Q2 2025 Earnings Call Transcript

Good day, and thank you for standing by. Welcome to Riot Platforms' Second Quarter 2025 Earnings Conference Call. [Operator Instructions] Please also be advised that today's call is being recorded. I would now like to turn the conference call over to Phil McPherson, Vice President of Capital Markets and Investor Relations at Riot Platforms. Please go ahead.

Thank you, operator. Good afternoon, and welcome to Riot Platforms Second Quarter Earnings Conference Call. My name is Phil McPherson, Vice President, Capital Markets and Investor Relations. And joining me on today's call from Riot are Jason Les, CEO; Benjamin Yi, Executive Chairman; Colin Yee, CFO; and Jason Chung, Executive Vice President and Head of Corporate Development and Strategy.

On the Riot Investor Relations website, you can find our second quarter earnings press release and an accompanying earnings presentation, which are intended to supplement today's prepared remarks and which include a discussion of certain non-GAAP items. Non-GAAP financial measures provided should not be considered as a substitute for or superior to the measures of financial performance prepared in accordance with GAAP and are included as additional clarifying items to aid investors in further understanding the company's second quarter performance. During today's call, we will be making forward-looking statements regarding potential future events.

These statements are based on management's current expectations and assumptions and are subject to risks and uncertainties. Actual results could materially differ due to factors discussed in today's earnings press release, in comments and responses made during today's call and in the Risk Factors section of our Form 10-K and Forms 10-Q, including for the 3 months ended June 30, 2025, which will be filed later today as well as other filings with the Securities and Exchange Commission. With that, I will turn the call over to Jason Les, CEO of Riot Platforms.

Thank you, Phil, and good afternoon, everyone. I'm excited to walk through the results of another strong quarter for Riot. But before we dive into second quarter earnings, I'd like to share Riot's strategic road map and provide some additional context to the development of our data center business and how we view all of our operations working together in a complementary manner. We are incredibly proud of the position that our company, Riot Platforms, is in today.

Over the last 7 years, we have scaled incredibly, both in terms of our size and our capabilities, representing the culmination of years of hard work, long-term planning and coordination, all with a view to taking ownership of our future and placing our destiny in our own hands. We have grown and evolved as a company, driven by our ability to develop world-class capabilities, including land and power procurement, Bitcoin mining at a globally significant scale, power management and trading at scale, engineering, manufacturing and servicing critical electrical infrastructure and significant access to global capital markets.

Recently, we have added a new world-class capability. With the hiring of Jonathan Gibbs, Riot's Chief Data Center Officer and other highly capable professionals from the traditional data center industry, we find ourselves at the beginning of another exciting chapter in Riot's story. With this new capability, we are about to undergo the next step of our evolution as a company. With the ability to build and develop high-performance compute data centers, we will transform Riot by establishing a robust and scalable data center segment.

Successful execution in this regard is Riot's top priority, and we recognize the importance of clearly articulating our approach to investors and stakeholders. To be clear, we are not pursuing a so-called pivot into AI HPC initiatives with a view of doing a "deal"." Rather, we have added a new data center development capability, which we will apply to as much of our power portfolio as possible and which will transform our company in the years to come. This mindset informs all of our decisions, enabling us to capitalize on this exciting opportunity with discipline and foresight. If I had to summarize our strategy into a simple elevator pitch, the pitch would be that Riot is in the business of monetizing megawatts with a view to utilizing as much of our power portfolio as possible and maximizing the value of our megawatts over the long term.

We will maximize the value of our operational assets, specifically optimizing our megawatts to use all available power. We have a great advantage with a portfolio of ready-for-service power, anchored by our operational flagship sites at Rockdale and Corsicana. These assets are not conceptual. They are active today, thanks to prior investments in Bitcoin mining infrastructure.

This enables more certain execution on our data center development initiatives compared to a stand-alone traditional developer. Our Bitcoin mining capabilities have proven integral to this strategy as they underpin our ready-for-service power portfolio. By utilizing our mining capabilities, we have put ourselves in the fortunate position we find ourselves in today, and we can secure new power sites by playing to our strengths, profitably managing risk and simultaneously creating a sustainable cycle of growth. Given the attractive economics and higher valuation multiples associated with data center leases to high-quality tenants, converting as much of our power portfolio to data centers remains our preferred end use for those assets.

The pace of transition from Bitcoin mining to data centers will be influenced by customer demand trends, the availability of financing and the general data center market. Our current efforts are laying a strong foundation for a pipeline of future transactions. We have many advantages that have put us in an incredible position because we offer a unique combination of significant scale of readily available power in high-demand jurisdictions, a strong balance sheet underpinned by holding more than 19,000 Bitcoin and $330 million in cash and with significant access to the capital markets, experienced hyperscale data center leadership and development capability, scaled efficient Bitcoin mining revenues, generating hundreds of millions of dollars in revenues and cash flows annually and battle hardened and experienced management and operations teams.

With this framework, our mission is clear. Riot will maximize value across our entire power portfolio with a view to ensuring no stranded capacity, aggressively shift power capacity towards data centers, strategically expand our power assets, utilizing Bitcoin mining where advantageous and increase our shareholders' exposure to value-accreting assets. We are strategically positioned at the convergence of surging compute demand and Bitcoin growth, offering compelling potential for shareholder value creation. Now turning to the second quarter.

