The Pomp Podcast - #1518 Asher Genoot | Bitcoin & AI Data Center Deep Dive

Episode Date: March 31, 2025

Asher Genoot is the CEO of Hut 8 Corp, one of the leading companies when it comes to bitcoin mining and AI data centers. In this conversation he breaks down what the business is, why it’s important, how they are thinking about deploying capital, how they are selecting the sites, and the difference between bitcoin mining and data centers.=======================BitcoinIRA: Buy, sell, and swap 75+ cryptocurrencies in your retirement account. Take 3 minutes to open your account & get connected to a team of IRA specialists that will guide you through every step of the process. Go to https://bitcoinira.com/pomp/ to earn up to $500 in rewards.=======================The future is being built today and the future of currency isn’t dollars, euros, pounds, or yen, it’s crypto. And Gemini thinks that’s a great thing. Because a future where money is decentralized, inclusive, and globally accessible, that’s a future that we are anxious to be a part of. Go where dollars won’t. With Gemini. =======================Pomp writes a daily letter to over 265,000+ investors about business, technology, and finance. He breaks down complex topics into easy-to-understand language while sharing opinions on various aspects of each industry. You can subscribe at https://pomp.substack.com/=======================View 10k+ open startup jobs:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://dreamstartupjob.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Enroll in my Crypto Academy: https://www.thecryptoacademy.io/

What's up everyone? This is Anthony Pompliano. Many of you know me as Pomp. You're listening to the Pomp Podcast, which is my effort to find the most interesting people in the world and sit with them for hours while I ask questions in an effort to learn. So it would mean the world to me if you would subscribe to the show on your favorite audio platform, watch episodes on YouTube, and tell your friends and family about the podcast. My goal is to help millions learn from the world's most interesting people. So let's get into today's episode. What's going on, guys? We've got a great episode for you today. It is with Asher Janute. He is the CEO of Hut8, one of the leading companies when it comes to Bitcoin mining and

AI data centers. In this conversation, he breaks down what the business is, why it's important, how they actually are thinking about deploying capital, how they're selecting sites, and what the difference is between Bitcoin mining and those AI data centers. This conversation is going to give you a peek into the future of energy and compute, and it's really going to highlight how important this intersection is right now. I hope you guys enjoy it. If you listen to the whole thing. I promise you're going to learn something new about these industries. And so here's my latest conversation with Asher. Anthony Pompliano runs Pomp Investments. All views of him and the guests on his podcast are solely their opinions and do not reflect the opinions of Pomp Investments.

You should not treat any opinion expressed by Pomp or his guests as a specific inducement to make a particular investment or follow a particular strategy, but only as an expression of his personal opinion. This podcast is for informational purposes only. trust Bitcoin IRA to secure their financial future. Getting started is quick and easy. It takes just three minutes to open an account. Once you're set up, their team of IRA specialists will reach out to guide you through every step of the process. Whether you're transferring an IRA from a legacy bank, rolling over an old 401k, or starting fresh with new contribution, the Bitcoin IRA team is here to help you get access to real crypto in your retirement account.

And here's the best part. As a listener to this podcast, you can earn up to $500 in rewards when you add funds to your account. That's right. Search for Bitcoin IRA in the App Store or visit BitcoinIRA.com slash Pomp to join 200,000 Americans on their journey to upgrade their retirement. BitcoinIRA.com slash Pomp to upgrade today and you can earn up to $500 in rewards when you add funds to your account. Today's episode is brought to you by Gemini. The future is being built today, and the future of currency isn't dollars, euros, pounds, or yen. It's crypto. And Gemini thinks that's a great thing, because a future where money is decentralized, inclusive, and globally accessible, that's a future that we are anxious to be a part of.

Gemini teamed up with futurists, technologists, and designers like award-winning artist Matt Griffin, known for his illustrations for Dune, to craft a vision of the future with crypto at its core. The creative theme, Go Where Dollars Won't, emphasizes exploration, growth, and the Crypto market's limitless potential, whether it's going on that Martian safari to capture an unforgettable photo of a herd of woolly mammoths grazing the red planet or using Bitcoin to pay for your lift pass to go strata skiing down the mountain peak of a breathtaking comet. Whatever adventures we'll be living, one thing is for sure. In a future this fantastic, the limits of traditional currency simply cannot keep up. Financial innovation will usher in this new frontier. And so go where dollars won't with Gemini.

