Core Scientific Closes $3.3 Billion Bond Deal but Hyperscaler Exclusivity Lapses, Leaving Key Sites Uncontracted
Q1 2026 Earnings Call — May 6, 2026
Core Scientific entered its first quarter earnings call with a headline-grabbing capital raise in hand but a conspicuous gap in its commercial story: the exclusivity agreement with a named hyperscaler across its most prized development sites, Pecos and Muskogee, has expired without a signed lease. The company moved quickly to frame the situation as a strength rather than a stumble, but investors should weigh both the genuine progress on financing and construction execution against the continued absence of a second anchor tenant.
The $3.3 Billion Bond: A Structural Win with Important Mechanics
The most consequential development of the quarter is the closing of a $3.3 billion project bond financing at a 7.75% interest rate, secured against the contracted CoreWeave cash flows. After closing costs and a required debt service reserve, net proceeds available to the company come to approximately $2.9 billion. CFO Jim Nygaard described the rate as "highly attractive cost of capital for a financing of this scale," and the structure is worth understanding in some detail. The bonds employ a lockbox mechanism where CoreWeave's lease revenues flow directly into a designated account that first covers operating expenses, then debt service, and then permits the distribution of the vast majority of remaining proceeds up to the corporate level. Unlike conventional project finance where proceeds are siloed to fund only the specific secured project, this structure allows Core Scientific to redeploy capital broadly across new developments. The practical implication is that roughly $2 billion of total capital expenditures planned for 2026 — including the Hunt County, Texas site acquisition that closed the day before the call, the Polaris acquisition at Muskogee, and early-stage development across roughly one gigawatt of new capacity — can be funded from this raise without needing new equity issuance or customer contracts in place first.
Hyperscaler Exclusivity Expired: One Door Closed, Three Opened
The most sensitive disclosure of the call was the expiration of exclusivity with a single hyperscaler across the Pecos and Muskogee sites. CEO Adam Sullivan was careful not to characterize this as a deal failure, noting that "three hyperscalers immediately engaged on those same sites" once the window closed and that the original counterparty "is still at the table and still interested." The episode nonetheless reveals a tension in Core Scientific's go-to-market strategy: the company spent a meaningful period keeping two of its most strategically valuable assets off the market for a deal that did not close. Sullivan acknowledged the lesson directly, saying the company has "migrated to a milestone arrangement method" for future exclusivity discussions that requires clear progress checkpoints and allows Core Scientific to bring sites back to market if cadence slows. "In a market like this, we will not keep high-value assets off the market longer than necessary," he said. Whether this represents a durable policy shift or reactive positioning will become clear in the quarters ahead.
CoreWeave Build: 243 Megawatts Billing, 590 Megawatt Target Intact
On the execution side, Core Scientific delivered a tangible milestone, announcing that 243 megawatts of billable capacity is now live across the CoreWeave portfolio, including the completion and full handover of the Marble, North Carolina facility at 65 megawatts and Dalton, Georgia Phase I at 30 megawatts. COO Matt Brown reiterated the target to deliver more than 450 billable megawatts by end of summer and the full 590 megawatts contracted under the CoreWeave agreement by early 2027. From a financial standpoint, Nygaard noted that 243 megawatts of billing equates to more than $350 million in annualized colocation GAAP revenue, with GAAP recognition occurring on a straight-line basis over the 12-year lease term, which pulls forward the economic value of built-in escalators. The company also raised its target cash gross profit margin on the CoreWeave contract to 80% to 85%, up from the original 75% to 80% guided when the deal was first signed two years ago. Nygaard attributed the revision to improved visibility into the actual cost structure now that a substantial portion of contracted megawatts are live and billing, describing it as "a true-up of that experience."