We continue to aggressively pursue further development of our data center business build-out and achieved a key milestone in our development plan. More specifically, we announced the hiring of Jonathan Gibbs as our Chief Data Center Officer. As we search for the right person to take leadership of this primary initiative for Riot, Jonathan's name repeatedly came strongly recommended to us by a number of different industry parties. The market for data center talent is incredibly competitive and professionals with Jonathan's level of expertise are in very high demand.

Jonathan's decision to join Riot and lead our data center platform is a testament to the unique opportunity set available to Riot and our ability to succeed. We are incredibly excited to have someone of Jonathan's caliber on board to drive our efforts. During the second quarter, we also continued to acquire additional land around our Corsicana site and now have a total footprint of 858 acres. Adding additional land ensures that we can fully utilize the large-scale access to power that we have on site without leaving any power stranded and therefore, maximize the value for Riot at which we believe is the premier data center development opportunity in the country.

We continue to see strong demand in the market, and we remain engaged in ongoing discussions with interested parties. With that said, in the second quarter, we also continued to make strong progress in our Bitcoin mining business, where we have made significant operational efficiency improvements that now place us among the most efficient operators in the industry, while also focusing on lower cost and maintaining a disciplined approach to capital allocation. Riot has also maintained our strong balance sheet, a long-standing key pillar of our business, ending the second quarter with over 19,000 Bitcoin and $330 million in cash on our balance sheet, representing $2.4 billion in liquidity today.

We continue to sell our monthly Bitcoin production in order to finance our ongoing operations while raising additional funds via a $200 million Bitcoin collateralized financing facility with Coinbase, allowing us to reduce issuance of stock through our ATM and fund our multiple growth opportunities, driving long-term shareholder value creation. I am proud of what we have been able to achieve in the second quarter. These results and the financial and operational strength of the company will allow us to continue aggressively growing our data center business in a way that will maximize long-term value for our shareholders.

I look forward to continuing to report on our progress throughout the rest of the year and beyond. With that, I would now like to turn the call over to Colin Yee, CFO of Riot Platforms, to present our second quarter financial update.

Thank you, Jason. I am pleased to present Riot's financial results for the second quarter of 2025. For ease of reference, we have highlighted key metrics on Slide 8, which presents a snapshot of key financial and operating metrics for the second quarter. During the second quarter, Riot increased its self-mining hash rate from 33.7 EH/s to 35.4 EH/s, representing a 5% increase over the course of the quarter, while global hash rate rose by 9% in the same period. Riot produced 1,426 Bitcoin in the second quarter, a slight decrease as compared to the 1,530 Bitcoin produced in the prior quarter, driven by the global network hash rate growing at a greater pace than Riot's deployed hash rate given our shift in strategic focus to developing our data center business.

Year-to-date for 2025, we have increased Bitcoin holdings per million fully diluted shares from 44.3 to 45.9 (sic) [ 47.4 ], representing a Bitcoin yield of 3.7% through the period ended June 30, 2025. For the second quarter, Riot reported total revenue of $153 million as compared to $161.4 million for the previous quarter, a 5% decrease quarter-over-quarter, primarily driven by lower Bitcoin production due to global hash rate increasing at a faster rate than our self-mining hash rate. Gross profit for the second quarter was $70.3 million as compared to gross profit of $73.6 million for the prior quarter.

Gross margin in the second quarter equaled 46%, flat with the prior quarter. Net income for the second quarter was $219.5 million or $0.65 per share compared to a net loss of $296.4 million or $0.90 per share for the prior quarter. This net income was primarily driven by mark-to-market adjustments due to the quarter end appreciation in Bitcoin price and marketable securities totaling $477 million. As a reference, the Bitcoin price at the end of the first quarter was $82,534, while the price at the end of the second quarter was $107,174, resulting in mark-to-market upward adjustment of $470.8 million for the quarter.

Net income for the quarter also included a $158.1 million loss on contract settlement as part of the Rhodium acquisition, depreciation and amortization expense of $83.2 million noncash stock-based compensation expense of $30.1 million and was positively impacted by the release of $26 million in restricted cash associated with the post-closing dispute settlement with Northern Data. Non-GAAP adjusted EBITDA for the second quarter was $495.3 million as compared to non-GAAP adjusted EBITDA loss of $176.3 million for the prior quarter, which included $470.8 million in unrealized gain on Bitcoin held.

Cash SG&A for the quarter was $45.8 million, including one time litigation expenses of $14.3 million and advisory fees of $2 million. Excluding these one time expenses, Riot's cash SG&A expenses equaled $29.5 million at the low end of our prior guidance of a run rate of $30 million to $33 million per quarter for 2025. For the second quarter, Bitcoin mining revenue totaled $140.9 million, in line with the prior quarter Bitcoin mining revenue of $142.9 million. Bitcoin mining gross margin for the quarter was 50%, an increase from 48% in the prior quarter. This margin expansion was driven by higher Bitcoin price.