Go check them out at Gemini.com slash go where dollars won't. Again, go check it out at Gemini.com slash go where dollars won't. I thought a great place to start the conversation is you guys provide energy and energy seems to be this thing that somehow has gotten demonized over time, but it is foundational to civilization and the success of humans in these civilizations. So talk a little bit about the importance of energy and then we can get into how you guys are kind of pioneering this new use of energy. I think that's exactly right. There's this kind of idea and belief that anything that consumes energy is bad. And if you look at kind of the progression of humanity, we've only consumed more and more energy on an exponential scale, right? One of the largest culprits and consumers

of energy is air conditioning system. Imagine living without AC units and living in our lives today. And so, and heaters, not just cooling, right? As we look through, so the World Bank kind of had this beautiful graph and it's GDP per capita of every single country and then energy consumption per capita. And there's essentially a linear correlation. The more energy you consume per capita, the higher your GDP is per capita as a country as well. And so I think that is extremely fascinating because there's this general idea of consuming energy is bad for the world, it's bad for the environment. But the reality is technology and every technological advancement we have requires energy in order to move forward. And if we really think about it, I mean, even the food

we put into our bodies, there's energy that went into producing that food, right? And so I think for us, as we think about the business, it's how do we sit at that intersection of power and technology? And how do we build infrastructure that supports the technologies of tomorrow, all the new innovations? So when we started the company, Bitcoin compute was a big thing is how do we create this decentralized system, but have really rigorous and robust security. And we essentially had proof of work and uh and we created large-scale mining sites and then there was and then now there is like ai compute and data centers for ai compute how do you power ai and i think it's really important that america is at the forefront of that and having the platforms

that do so but those two types of data centers are very very different load profiles right a data center for bitcoin computes you can shut off the data center as long as power prices are high because you're not you're price sensitive you're agnostic on the mission critical of that infrastructure a data center your mission critical and you're not shutting down for power prices so you're running through it so where you place a tier 3 ai compute data center versus where you place a bitcoin compute data center are different um and your the models are different so for us it's really understanding the technology the low profile of that consumption and then how do we build infrastructure that supports it and connect that into the grid and different power systems so

as a company we have sites connected front of the meter just directly into the transmission grids behind the meter at power generation facilities and then we've also spent some time looking at islanded facilities because we own four power plants ourself in ontario and so we've been looking a lot about what if you built power generation for the consumption and it's not a new or interesting novel concept like that's happened historically if you think about aluminum smelter steel mills, paper mills. We've done this in America before when these industries existed and were thriving back in the day. So there's old use cases of energy, and then there's what I'll call net new or new industry use cases for this energy.

You mentioned Bitcoin and AI, which are the two that you guys are running full speed at. Those seem to be the two largest consumers, but also the most important from a market cap creation of value standpoint. Do you think there will be other ones as well before we get into talking about your Bitcoin and AI business. Is this, okay, at first it was Bitcoin, now AI, and there's going to be five or six other of these over time? Or is it pretty much Bitcoin and AI is what you see on the horizon? No, I think so. I mean, if you asked me two years ago, what would be the next use case? I mean, we were going deep in green hydrogen, carbon capture, industrial batteries. We were looking at the re-domestication of manufacturing at scale for these large industries. And so,

I mean, we weren't thinking data centers because until recently with AI, you couldn't go place a data center outside of a main tier one market. Now, because of training in the GPU clusters, data centers are going to new regions. We just announced a deal in Louisiana that we purchased a project for, and that's focused on the AI side of the business. And so AI has been this new massive consumer of this demand. And obviously, the technology merits the amount of hype around it. Uh, but I do believe that there will continue to be new in, uh, large use cases for power at scale. What, like, why is Louisiana not an area that people have previously gone?

