

22:59Will Cherlin: them. So as we wrap up here, Joel, before I go too crazy on my tangent, okay. As we wrap up here, Joel, what do you kind of see as

23:07Unidentified speaker (UNKNOWN): the

23:08Unidentified speaker (speaker_2): benchmarks

23:08Will Cherlin: to

23:09Unidentified speaker (UNKNOWN): be

23:09Unidentified speaker (speaker_2): where you

23:09Unidentified speaker (UNKNOWN): want to

23:09Unidentified speaker (speaker_2): be in three, five years from now? What are the goals that you

23:12Unidentified speaker (UNKNOWN): are

23:12Will Cherlin: trying

23:12Unidentified speaker (UNKNOWN): to hit in the

23:13Will Cherlin: next 12-18 12,

23:14Unidentified speaker (speaker_2): 18 months to really be where you wanna be in, say, the year 2030?

23:18Unidentified speaker (speaker_1): Yeah, no, it's a good question. Well, we are laser focused right now on securing our our first tenant or tenants. And I think a lot of it stems from that. And you've got to look at what makes a good tenant. On one end, you need a tenant that's gonna wanna work with you, be reliable, be with you for the long term. You've got to understand the financial health of the tenants because the credit markets and the broader ecosystem in terms of being able to raise capital to deploy in these CapEx initiatives are significant. And not all tenants are created equal. And so there's a bit of a discerning process on both sides. From the tenant side, they're looking to companies like ourselves. Do you have access to power and land? Can you execute on this? Do you have the team around you that can develop and build these data centers and deliver on time? And those are all fair questions that the ones that we are excited to. be measured up against. As I kind of look to kind of the medium term, let's say beyond 12 months, our vision has been from day one since we've started Dark Horse of building a platform where we can really tap into under underlooked at areas, underserved kind of properties and portfolios, and really create a repeatable model where we can bring power online in a community friendly and forward way as quickly as possible. to to serve the the needs of tenants and demand.

24:30Will Cherlin: Joel, you've built a bunch of companies, done

24:32Unidentified speaker (speaker_2): a lot of cool things.

24:33Will Cherlin: What gets you out of bed in the morning that this is kind of what you want to

24:36Unidentified speaker (speaker_2): spend your time doing and this is what you want to be working on?

24:38Unidentified speaker (speaker_1): I think there's a couple things. I think it's the intersection of the opportunity, like the industry. I think it's it with two components. One is the challenge. It's a challenge. A lot of people are racing toward this. There's a lot of money, a lot of CapEx. You're seeing the hundreds of billions or trillions of dollars being forecasted toward AI. And I kind of always enjoy working towards challenges and solving problems. I think the third thing is the people around me. I think I'm at a point in my career where I enjoy the work and I want to enjoy it. I want to enjoy the people I'm working with. And I've had the ability to kind of inherit some teammates, bring some new new people into the mix and really build the type of organization that I enjoy getting out of bed and working with every day. And uh it keeps me going throughout the day and early and late into the evenings.

25:23Will Cherlin: Yeah, no, it's it's a very exciting time where about a day after anthropic release their S1 that kind of was almost like the uh beat earnings almost to when people are looking at it and trying to decide how much cash do these AI companies have, how big are their obligations. And a lot of fears were put to rest, it seems like in the last 24 hours or so. So it's a very exciting time for you.

25:45Unidentified speaker (speaker_1): Yeah, no, I I that that was a big announcement. I think people are still trying to read it. The document is quite long and thorough, but there's been some good analysis on it. Well,

25:53Unidentified speaker (speaker_2): yeah, you know, you got to think of a cyclical business idea like that, that everyone had to put this 400-page 400 page S1 into their clod

25:59Unidentified speaker (UNKNOWN): and

25:59Unidentified speaker (speaker_2): use up their CLOD tokens to get through it. It was a master stroke. by them to uh pad their own pockets. So it's genius.

26:06Unidentified speaker (speaker_1): Fulfills the cycle.

26:07Unidentified speaker (speaker_2): Yeah, exactly.

26:09Will Cherlin: Well, as we wrap up,

26:10Unidentified speaker (speaker_2): Joel, people want to hear more about dark horse. They want to learn about more of what you're up to. Where should I send my audience?

26:15Unidentified speaker (UNKNOWN): Uh

26:16Unidentified speaker (speaker_1): I think there's two places. Um we have a handle on X that we've become more active on uh under Dark Horse or DRK as is the ticker. And then our website does a pretty good job um keeping people up to date on releases and those components. That's darkhorse.inc.

26:30Unidentified speaker (speaker_2): Awesome. Well, Joel, thank you so much for coming on the podcast. And for those of you still listening, please remember to subscribe and review. Thanks.

26:37Unidentified speaker (speaker_3): This podcast is for informational purposes only and does not provide financial, legal, or any other advice. The information is not investment advice or any offer to buy or sell any securities or to make any investment. The view views expressed by the guests are their own, and any reference to third-party products, services, or information does not constitute an endorsement therefore by NanoCap Podcast or its affiliates. The NanoCap Podcast Express disclaims all liability for any individual's use of the information presented in this podcast.

This page was produced automatically: the audio was transcribed with nvidia/parakeet-tdt-0.6b-v3+sortformer-v2.1 and interpreted by Qwen3.6-35B-A3B-UD-IQ3_S.gguf. Transcription and interpretation can contain errors; always confirm against the original episode. Nothing here is investment advice.
All insights