Most notably, Riot's year-over-year hash rate utilization increased from 61% to 87%, demonstrating our strategic focus on improving operations across all of our sites even as we significantly scaled our operations and now positioning us among the most efficient operators in the industry. Direct cost to mine, excluding depreciation in the second quarter totaled $48,992 per Bitcoin, of which power costs amounted to $37,767 per Bitcoin or 77% of total direct cost per Bitcoin. Direct non-power costs, which include direct labor, miner insurance, miner and miner-related equipment repairs, land lease, property taxes, network costs and other utility expenses totaled $11,225 or 23% per Bitcoin mined, increasing quarter-over-quarter when direct non-power costs accounted for 18% of total costs.

This increase was almost entirely attributed to the 1-year anniversary of the completion of phase 1 construction at Corsicana and the resulting property tax bill assessment, which totaled $3.8 million for the quarter, adding an additional $2,650 per Bitcoin in direct non-power costs. We anticipate this cost will remain constant at $1.7 million per quarter going forward in our direct non-power costs. Despite this increase in our direct cost to mine, gross profit per Bitcoin mined for the quarter remained in line with the prior quarter given the higher average price per Bitcoin seen in the second quarter.

I would now like to turn the call over to Jason Chung, EVP of Corporate Development and Strategy.

Thank you, Colin. As we continue to develop our data center business, we believe that providing greater clarity on our Bitcoin mining business on a stand-alone basis is important information for the market. On Page 11 of our second quarter earnings presentation, we have outlined the underlying run rate profitability of our Bitcoin mining business for the second quarter of 2025. The column outlined in the middle of the slide provides a step-by-step walk-through of key profitability drivers for our Bitcoin mining business, ultimately culminating in run rate EBITDA for the quarter. Top line revenue drivers include the average global network hash rate, Riot's average operating hash rate, average network hash price and our total Bitcoin production for the quarter, which, when taken together, result in a reported second quarter Bitcoin mining revenue of $140.9 million.

As highlighted on the prior slide, total direct cost per Bitcoin for the second quarter was $48,992. And when applied to the 1,426 Bitcoin we produced during the quarter, equates to our reported Bitcoin mining gross profit of $71 million or 50% on a gross profit margin basis. In order to determine run rate cash SG&A for the quarter, we exclude from total SG&A, the impact of noncash charges, which are primarily comprised of stock-based compensation, cash SG&A related to our engineering business and nonrecurring expenses, which are primarily litigation and advisory related. Run rate EBITDA for our Bitcoin mining business for the second quarter equaled $45.6 million, representing a 32% margin.

These results are based on the average network hash price for the second quarter of $51 per petahash per day, while hash price today is currently closer to $60 per petahash per day. Our Bitcoin mining business demonstrates strong leverage to changes in hash price. And as an illustration, applying current hash price of approximately $60 per petahash per day to the second quarter results would have resulted in a 70% increase in our run rate EBITDA for the quarter. At the same time, we continue to focus on controlling and reducing costs.

Noncash charges, which are primarily comprised of stock-based compensation, are temporarily elevated at present, but will be meaningfully and dramatically reduced from mid next year onwards. And we will provide more detailed guidance on the expected reduction in stock-based compensation in the next quarter. As Colin previously mentioned, litigation expenses represent the bulk of our nonrecurring cash expenses for the quarter, constituting $14.3 million out of the total $16.3 million. While litigation expenses can be difficult to forecast, we continue to work to reduce these expenses as well.

For instance, our recent acquisition of Rhodium's assets and settlement agreement during the quarter have eliminated litigation costs associated with this dispute. It is important to keep in mind that these results are specific to our second quarter and historically, the third quarter has been the period during which we have typically seen the greatest reduction in direct costs and therefore, the greatest increase in profitability as that quarter is when we have typically been able to most fully employ our power strategy. I will now turn the call back over to Colin to continue with the second quarter financial update.

Thanks, Jason. Before diving into the financial results of our engineering business for the quarter, it would be helpful to discuss the underlying significant strategic benefits that this business brings to Riot. Our engineering business provides critical long lead time items directly applicable to developing large-scale data center infrastructure. By directly controlling this business, we can ensure timely, cost competitive availability of critical electrical components, representing a key competitive advantage in planning for ongoing development of both our Bitcoin mining and data center businesses at a time when other developers face supply constraints.

Further, through our acquisition of E4A Solutions last year, the engineering business also brings added in-house expertise in commissioning, operating and maintaining electrical infrastructure, allowing us to better maintain existing equipment, which reduces downtime and extending the life cycle of our equipment, which reduces additional CapEx spend. Direct savings to Riot on CapEx spend associated with ESS Metron since its acquisition in December 2021 already totaled $18.5 million to date, and we anticipate additional ongoing cost savings well into the future. Now let's dive into the financials.

During the quarter, the engineering business achieved a record in order bookings, taking our backlog to $118.7 million and setting the stage for a strong second half of 2025. During the quarter, engineering revenue totaled $10.6 million, a 14% decrease relative to the prior quarter revenue of $13.9 million. Total revenue excludes $5 million of intercompany purchases made in the second quarter by Riot for CapEx. With that, I would now like to turn the call back over to Jason Les.