And you mentioned that now it is possible, but like, walk me through that deal, right. In terms of whatever you can share, uh, to really highlight how these new consumption components, uh, for this energy, we know flare gas, uh, mitigation sounds like there's other things with AI, but like this really changes kind of the energy map, if you will. And so talk me through that Louisiana deal. Sure. So when you think about kind of the data center market, historically, there were kind of tier one markets, tier two markets, and tier one hubs are like Ashford, Virginia. I think a lot of that first started with like government demand, and then you built these clusters. And ultimately, if you think about interconnectivity is important when

you think about data centers and how data can go from one location to another. And so the more of a cluster of a data center market you can have, the better. And then the fiber lines were built around that. I mean, there's no secret sauce on why were these a data center market. They were built there. The fiber connectivity was built there. The interconnections between these different places built a lot of fiber to connect these different hubs. And that's why you had kind of these tier one markets. And so Louisiana was never a data center hub, right? And so when GPUs became a thing, the reason why being a tier one market matters is your latency on your data center and how quickly you can ping that data center, transfer information, and so forth.

With GPUs and AI training, it was a little bit different because the real need is how much can we train for as cheap and as large as we can, and then we can send that information somewhere else where there can be inference on that in order to run your application layer. And so there was this openness with kind of NVIDIA obviously becoming one of the most valuable companies in the world, OpenAI, XAI. And there's this AI race to build these large training campuses. And the requirements were less so on latency and more so on power and scale of facilities. And so in areas that are already pretty tapped out, it's hard to find that scale of land and power. And so people started saying, you know what? We're open to looking at other

markets. And so for HUD-8, we have a large footprint in Texas. And we were like, okay, we have exposure in Texas. We have some other sites in Canada, upstate New York. What are other markets we think could be really, really interesting data center markets? And so we kind of looked around the whole US. And one of the markets we isolated into was Louisiana. And the primary reason is when you think about what makes a good data center market, like one, what does a fiber connectivity look like? And so Louisiana sits right in between Texas, Alabama, Atlanta. And so Alabama is trying to be like a Louisiana where it's really... Atlanta is a massive data center market. If you think about Dallas is a massive data center hub.

And so Louisiana sits right in between that fiber corridor. And so from a fiber perspective, can you be either an overflow market for Texas, or can you be an alternative to the congested the land to markets to hit into kind of the whole southern region of the world right and so that was one from a fiber connectivity perspective is actually really interesting and there are a lot of long-haul fiber arrives with with good latency the second is like louisiana is also one of our biggest energy hubs in the u.s right you have so much gas being coming out of louisiana that gets transported to the rest of the u.s so instead of transporting that gas to other places in the u.s where then they turn that gas into electricity,

that electricity into AI compute. Why not just like keep it at the source? And so really, really interesting energy market. And then three, the regulatory side. You have a new governor, Governor Laundrie. He's been amazing and so pro-business. Him and the legislator passed last year a data center sales and use tax exemption on data centers.

It's trying to welcome the industry into the market. So you had a very pro-business, pro-AI data center political regime. And so for us, we're like, this is a really interesting place. So we started doing a lot of work on the project. And then Meta announced their big site in northern Louisiana. And that really further validated because we still have to try to convince people, no, this could be an interesting market. People are like, I don't know. It's Louisiana. I'm not sure. There's no data centers there. When Meta kind of announced it to help put it on the map, and obviously, we were already deep, deep, deep intertwined. I think even probably same time or

pre even uh meta and so we're excited for louisiana we think it's an interesting new market we think we uh have seen a lot of demand uh for for that area we think aside from just our projects that there will be more projects in the long term it's a regulated utility it's not a competitive choice market and so unlike ericott like you have energy louisiana um and that needs to be very thoughtful to how does this low profile come and how to make sure rate payers are protected It's a little bit of a different dynamic and maneuvering. It's a hard energy market to navigate. And it's funny because people deep in the energy space have given us a lot of kudos for navigating the area, but they've been awesome and great partners, both from a utility perspective, local state parish president, Kenny, state official, the governor himself, and they've been all really supportive.

And so that's a great example of a project in our development pipeline that we take a bet on for a multitude of reasons. And our site specifically sits in a very, very generation-heavy area. We bought almost 600 acres of land there. And we have massive transition lines that go through the property. And so we have to get like 14 landlords to sell and kind of put this piece of property together. And the line that we're connected to is directly into the nuclear substation. And so that's exciting too, that it's connected to the grid and going straight into that substation. And so overall, we think that's an interesting site.