Pecos and Muskogee: Gigawatt-Scale Ambitions, But No Tenant Yet
The two centerpiece development announcements of the quarter were expansion plans at Pecos, Texas and Muskogee, Oklahoma, each charting a path to 1.5 gigawatts of gross power capacity. At Pecos, the company is converting the existing Bitcoin mining site to high-density colocation with construction already underway and a 12-month runway to ready-for-service on the initial 185-megawatt, 431,000-square-foot building. Long-lead equipment has been secured, vertical construction is beginning with precast walls arriving on site, and the company is building a new regional fiber interconnect in Midland, Texas linking back to the campus. The power expansion strategy includes securing an additional 300 megawatts of grid capacity and progressing a natural gas pipeline to the campus to enable behind-the-meter generation. At Muskogee, a general contractor is already mobilized and the first 82.5-megawatt building is under development with initial delivery expected in the second half of 2027. The Polaris acquisition adds approximately 440 megawatts of power rights to the site, and Oklahoma's recently enacted behind-the-meter legislation, publicly championed by Governor Kevin Stitt, provides a regulatory tailwind. Brown noted that air quality studies for behind-the-meter deployment are already "far down the path" at both sites. The critical caveat: neither site has a signed customer contract, and the company is deliberately building ahead of lease execution.
Building Ahead of Contracts: Strategic Differentiation or Speculative Risk?
Sullivan laid out the company's rationale for pre-contract capital deployment with some precision. The strategy is to take each first data hall at new sites to full ready-for-service status — securing labor, trades, and long-lead equipment — so that when a customer signs at any point along the timeline, the on-site workforce can simply continue without interruption. "We are not waiting for deal negotiations to conclude before advancing sites," he said, arguing this shortens perceived delivery timelines for customers to the 12-to-14-month window that hyperscalers are "actively trying to solve for." Brown reinforced the competitive logic, noting that construction labor is "one of the primary constraints in the market" and that having large general contractors already mobilized across multiple sites is a genuine advantage. Sullivan argued directly that showing active construction photographs and equipment order confirmations "dramatically changes the dynamic" of customer discussions compared to presenting undeveloped land. The risk, unacknowledged on the call, is that the company is now deploying significant capital across multiple sites simultaneously without signed revenue commitments on any of them outside the CoreWeave portfolio.
Greenfield Over Brownfield: An Operational Lesson with Lasting Implications
One of the more instructive exchanges of the call came in response to a question about lessons learned from the CoreWeave build. Brown was candid that brownfield site conversions — retrofitting existing Bitcoin mining facilities — proved far more complex than anticipated. "Brownfield sites are highly unpredictable. They require a lot of customization. It's a lot of effort to try to retrofit an existing building," he said. The company has responded by standardizing all forward development on greenfield sites with a repeatable basis design, which Brown said enables supply chain leverage and delivery predictability. The team executed more than 150 design changes across the CoreWeave portfolio during construction and treats that iteration as institutional knowledge that competitors building their first high-density facilities will need to acquire the hard way. The Pecos conversion from mining to colocation, however, complicates this narrative slightly and will be worth monitoring as a potential source of execution friction.
Bitcoin Mining in Managed Decline
The company continues to run Bitcoin mining operations primarily to offset contractual power costs during the transition, but the business is in explicit wind-down mode. Nygaard flagged a "meaningful step down in miners online in the second half" of 2026, and Sullivan indicated that by year-end, only one or possibly two sites will still be operating Bitcoin mining infrastructure. The company has already monetized the majority of its Bitcoin treasury holdings, retaining only a modest residual balance. Mining is no longer a strategic priority and is being managed as a cost mitigation tool rather than a revenue driver.
Commercial Pipeline: Breadth Without Specificity
Sullivan described the company's commercial conversations as having "increased significantly since the beginning of the year," spanning hyperscalers, chip makers, AI labs, and Neo-cloud providers. He characterized the company as being in discussions with "all of the major groups" and noted that all five non-CoreWeave sites have first data halls targeting RFS in 2027, a spread of geographies and scales he described as "an incredibly unique position." On pricing, Sullivan said contract rates continue to firm due to labor and equipment inflation, suggesting new leases will price above earlier vintage deals. For Neo-cloud and AI lab customers, he flagged that these groups "often require additional credit support" and that the company is actively working with financing partners on structures to accommodate them. No specific deal timelines were provided for any customer segment, and the company offered no formal guidance on when a second anchor tenant announcement should be expected.