Thank you, Colin. As I discussed in my opening remarks, Riot's strategy is to maximize the value of the megawatts that we currently have readily available. With the closing of the Rhodium asset acquisition during the second quarter, we now have access to an additional 125 megawatts of power capacity at our Rockdale facility. Following careful evaluation, we determined that the optimal use for this additional capacity in the immediate term is to upgrade it to support enhanced Bitcoin mining use.

As such, we have recently entered into purchase orders with MicroBT for new miners to be deployed at both Rockdale and Kentucky. In total, this order consists of 10 exahash of MicroBT's most efficient miner, the M60S++ with an efficiency rating of 15.5 joules per terahash. At current hash prices, coupled with Riot's low cost of energy, we anticipate a relatively quick payoff period on this purchase. Given the attractive economics and higher valuation multiples associated with data center leases to high-quality tenants, our long- term goal for this additional capacity is to transition it to data center use when appropriate.

These capital expenditures are fully funded through year-end 2025 with Riot's current cash on hand. As a result of this increase in 2025 CapEx, we are raising Riot's fourth quarter 2025 hash rate forecast from 38.4 exahash to 40 exahash, representing a year-over-year hash rate growth of 26%. A portion of the new miner order previously highlighted will be deployed during the first quarter of 2026. And as such, we are also providing an initial first quarter 2026 hash rate forecast of 45 exahash. This pace of hash rate growth is anticipated to allow Riot to maintain our approximate 4% share of the global Bitcoin network into the first quarter of 2026, while we continue to focus on the development of our data center business.

In January 2025, Riot formally announced our pivot to utilize the available 600 megawatts of power at Corsicana for data centers that serve high-performance computing. In just 7 months, Riot has accomplished the following: one, engaged Altman Solon to perform a comprehensive evaluation of the Corsicana site; two, expanded our Board to include key data center and infrastructure development expertise; three, engaged financial advisers to insist in our go-to-market strategy, financing and strategic partnership exploration; four, continued development of the 600-megawatt substation at Corsicana with 400 megawatts on track for the first quarter of 2026 and the second 200 megawatts expected to come online in the second half of 2026; five, building internal expertise, recruited and hired Jonathan Gibbs as Chief Data Center Officer, along with other veteran data center talent; and six, progressing on the basis of design for our data centers.

All of these steps are being taken in a methodical step-by-step manner in order to put us in the best position possible to secure a lease with the tenant and build a sustainable data center business. Further, when combined with our Bitcoin mining operations and resulting ability to monetize power land as well as our strong balance sheet, we are well positioned to expand our power portfolio further as attractive opportunities arise. Building a world-class data center team starts with the right leadership. In June, Jonathan Gibbs joined Riot as our Chief Data Center Officer, bringing more than 15 years of global experience leading end-to-end data center development and operations.

Throughout his career, Jonathan has driven multiple aspects of leading edge data center development, spanning capital planning, infrastructure delivery, operations and customer engagement across North America, Europe and Asia. Jonathan has led cross-functional teams responsible for design, construction, procurement, critical operations, ESG, EHS and sales engineering and has successfully led development of over 1 gigawatt of capacity, representing more than $17 billion in global investment. Most recently, he served as Executive Vice President of Product Delivery at Prime Data Centers, overseeing the execution of hyperscale and enterprise data centers across the United States.

Having the right expertise and experienced leadership in place is a critical step towards engaging potential data center tenants and negotiating leases from a position of credibility and strength. As highlighted on the prior side, building internal expertise represents a key milestone in the ongoing development of our data center business. And with Jonathan now in position leading the team, we continue to aggressively push forward in completing our basis of design and ultimately securing the lease in a manner that maximizes value for Riot shareholders. We are excited to have Jonathan at the helm of our data center platform and look forward to sharing more of his team's progress and vision in the quarters ahead.

Altman Solon's feasibility study identified the footprint of our existing site as a potential complicating factor to fully utilizing the entire 1 gigawatt of power availability at Corsicana for data center use in a lowest development cost way due to the different density requirements in comparison to Bitcoin mining. We quickly moved to address this. And in May of this year, we announced that Riot acquired a 355-acre parcel, expanding our available footprint for additional development. In July, Riot acquired a second 238-acre parcel adjacent to the previously announced 355-acre parcel, creating a 593-acre contiguous collection of land in close proximity to our existing site.

Collectively, Riot now controls 858 acres of potential development area in Corsicana. Our goal is to assemble a portfolio that ensures we have maximum flexibility to accommodate any design specifications and requirements of potential tenants. We are frequently asked about our time-to-market strategy and the "window of opportunity" that we see. Our observation of market dynamics suggest that power availability will remain the key constraining factor to the explosive demand for data center development that we are witnessing and that these dynamics will remain in place for many years to come.

On Page 20 of the earnings presentation, there are 2 charts. The chart on the left-hand side of this slide demonstrates from 2008 to 2023, U.S. on-grid energy demand growth was nearly flat, resulting in minimal investments into the grid infrastructure upgrades. Contrast that with projections of 2.2% compounded annual growth in demand for the next 5 years, representing a greater than 10x increase in annual demand relative to the prior 15-year period and demonstrating a significant and growing gap between this increased demand and more limited growth in supply. Concurrent to this growing gap in demand for power and relative to supply, time lines to procure power in key markets across the United States are significant, with analysts pointing to lead times in the Dallas and Austin markets where our Corsicana and Rockdale sites are located of 36 and 42 months, respectively.