But look, it's one of the many sites in our development pipeline. This was announced to the world earlier than I wanted to. We had a public hearing and it got leaked and then it was all over the news. But we have a lot of different sites that we have that we have a similar philosophy and like, how do we build and where are these data hubs and where can we be competitive and have a competitive edge to other peers? So you go and you acquire the 600 acres, a bunch of work goes into it. How do you decide whether you build a Bitcoin data center or a AI data center?

So there's a couple of big drivers. In an ideal world, you find a site that can do both, right? And so a lot of our sites in Texas, we can put Bitcoin because power is cheap. You can curtail around the power volatility to drive down your price down. And then there are AI use cases like one of the Stargate campuses, they're building a mega campus in the Abilene site in Texas. And so Texas is a really interesting market for that because you can drive down your cost the cost of energy. And as a Bitcoin miner, your cost of energy is what you care about most.

Louisiana is a regulated energy market. And so you have less of that ability to curtail around that energy volatility because your capacity payments are higher. It's the slippage between just your pure energy rate and your transmission delivery fees. In Texas, if you're pulling at a transmission level, you can drive those fees down a lot. And really, the majority of your payments are for the energy component. In a place like Louisiana, that's not the case. And so for Bitcoin, it's a lot harder to operate there because the power prices are higher and you can't curtail around that volatility. And so this site was purpose-developed for AI and traditional data center development in areas like we have this Vega campus, a 205 megawatt campus in Texas, and Amarillo, interconnected with the wind farms, connected to the grid as well for the redundancy.

and that's a great example of like we were able to get that site who said we can pull power here within three quarters and they're like no data center can do that we came in we locked it up with bitcoin and procured the power and then over time we can do whatever use case we want there yeah and what's the cost difference between building these data centers right the bitcoin ones uh the bitcoin ones are one-fifth one-tenth one-fifty at the cost like what does that look like yeah massive difference so the vega project we're building we're building for a little over $400,000 a megawatt. And that's all in

cooling systems, racks, foundations, civil, electrical, everything, not including the chips themselves. That's probably around half the price of anyone else building it. We vertically integrate a lot of designs in the equipment ourselves, so that's not kind of the industry standard. I think industry standards are probably closer to $700 to $1 million

a megawatt there. A traditional tier 3 data center liquid cooled is somewhere between $10 to $13 million a megawatt. And prices have gone up because labor is hard to get. So you have to pay per diem and overtime, obviously materials, a long lead time item. So it was, I think, traditionally probably closer to the 10. You're seeing kind of 10 to 13 because people are trying to build quickly and they're overpaying

to build, but much, much more expensive. And what is the profitability of the difference of these things, right? obviously people know can be really, really profitable. At times though, if you got the wrong power costs, you can be on the wrong side of that. AI, I think people see all the hype and they're like, well, somebody's making money somewhere. So how do you think about revenue or profit per megawatt or something? So the best way to think about it from my perspective is we look at three big metrics and there's a multitude of others, but one is what is your simple payback or i.e. your unlevered returns, right? So if I'm not putting debt on this thing,

how much am I investing? What's my profit margins and how quickly can I get a payback? Then the second question is like, what about if I have leverage? What is my levered return on the equity I invest? And the third is like, what is the actual market value of this project once it's stabilized as a stabilized asset? How much could you sell this for? And so I'll kind of go through each of them and how they differ. When we think about the Bitcoin compute and data centers for Bitcoin compute, the problem is volatility, right? You can make a ton of money when Bitcoin's ripping and hash prices are high, or if Bitcoin goes down and your energy costs are not that great, you can be squeezed. And so from an unlevered perspective, you're asking for a

much faster payback because of the risk profile of the consumption and the use factor. And so you're trying to target really like, call it a two to three-year paybacks on the infrastructure you invest into. On the AI side, I mean, you're looking at signing 10, 15-year deals with companies that have better credit profiles than banks. And so in that scenario, your payback is a much longer period because you have a guaranteed fixed return for a 10 or 15-year period, right? So you're maybe looking at like, call it a six to seven year simple payback. So if you look at it kind of apples to apples, you're like, okay, well, Bitcoin is way better because you get your money way quicker.