Core Scientific Deep Dive
Business Model and Monetization Strategy
Core Scientific is executing a complete business model metamorphosis. Historically recognized as one of the largest digital asset mining firms in North America, the company is systematically winding down its proprietary digital currency operations to repurpose its immense power footprint for high-performance computing and artificial intelligence workloads. By the end of the first quarter of 2026, the company began liquidating virtually all of its remaining digital asset treasury, which previously stood at roughly 1,900 tokens, signaling a permanent departure from the volatility of cryptocurrency mining. The new monetization model is centered on High-Density Colocation. Instead of capturing variable block rewards subject to global network difficulty and fluctuating hash prices, Core Scientific signs 12-year, fixed-rate, dollar-denominated leases with major artificial intelligence hyperscalers. The company provides the physical real estate, direct-to-chip liquid cooling systems, and massive power loads required to operate next-generation semiconductor clusters. This fundamentally transforms its revenue profile from a cyclical commodity producer to a highly predictable digital real estate utility.
Key Customers, Competitors, and the Supply Chain
The company currently navigates significant customer concentration, with the artificial intelligence hyperscaler CoreWeave serving as its undisputed anchor tenant. CoreWeave has contracted for 590 megawatts of critical IT load, an agreement projected to generate over $10 billion in cumulative revenue over its lifespan. However, Core Scientific is actively mitigating this reliance, advancing negotiations to allocate approximately 400 megawatts of near-term capacity to new third-party hyperscaler and enterprise clients throughout 2026. On the competitive front, the landscape is defined by other legacy infrastructure miners attempting identical transitions. Peers such as Iris Energy, which recently secured a $9.7 billion cloud deal, alongside TeraWulf and Hut 8, are aggressively competing for hyperscaler contracts. The supply chain for this industry is heavily constrained not by real estate, but by high-voltage electrical equipment and grid approvals. Key suppliers include regional transmission organizations, primarily the Electric Reliability Council of Texas, and original equipment manufacturers of long-lead electrical switchgear and liquid cooling infrastructure. The procurement timeline for these critical components dictates the pace at which raw land can be converted into billable computing power.
Market Share Dynamics and the Infrastructure Premium
Measuring market share in the emerging high-density colocation sector is distinctly determined by energized power capacity rather than conventional metrics like server count or software licenses. Within the specific niche of infrastructure conversions, Core Scientific commands a dominant leadership position. While the company previously controlled roughly 18 exahashes per second of digital mining capacity, ranking it among the top ten globally, its strategic transition has repositioned it as a top-tier independent data center operator. The company holds a massive 3 gigawatt pipeline of leasable customer power. At present, Core Scientific has successfully energized approximately 350 megawatts of high-performance computing capacity, with nearly 245 megawatts generating active billing revenue in the first quarter of 2026. This sheer scale of deployment places the company structurally ahead of peers who are still navigating early-stage zoning approvals and initial pilot facilities.
Competitive Advantages and Barriers to Entry
The primary competitive moat for Core Scientific is speed to market, an asset currently commanding a premium over pure capital availability in the accelerated computing sector. Traditional greenfield data centers typically require 36 to 60 months to secure grid interconnections, land permits, and construction approvals. By leveraging existing pre-approved power loads from its legacy operations, Core Scientific has compressed its Ready-For-Service timeline to just 12 to 14 months. The company operates under an aggressive capital deployment strategy, termed Operation Forward Observer, wherein management procures long-lead electrical equipment and advances site commissioning before final customer contracts are even signed. This speculative procurement is a calculated operational risk that dramatically shortens delivery windows, fulfilling a critical requirement for hyperscalers desperate to deploy tens of thousands of next-generation graphics processing units immediately.
Industry Dynamics: Opportunities and Structural Threats
The foundational opportunity underpinning this sector is the severe power deficit across the North American electrical grid. Artificial intelligence workloads require rack densities exceeding 100 kilowatts, far surpassing the 10 to 15 kilowatt limits of traditional enterprise data centers. This dynamic forces compute demand directly into the hands of operators with pre-existing access to heavy-industrial power footprints. However, this transition is fraught with structural threats, primarily the staggering capital intensity required to build out the necessary infrastructure. Converting or building high-density facilities currently costs approximately $11 million per megawatt. To develop a 1 gigawatt pipeline, an operator must secure $11 billion in total capital, which necessitates relying on project-level debt for up to 80% of the build cost. If hyperscaler demand proves cyclical rather than secular, or if customer credit quality unexpectedly deteriorates, operators carrying such heavy leverage profiles could face severe liquidity challenges.