Riot's fully permitted and readily available power located in important in-demand markets positions us to be in the right place at the right time to capitalize on these market dynamics to the benefit of our shareholders. In closing, we have many advantages that have put us in an incredible position because we offer a unique combination of significant scale of readily available power capacity in key high-demand jurisdictions, experienced, credible hyperscale data center leadership and development capability, strong balance sheet underpinned by more than 19,000 Bitcoin and $330 million in cash and significant access to capital markets.

Large-scale efficient Bitcoin mining operations generating hundreds of millions of dollars in revenues and cash flows and battle-hardened and experienced management and operations teams. With this framework, our mission is clear. Riot will maximize value across our entire power portfolio with a view to ensuring full utilization of our available power capacity and pipeline, leaving no stranded capacity behind, progressively shift power capacity towards data centers, strategically expand our power assets, utilizing Bitcoin mining where advantageous and increase our shareholders' exposure to value-accreting assets.

We are strategically positioned at the confluence of surging compute demand and Bitcoin growth, offering compelling potential for shareholder value creation. We will now open the call up for questions. Operator?

[Operator Instructions] Our first question coming from the line of Greg Lewis with BTIG.

There's definitely a lot to chew through on the HPC opportunity ahead for Riot. But I did want to ask about the decision. It was clearly a good quarter for generating Bitcoin. But clearly, from the action, we took that Bitcoin generation to really -- we sold that to monetize.

Could you talk a little bit about that decision to do that and how you're thinking about the HODL strategy in the back half of the year or even longer term?

Thanks for the question, Greg. This is Jason Chung. Maybe I'll take a stab at that one. So I think this quarter is an interesting representation of how we think about our financing strategy and the different levers available to us.

And just looking at for the past quarter, the 2 levers that we exercised most heavily, one was the sales of our Bitcoin production. And the second was leaning into our Bitcoin Stash to borrow and enter into the Coinbase facility for $200 million. The sale of Bitcoin production allows us to more than cover our operating costs and therefore, frees up the additional capacity or minimizes our requirement to issue into the ATM and really allows us to focus any financing raised through that very specifically towards growth opportunities, which we believe are going to be value accretive to our shareholders.

And so I think that's kind of how we think about -- thought about things for the quarter and probably a good reflection of how we currently think about things as well. As Bitcoin prices increase, that does give us additional room for comfort around our leverage levels and the ability to consider expanding the amount of financing we draw upon there as well. So I think as we continue to see how Bitcoin prices evolve, you'll see us continue to take advantage of different market conditions as we think about what's optimal from a capital perspective for the quarter.

Okay. Super helpful. And then just realizing you're probably limited in what you can say, maybe we can talk a little bit about what we're seeing in the market in terms of the available power transactions or availability to electricity signing for -- with HPC. If you could kind of talk to the pricing dynamics, how things have been trending?

I feel like more recently, it was kind of in the $120 a megawatt range is some of the things that we've been hearing. Kind of curious if that's kind of where you're hearing the market is. And then really the question I have is, as we think about sizing, is there a premium that you're seeing in terms of having larger amounts of power available, i.e., if we're looking at a couple of hundred megawatt power deal versus, say, 0.5 gig plus, is there any kind of premium for that larger power deal just in thinking about how potential transaction could shake out?

Yes, Greg, this is Jason Les. I think at a high level, we're seeing very robust demand in the data center market. Our view continues to be that what exists out there in terms of power and infrastructure is really not close to sufficient to meet what's forecasted demand and hyperscalers continuing to announce higher levels of CapEx budgets, they have serious demands for more data center capacity that really cannot be satisfied by new power that's expected to be available. What we see is the implications of the AI arms race being very clear here.

There's a trend for more compute demand. That's very clear. So we believe demand is going to continue to be robust, and we are building a business here, building a platform to be able to serve it. As far as monthly rates go, I think there's a lot of different components that go into what an ultimate lease might be.

And it's important to look at a deal like this as a sum of all of its parts, maybe instead of just a single metric. You'll see a range of rental rates, and those will have somewhat of a correlation to the type of tenant that you're getting. There's a bit of credit risk often built into what those monthly rates are. And you'll see term and other components of these agreements.

So I think it's important to look at these in all of the parts that comprise them and not necessarily just what that monthly rate will be. It can be a range and other components could enhance that deal or make that deal worse off from the perspective of the lessor. Now I think the last part of your question was, is there a premium for large-scale power? I don't know if I can comment right now if there's a premium for that power.

But what I can say is that there's a premium of interest for large scales of power. So for tenants, everyone is massively scaling. Hyperscalers are looking to take down gigawatts and beyond now. And as everyone else increases their demand for compute, we're now seeing neoclouds taking down capacity at levels that hyperscalers once did and now enterprise tenants taking down capacity at those significant levels as well.

So what any long-term growth-oriented tenant is going to be thinking about is their pipeline for expanding instead of having to solve a problem for capacity over and over and over again with different providers, what we see is customers who are interested in capacity available beyond just what their initial lease might be. So when you talk about a premium for capacity, that's how we think about it. There's a premium that is, in essence, garnering customer interest because that -- because they see an ability to expand beyond just what an additional phase of a development or lease might be. And that we have found is very helpful for having productive discussions.