The answer is yes, but is it better from a risk adjusted basis? Because one is you're guaranteeing your return. The other one is you're taking market risk, both on energy and on the revenue, because on the AI side, you're passing through the energy to the customer. You're really just taking a return payback on your infrastructure costs. Then when you look at it on a lever basis, cost of equity and capital is really expensive and debt on Bitcoin mining. And so if you can get it, if they'll even give you that, because if, and so, but on

the traditional data center side, I mean, you can get 70 to 80% LTC construction financing if you have like a, call it investment grade counterparty as your customer. And so you're really only putting up 20% to 30% of the actual equity rather than 100% on the mining side. And so on a lever perspective, your returns actually look pretty great, right? And then the last piece is like, what is the actual stabilized asset value? Like your data center, you trade at 15, 20x multiples. Your data center, you're a Bitcoin company, you're not trading at those multiples. And so because your kind of stabilized cap rate is really, really great.

And so you go build a 200 megawatt IT data center. I mean, that's two plus billion dollars in equity value that you're creating. And so that's why you see such a big focus and shift on wanting to build these AI data centers. There's a lot of value there. Our perspective is like, it's not one or the other. If you have enough power, go and capture these different market opportunities. we have different teams that run these different businesses. We have different people focused on it.

We have such a big development pipeline that we can figure out where they go that drive us the most value in the short term and in the long term. When you look at what is being built, one of the risks or one of the data points that people say, oh, hold on a second here. This may be overbuilding.

Microsoft has started to cancel some of their leases, but it's a little unclear, I think, to people who are not in the day-to-day industry as to why they're doing that. And so talk through a little bit of these different maybe market approaches. You can go build your stuff, you can let somebody else build it and lease it,

or are they walking away saying we have enough space? Like just talk through some of that nuance. Yeah, I think that situation a bit overblown out of proportion. And at the end of the day, you have a couple of large scale consumers of these products, right? And I think the question that the markets keep on asking is,

is this demand real? And if so, how long? And is this sustainable? And when you hear a Microsoft canceling contracts, oh no, the end is here. There's no more demand. At the end of the day, every company will have their own demand signals

or have their own engineering things that they have to solve for and whatnot. And so overall from our side, whether it be the DeepSeek news, the Microsoft news, we continue to see demand. We continue to see customers going for these data centers. Obviously every customer has a different demand profile. Some customers, they have platforms

that they need to procure from. some of them have customers. And so like any other business, it is very company specific. I mean, if you look at Microsoft, OpenAI, they're tied by the hit, right? And that demand was going Microsoft, Microsoft, Microsoft. And then they go and stand on stage with President Trump on a $500 billion announcement of net investments, and it's OpenAI with SoftBank and Oracle. And so there's obviously some dynamic changes there in that relationship and OpenAI now being able to work with more than just one kind of counterparty. And so I would say overall, the big question I ask is, has AI truly penetrated its full use case? And I think

not even close is my opinion. And so I think we're still in the early stages of that. The question is, how many large-scale campuses do you need? Who are going to be the main consumers of those large-scale campuses, and how much of that is going to be training versus inference. But at the end of the day, I mean, the demand is still strong. We're still seeing appetite. A lot of it, obviously, is we want these things yesterday. And so we continue to see demand being there. And it varies from company to company. And there are shifts. One company might say, I have this demand. They fill it. They don't have that for a month. And then they have another a demand signal. And so it's really dependent on the actual consumer itself. And who's like

financing all this stuff, right? Like you're a publicly traded company. There's many other Bitcoin miners or AI kind of data center businesses that are public. I imagine there's plenty private. Is this like the full capital stack, everything from private equity to public markets, debt, kind of everything is at your disposal or are there specific financing mechanisms that people are pursuing here? No, everything. Our CFO was the Managing Director of Power Utilities and Renewable at Citibank. And he came after being there for 13 years. And he said he's never in his career seen such a large influx and pouring of capital on one specific asset. And the amount of demand on data center infrastructure to finance is incredible. Public markets, private

markets, equity, traditional bank financing, project level financing, private equity insurance funds, sovereigns, hedge funds, the whole gambit. There's an incredible amount of capital. I think like any industry, capital gets smarter and smarter and says, what is a good deal? What are the deals we want to finance? How do we drive our return? But there's a lot of capital financing these industries today. What is the thing that you're paying attention to most right now in building out AI or the Bitcoin data centers? Is it getting them financed? Is it site selection? Is it the structure? Liquid cooling was not very popular a decade ago. Now that's all the rage. What are the things that either keep you up at night or that you go into the office every day