New Technologies and Growth Drivers
To circumvent grid congestion and accelerate capacity growth, Core Scientific is actively integrating behind-the-meter power solutions into its future developments. A prime example is the company’s Pecos, Texas facility, which is currently undergoing a massive expansion aimed at reaching 1.5 gigawatts of total gross capacity. Management is advancing the integration of localized natural gas infrastructure to generate power onsite, effectively bypassing the bottlenecked regional transmission grids. Similarly, the Muskogee campus is being scaled to 1.5 gigawatts through a combination of grid expansion and private behind-the-meter generation. Additionally, the development of Application-Specific Data Centers engineered specifically for advanced thermal management is a major technological growth driver. At the Denton campus alone, this custom facility architecture is currently supporting over 50,000 extreme-density graphics processing units, proving the viability of their proprietary cooling integration.
The Threat of Disruptive New Entrants
The concept of new entrants in this sector presents an unconventional dynamic, as traditional data center real estate investment trusts are paradoxically the legacy players struggling to adapt, while independent power aggregators are the disruptive forces. Traditional operators are fundamentally constrained by older, air-cooled facility designs that cannot be easily retrofitted for the extreme thermal demands of modern accelerated computing without prohibitive capital expenditures. The true disruption comes from heavy-industrial energy developers and specialized infrastructure firms bypassing the legacy data center model entirely. By building dedicated computing islands directly adjacent to natural gas, nuclear, or large-scale renewable generation sites, these independent power producers can vertically integrate and offer co-location services directly to hyperscalers, challenging the dominance of operators reliant on municipal grid structures.
Management Track Record and Capital Allocation
The executive leadership team has executed one of the most rigorous corporate turnarounds in recent infrastructure history. Emerging from Chapter 11 bankruptcy in early 2024, management ruthlessly streamlined operations and recognized the structural ceiling on digital asset economics. Their defining strategic moment occurred in the second half of 2025. CoreWeave, seeking to vertically integrate its operations, launched a $9 billion all-stock takeover bid for Core Scientific. Recognizing that the prevailing market significantly undervalued their immense power pipeline, management and major shareholders decisively blocked the acquisition in October 2025. This decision proved highly accretive to shareholder value. The company maintained its independence, preserved the long-term equity upside of its 3 gigawatt pipeline, and successfully priced a $3.3 billion senior secured notes offering in May 2026 to independently fund its next wave of expansion. Management has demonstrated severe capital discipline, pivoting definitively away from commodity operations to focus entirely on high-margin, contracted infrastructure returns.
The Scorecard
Core Scientific has successfully architected a fundamental operational pivot, transforming from a highly cyclical digital asset miner into a premier provider of high-density infrastructure for artificial intelligence workloads. The company's strategic decision to abandon its 18 exahash mining operation in favor of 12-year, dollar-denominated colocation leases fundamentally upgrades the quality, predictability, and duration of its cash flows. By leveraging its vast, pre-approved grid interconnections and employing an aggressive forward-procurement strategy for long-lead equipment, the company has established a formidable speed-to-market advantage, compressing delivery timelines to 12 months in an industry accustomed to multi-year delays. The sheer scale of its current deployment, encompassing 350 energized megawatts and a forward pipeline of 3 gigawatts, solidifies its position as an indispensable infrastructure partner to major hyperscalers facing acute power shortages.
However, this aggressive expansion introduces a distinctly new risk profile characterized by extreme capital intensity and relentless execution pressure. Building out the targeted capacity requires roughly $11 million per megawatt, demanding billions in project-level financing and resulting in highly leveraged capital structures. The company’s immediate future is also heavily tethered to the creditworthiness and sustained growth of a single anchor tenant, CoreWeave, though management is actively executing plans to diversify its counterparty exposure. Ultimately, the executive team’s successful defense against an opportunistic $9 billion takeover in late 2025, combined with their ability to independently secure $3.3 billion in development capital, demonstrates a sophisticated understanding of their inherent infrastructural value. If management can maintain disciplined deployment and execute on behind-the-meter generation strategies, the company is exceptionally well-positioned to capture the structural premium afforded to mission-critical energy assets in the accelerated computing era.