Our next question coming from the line of Nick Giles with B. Riley Securities.

I think it's become clear that Riot is not going to rush to get a deal done. So I want to commend you for your measured approach. But I think in recent months, forming a basis of design has been at the core of Riot's efforts towards the data center side. And so I was wondering if you could provide any detail on what aspects of that document are clearly defined versus ones you may still be working on?

I think factors that come to mind are cooling resources, redundancy, security layout. Any color that you can add there would be great.

Yes. So first off, bringing an experienced data center executive like Jonathan Gibbs on board alongside other talent that's been recruited with significant experience in data center development has aided us considerably in building the basis of design. Of course, it is this team -- this data center team's project and objective to accomplish here. And this basis of design is very foundational to being able to go to market.

What we're putting together here is the technical strategy, design elements that we can then take and then have something concrete to be able to discuss with potential customers, with potential tenants to ultimately arrive at a more customized design and then a lease. So we see this as one of multiple milestones, but a very key milestone in progressing towards getting a lease here. We have been working on this quite a bit. Jonathan and his team has rather, and there's been significant progress made already.

We expect that we will be able to complete this basis of design by the end of this quarter -- by the end of the third quarter, that is, and be moving on to next steps in our data center strategy.

Jason, thanks for all that color and that reminds me I want to congratulate Jonathan on his appointment. My second question was, obviously, long lead times are a key determining factor in development time lines. So have you submitted any RFPs to contractors? I mean, how much is the tariff landscape ultimately playing into the timing of that?

So first, for some of the critical infrastructure that's needed to build this capacity, we have already secured. I'm referring to the 600-megawatt substation that's being built that's expanding the site to 1 gigawatt. We have already procured that equipment. That equipment is already arriving, and that is going to take our Corsicana site to 1 gigawatt in 2026. So we are very well positioned on that critical equipment there.

As far as other equipment goes, we are pretty confident in the steps that we're taking to prepare for that. We are looking at long lead times for other equipment, but the time lines for these are not surprising to us. That's kind of expected. And the process of procuring these long lead items is already underway.

And with Jonathan and his team on board, we feel like we're approaching this in a very strategic way. And ultimately, we don't believe that the lead times for an equipment is going to impact our ability to secure a lease.

Our next question coming from the line of Darren Aftahi with ROTH Capital.

Just following up on the master site design time frame being completed by the end of this quarter. Can you speak to the potential tenants that you're engaged with? And I guess, like how critical is that master site design in terms of their willingness to kind of continue negotiating? Said another way, like is that something that will accelerate negotiations for you?

Or are there folks that have already kind of parallel diligence the things they need while waiting for that master site design? And then my second question on Rockdale, I know you're upgrading some rigs there. But can you just give us some general long-term thoughts on what that campus potentially could be used for other than Bitcoin mining and kind of where your head is at? Is it you have too much to [ Audio Gap ] with Corsicana and it's kind of back burner or you could do these things simultaneously and potentially market all your power as one campus.

Yes, Darren. So the first part of your question, so one thing I want to make clear is we are making a basis of design that we believe can serve a wide range of customers. It can serve hyperscale customers, it can serve enterprise customers or neocloud customers. What we want to do is maximize our flexibility.

I think that's a theme you've heard us talk about on our earnings calls a couple of times now, taking different actions, making moves in order to maximize the flexibility of our site, of our data center [ Audio Gap ] and secure the best possible deal here. And if you're talking about engaging with serious counterparties, this is the type of information that they [ Audio Gap ] you to come to the table with in order to advance discussions substantially. And that's why we view the building out of this team here, especially led by Jonathan Gibbs and his onboarding is very critical and a very important step we've made to building up this platform.

I can't comment on ongoing discussions. I would say that all types of customers are different and maybe approach conversations in different ways and [ Audio Gap ] milestone in order to have serious discussions here. So we look forward to sharing more about this with the market as it's completed and being transparent and sharing our milestones and our road map to building out our platform here and ultimately securing a lease. With respect to long term -- I'm sorry, with respect to Rockdale, our primary focus is scaling our data center business and maximizing the value of all of our power assets.

Because of that, because of the economics that you can get with data center leases and how the market values that, data centers are the ultimate ideal use for us for all of our power capacity. What's great about Riot is we have a lot of power capacity to work with. Megawatts alone, which is our available capacity at Corsicana, that represents a very substantial data center campus in its own. At the same time, we are open to doing -- finding deals at Rockdale as well.

I think what we're just doing right now is prioritizing what we see as the [ Audio Gap ] with. And as we get our data center platform off the ground and we continue to make more progress, then that makes all of our power assets -- that positions all of our power assets in the pipeline for growth of the data center platform ultimately. So you can think of our strategy as using Bitcoin mining at sites like Rockdale to monetize that power to ensure that no power stranded and wasted, turning that into meaningful cash flows for the company and then ultimately looking to transition that capacity to data center leases when the time is right.

Our next question coming from the line of Brett Knoblauch with Cantor Fitzgerald.