and you're like, this is what I'm paying attention to? So I see it in stages. There's a couple of things that I foresee, but the focuses are at different stages of our company. Stage one of last year was how do we continue to scale up a robust development pipeline? So our team has done a phenomenal job building a really, really great pipeline. And I told in my last earnings call, I'm like, if Louisiana didn't get leaked, I wouldn't even be talking about it right now. But because it did, I have to address it. So we have a lot of sites in our development pipeline that we're extremely excited by. So that was kind of phase one, which is like, have really great assets, whether it be for AI, Bitcoin, like have great assets that we can

monetize and capture the power and own that power. Today, we need to show the market that we can build and deliver a large scale AI campus. And so it's about getting a customer deal done and delivering that data center. That's the phase we're in today. We have to show the market that we can deliver. We have to show customers we can deliver. And we have the assets that we believe can be monetized in order to do so the phase after that is really like where where i believe is like where we'll really shine data centers cost so much money to build like there is no reason why we should be building a liquid cool data server for a little over four i think it's like four hundred eleven thousand dollars a megawatt right there's no reason we're building this for that much and so

So I think long-term, as the power crunch and the supply and demand kind of match out maybe, the competitive edge is how do we build these things way more efficiently and way cheaper? We think about the Bitcoin mining ecosystem, what I take pride in is I think we can build Bitcoin mining data centers that power Bitcoin compute much cheaper, more efficiently maintaining quality than our peer set. and that's the same type of mentality we want to bring to traditional data centers but that's not the problem we need to solve today the problem we need to solve today is people need

access to power they want to build the product that they know so let's go and deliver that to them and lock in those economics you guys also have um depending on where the price is about a billion dollars of bitcoin on the balance sheet i think it's somewhere around 10 000 bitcoin or something. That's a lot. It makes you one of the larger holders in public markets. Is that just a, hey, we're going to keep putting Bitcoin on the balance sheet and Bitcoin goes up? Or what was the thought process there? So I started a company called US Bitcoin Corp. We merged with Hudate at the end of 2023. And Hudate was one of the first public, I think the first public company, if not one of the first public companies that held Bitcoin on the balance sheet.

And so circa kind of back in even 2017, starting to accumulate and holding Bitcoin. And a lot of the shareholders really believe in that HODL strategy. When I came in, I'm a lot more kind of practical, which is like our jobs is deliver returns for our shareholders and our job is go build a great business. And so when I started, I said, you know what? Our stock was like trading really low in close to cash. I mean, it's not doing so great right now with Bitcoin being down these last two months. But I said, if there's an opportunity to invest to grow the company, we will sell Bitcoin to advance those developments and growth because the cost of our equity is so expensive. Then towards the end of last year and even today with debt financing available, there are sources of cheap capital.

So we will use those sources of cheap capital to grow the business because we believe in Bitcoin. We believe in the appreciation of Bitcoin over the long term. And so I think as a company today, if we think about the three layers of our business, layer one is energy infrastructure. Our long-term conviction is energy will continue to be a rare and scarce asset. And we want to accumulate as many megawatts as we can in our energy and power layer. The second layer is digital infrastructure. We want to figure out how to monetize that energy.

So we'll be building data centers. Today, we're building tier zero data server, Bitcoin compute, tier three data centers for AI compute, and whatever other applications and use cases. then the third layer is compute in our tier three data centers we have gpu compute so we spun out a company called high rise that ai and it's a separate company that builds because we couldn't invest continue capital buying gpu so we're like let's finance that separately then there's a6 for bitcoin compute that goes into the actual data centers today we finance that off that balance sheet we've spent a lot of time thinking about is that better on balance sheet off balance sheet

does it more operate better if it's like a high rise dot ai right so we've been thinking about a lot of that and i shared that with the market do these companies work together and as a part of that, it's like the Bitcoin treasury management huddle strategy. And so I think we're believers in Bitcoin. Otherwise, obviously, we wouldn't be investing in building all these dollars for it. And I think we'll continue to get a lot smarter on how do we structure our business, that we can give investors the type of exposure they want, the volatility they want. Meanwhile, as executives, maintaining the ability to capture as much of the value in all these different areas we believe in. And what have been the downsides to holding Bitcoin on the