Maybe an update on kind of your Bitcoin mining outlook. I know you guys kind of raised guidance for the end of this year in the first quarter as well. Network hash has kind of been stubbornly continuing to go up. Maybe high level, where do you see network hash going?

Is there a level where you think maybe it kind of plateaus a bit? And I know you talked about being 4% share. Is that kind of like a goal that you guys want to maintain for the long term? Or how should we think about that?

I think the 4% share is not a mandate that we have. That's something that we see ourselves being in just based on the growth that we've outlined and kind of a near-term estimate of global network hash rate in the next 6 to 12 months. But by no means are we intending to always maintain a certain percentage. But going back to the first part of your question, I think Bitcoin miners will face the same types of scaling challenges that data centers are.

There's very limited amounts of power. And from what I think we're seeing and we're excited about is data center customers are paying a lot more for that than Bitcoin miners ultimately would. So while Bitcoin miners have other options for power that data centers don't, I think they will also be constrained in how they scale, which has the potential to have a positive impact on hash price in the future. At Riot, what we're focusing on is maximizing the value across our power portfolio, trying to maximize the value of all of our megawatts not straining any capacity.

So what we shared with the growth that we have going on in Kentucky and the hash rate growth that we have at Rockdale, that -- those are moves in accordance with that strategy, and I think represent measured growth of our Bitcoin mining segment. We're looking at approximately 26% year-over-year growth from '24 to '25 and then approximately 10% growth from 2025 to 2026.

Awesome. That's helpful. And then maybe just on the maybe Corsicana, I think a lot of the conversations we've had kind of suggest that is maybe one of, if not the best potential AI HPC data center sites out there. Are you guys getting kind of like similar feedback when you guys are looking at potential customers or kind of what to do with maybe the remaining 600 or full gigawatt there?

What we're focused on with launching this data center platform is building a strong foundation. We want to get off on the right foot here. And building that strong foundation means getting the right deal on which we can build the pipeline on top of from the start. Now that doesn't mean that we need all of that 600 megawatts or sign a lease for all of that 600 megawatts to build that first foundation -- that first step to build that strong foundation.

We are looking at this capacity and building it out as a phased approach. We see building this out in different segments, and we'll be talking about that more in the future. And the fact that the site has so much capacity means that ultimately, there may be one tenant that wants all of that. I discussed an earlier question.

The fact that there's so much growth in one site is, we believe, very interesting to lots of customers out there who have very robust demand forecast. So it's to be determined how this is all segmented out, but we are approaching the market with a design that we believe can serve a wide range of the market, hyperscale customers, enterprise customers and neoclouds. And what's important to us is getting this off on a solid foundation to start. And then ultimately, like I stated again, there's lots of room to grow here and the potential to do a larger deal from there.

Our next question coming from the line of Paul Golding with Macquarie.

I wanted to ask about Corsicana and drill down to some of the infrastructure components. I noticed in the slide on 2025 CapEx that there's a waterline project expected to be completed in Q2 '26. And just overall looking at the substation development line item for Corsicana, I was wondering if you could expand on any of the infrastructure components for Corsicana that are maybe factoring into the conversations still pending with potential tenant counterparties as opposed to these deals having been signed already. And also just to help us understand the extent to which water access has already been secured given the water retention pond that you have and the importance of that for HPC and AI liquid cooling.

So starting on water, Paul, as you noted, we have a significant size retention pond that allows us to use a lot of the water that's just naturally generated on site. It's Texas, but still gets a lot of rain. We have secured the plans and the approvals to build out the water line, and that will ultimately -- that's a part of giving us maximum flexibility to serve customer demands. What we're seeing on the data center technology side is cooling technologies becoming more and more water efficient as time goes on.

In order to be flexible, we didn't want to bank on that. So we're securing enough water that we believe would be ample for a full 1 gigawatt development if someone needed that amount of water in order to achieve the cooling strategies that they have or that they require. As far as the infrastructure for Corsicana, I think we are in a great position and probably have a considerable leg up on what other data center developers might be at, at this stage. We've already made the decision years ago really to be procuring this equipment.

So it's already coming in now. That's significantly, I think, derisk the amount -- I'm sorry, derisk the time line to getting that power online, also combined with the fact that we have this approved already. We have FEA for this already. It is all baked in and ready to go.

So That, I believe, puts us in a great position when we have conversations with tenants because this power is not theoretical. This power isn't pending certain steps happening. This power is coming in the next 6 months and scaling up from there.

Great. And maybe a follow-on to that. We've talked on the call around -- about price, potential pricing in the marketplace and premiums or premium for demand. You've spoken on the call about data center customer requirements, and that's factoring into this build concept.

As you have these conversations, just wanted to verify, is the plan still or is what you're pursuing still the option to construct the facility and the power infrastructure for these tenants in a yield-on-cost or build-to-suit scenario? Or are you getting inbounds? Or are you considering inbounds where someone else is building it and leasing the power and the infrastructure?

So our philosophy at Riot has been to maximize the value of our assets. And we believe that build-to-suit model is going to be the best way to maximize the value of our portfolio of assets, especially at Corsicana. That being said, we do not intend on building out the site beyond an initial stage without a lease. We're not looking to build out a site on spec.

We believe that by finalizing the design here, understanding what that is with customers and then being able to take initial steps to get things off the ground, which we already have done with building out the substation in the water, these would be foundational steps in any data center, we are willing to invest in order to get things moving off the ground and getting to the point of getting the lease, but we are not looking to build a suit, a site on spec and take on all that risk without having a lease in hand.

Our next question coming from the line of Reggie Smith with JPMorgan.

Jason, congrats on the quarter. I guess I'd like to follow up on the last question, and I appreciate you guys wanting to actually build to suit. But I guess my question is if there's more demand today for people just looking to buy power outright. So like if that were your strategy, do you think this plot or your capacity would have been sold now if that makes sense?

I'm trying to figure out like is the hang up that -- or the delay in the deal being done, the fact that there may be some haggling over whether a miner just sells power outright versus a build-to-suit type of situation? And then I have one follow-up question.

Yes, Reggie. So we're really -- we believe what we have is incredibly valuable. And I think all the data that we're seeing in the market on data center leasing validates that belief. So it's important to us to maximize the value of that.

If you're talking about doing something like leasing-powered land, yes, there is a ton of demand for lease-powered land, but the value that you can expect to extract from that is going to be, I think, pretty significantly mismatched with what I think investors are expecting from this type of data center opportunity. With the assets that we have, with the balance sheet that we have and now with the team that we have, we are in a great position to build a data center platform and be able to pursue the value maximizing approach that we see with this build-to-suit model. We are open to anything that will maximize the value.

So we're not close up to any type of discussion, but this is the avenue that we see as the best pursuit going forward, and that's why we're approaching things in this manner.

That makes sense. And if I could ask one more question. One of the points that we've talked about that we thought has distinguished you guys from other operators is that you're located so close to Dallas and Austin. As you kind of assess or appraise your assets, how important is that distance from one of those cities in determining the attractiveness of partnering with a Riot versus someone else?

Is there still a premium for location, I guess, is what I'm asking you.

Yes, Reggie, the location is very important. Dallas is one of -- is a Tier-1 data center market. It is one of the most in-demand data center markets in the country. That's why I think Corsicana is so valuable.

You have the great connections, low latency and ability to get people and talent to that site relatively easily as -- I'm sorry, as opposed to more remote locations. For that reason, we think Rockdale is also an attractive site. Now Austin, San Antonio, those aren't Tier-1 markets yet, but with the investments that we see in data center CapEx, with the revenue forecast for AI software and the margins that AI software service providers are forecasting and able to get, we think that will change over time. So by having these 2 sites, both near one Corsicana near major market today and Rockdale near what I would say is an emerging up-and-coming market, I think, makes those sites very attractive and allows them -- because of those elements allows them to command perhaps better economics than other projects out there.

Our next question coming from the line of Mike Grondahl with Northland Capital Markets.

Congratulations on hiring Jonathan Gibbs. What would you say his top 2 priorities are this summer and fall?

So our #1 priority is building this data center platform bar none. And I would break that down into 2 priorities on accomplishing that. One is building out the team. Jonathan is bringing the critical leadership to making that happen.

We've added other individuals that are veterans of data center design and development. We are bringing more talent on as we speak. This is important because we want to build up our expertise. We want to build up our platform so it looks and feels and acts like a way that hyperscale and enterprise and neocloud customers expect.

So that's the number one priority. I guess second priority in parallel, I'm not ranking one over the other, is completing this basis of design at Corsicana. This will allow us to have more substantive discussions with potential tenants, allow us to advance the design further, work in different customers when necessary and really get the critical parts of negotiations happening. So 2 priorities. #1 priority, building the data center platform, [ Audio Gap ] building the team and building the design.

Our next question coming from the line of tephen Glagola with Jones Trading.

How will the new requirements in Texas Senate Bill 6, such as like grid upgrade, cost sharing, mandatory backup generation disclosure, curtailment obligations and so forth affect the cost structure and operations of your mining and your HPC activities at both Corsicana and Rockdale.

So first, important to note is that both of these sites, we have FEAs already in place. So we do not expect to need to renegotiate those FEAs in any way as a result of this change or as a result of this new legislation. This legislation launches a lot of exploratory work and information gathering. That's something that Riot, our very capable public policy team, our power team and our industry partners are all very involved in.

One of the parts of SB6 is looking at the 4CP program. That's something that Riot participates in order to reduce our transmission charges. That program may see changes as the working groups from this legislation progress. We don't -- we hope and we're working to ensure this doesn't have too much of an impact on our transmission charges ultimately.

There's lots of different ideas of how changes to that program could take place. So it's really too early, I think, to speculate on that. As far as the other requirements go, I think that is probably going to impact new FEAs and new interconnection agreements more than it is us, but it's something we're staying very close to and making sure that we're good stewards of the grid, we're good industry partners and we're doing what we can to support the grid and give them the data and the reliability that they need.

I'm showing no further questions at this time. I will now turn the call back over to Philip McPherson for any closing remarks.

Thank you, operator, and thank you, everyone, for joining us on our second quarter call. We look forward to updating you on further progress on our business on the third quarter call in October.

This concludes today's conference. Thank you for your participation, and you may now disconnect.