balance sheet, right? Because everyone always talks about the upsides, but obviously when you're in the mining business, there's kind of booms and busts with the market cycles, but Bitcoin also goes up and down, obviously. And so talk a little bit about what some of those downsides have been. Volatility is exactly what you said, right? We've got hundreds of millions of dollars of appreciation of Bitcoin by holding it from end of 22, beginning of 23 to today. At the same time, if we look at it from end of last year to now, three months later, Bitcoin went from about 100,000, 100, whatever thousand now to 80 something thousand. So in theory, we lost 150, 200 million, right? That volatility is real. We're still net positive. And our average

Bitcoin on our balance sheet is under $30,000. And so we're still very much in the money, right? But there's volatility. We believe in the upside, but Bitcoin is still a very volatile asset. And that is kind of the pros and cons. I think the positive is, if we were to hold a billion dollars of cash on our balance sheet, I think people would freak out. Like, why are you holding all this cash? This is not a good use. Give it in and out to your shareholders and so forth. But when you think about the types of deals we're trying to do, we're trying to do multi-billion dollar deals with large counterparties. When we tell them, hey, we have about a billion dollars of liquid assets between cash and Bitcoin, that gives us a lot of credibility because they know

we can actually follow through with these projects and we don't have to go raise the money on the back of them giving us a contract. And so there's pros and cons like everything, but overall, it's done well for us. How do you describe the business? When you're talking to shareholders or potential partners, you and I have been working together now for a while. I'm an advisor to the company. And one of the things that I always talk to CEOs about is just like, the words you use are really important. And I think if you go back in history, it was a Bitcoin mining business. Today, it's not, right? They do that business, but it's something that seems much bigger. So how do you describe it?

Our elevator pitch, you call it, is HUD-8 is building a next generation energy infrastructure platform at the intersection of power and technology. That's a business. It's not just Bitcoin mining. It's not just data centers for AI or any other use cases. It's that intersection of power and technology and whatever can drive the most value. And my belief is what drives the most value of the technology is that empower humanity most. We will be building infrastructure to help support that. And you've got that energy infrastructure. I think that's the other key piece here is like, you guys want to own some of that infrastructure. You also want to own kind of the data centers, Bitcoin, AI, whatever. But by putting them together,

that vertical integration really is something that kind of sets you guys apart from a strategy standpoint, it feels like. That's right. I think because we took this power first approach when we started, I kept on telling people we're an energy company, not a Bitcoin company. People said, screw off. That's not true. You're a Bitcoin company. And I think finally, the story is starting to build legs. I kind of feel like it was like Jeff telling people he was not a book company like you're selling books you're a book company i don't know what you're talking about um but because we thought of ourselves as an energy company i mean we went and recruited people from the energy background to be on our team so today we have like deep energy expertise

within the company right greg who runs our corporate development team was that next era for over a decade sean was in power and utilities that city for over a decade brad um chris i can name all these people and then even our board members i mean mayo shattuck ran constellation energy and excellent and he was on our board like we brought on energy people when we started the company and we built our board it was all around energy focus and so i think that's paid off because at the time people didn't really get it or believe it but now as we're starting to build more verticals i think that story's coming together and then today i think people are like okay you're trying you're in bitcoin company trying to be a data center company i think when we build the

next thing then they'll finally get the full picture yeah it's like uh one is the data point right two is kind of like a line three is the trend exactly you gotta get to the third one that's right where can we send people to find out more about the business yeah i think just type in huddy presentations we're starting to put out a lot more investment presentations and decks and update in real time uh we've had a high velocity of new deals happening and so we try to do as much as we can and update our investor presentation as much as we can and so just go to our website and you'll find a lot of information on there. And we spent a lot of time telling our story

through our earning scripts and we record them and have them live. So the last two, we've been really talking about the transition, what the vision is. I think unlike some companies, like we have purposefully focused a lot of those scripts on what are we trying to build, not just what are the financial results for last quarter.

And so you'll hear a lot of kind of the vision and the strategy in those scripts. Amazing. Thank you for doing this. We'll definitely do it again in the future. Yeah, thanks for having me. appreciate it.