8-K: WhiteFiber Unveils AI Infrastructure Growth Strategy
Investor Presentation
WhiteFiber, Inc. released an investor presentation detailing its strategy for scaling AI data center and cloud solutions, highlighting significant market opportunities and a robust project pipeline.
Summary
- WhiteFiber, Inc. is an AI infrastructure company providing data center and cloud solutions for AI workloads.
- The company targets a global data center capacity demand of over 298 GW by 2030E and a global Cloud AI market of approximately $454 billion by 2031E.
- Its strategy focuses on retrofitting existing facilities, which is faster and up to 40% cheaper than greenfield solutions, with buildout costs estimated at $7M-$9M per gross MW.
- The current portfolio includes MTL-1 (4.0 MW), MTL-3 (7.0 MW), and NC-1 (99.0 MW), with a total expected capacity of 115 gross MW within four years.
- Key projects include MTL-3, a 7 MW facility fully operational in Q4 2025 and contracted with Cerebras Systems, and NC-1, a 99 MW facility with 40 MW contracted by Nscale under a 10-year, $865 million TCV agreement, expected to be operational by mid-2026.
- The company reported combined revenue of $47.6 million for FY2024 and $70.5 million for LTM Q3 2025.
- Net income for FY2024 was $1.37 million, while 3Q 2025 saw a net loss of $15.75 million. Adjusted EBITDA was $21.93 million for FY2024 and $2.28 million for 3Q 2025.
Sentiment
Score: 6
Explanation: The presentation highlights strong growth opportunities, a differentiated strategy, and significant project wins, indicating a positive outlook. However, recent financial results (3Q 2025 net income, EBITDA, and Adjusted EBITDA) show a notable decline, and one project (MTL-2) is on hold, tempering the overall positive sentiment.
Positives
- Significant and growing market opportunity in global data center capacity demand (>298 GW by 2030E) and Cloud AI market (>$454B by 2031E).
- Differentiated and efficient retrofit strategy offers faster time-to-market and up to 40% lower buildout costs ($7M-$9M per gross MW) compared to greenfield solutions.
- High unlevered returns for data centers and cloud services.
- Robust HPC data center portfolio and pipeline, with four facilities totaling 115 gross MW capacity expected within four years.
- Strong anchor contracts, such as the 10-year, 40 MW colocation agreement with Nscale for NC-1, valued at $865 million TCV.
- Experienced senior operating team with an average of 15 years in data center and cloud services industries.
- Clean capital structure supported by low-cost bank debt and ordinary shares.
- MTL-3 retrofit for Cerebras Systems was delivered in 6 months, significantly faster than typical greenfield builds (18+ months).
- NC-1 site benefits from a low energy cost of approximately $0.05 per kWh from Duke Energy.
Negatives
- Development of the MTL-2 5.0 MW facility has been put on hold.
- Net income for 3Q 2025 was a loss of $15.75 million, a significant decrease from a net income of $1.37 million in FY2024 and a loss of $0.37 million in 3Q 2024.
- EBITDA for 3Q 2025 was negative $9.32 million, down from positive $18.76 million in FY2024 and positive $4.34 million in 3Q 2024.
- Adjusted EBITDA for 3Q 2025 was $2.28 million, a substantial decrease from $21.93 million in FY2024 and $5.64 million in 3Q 2024.
Risks
- Forward-looking statements involve substantial risks and uncertainties, including the ability to scale the business, acquire new data centers, implement business models, commercialize and grow by acquiring customers, compete effectively, and manage expenses and capital requirements.
- The company operates in a very competitive and rapidly changing environment, with new risks and uncertainties emerging that could impact forward-looking statements.
- Actual results may differ materially from projected financial information due to inherent uncertainties and a wide variety of significant business, economic, competitive, and other risks.
- The financial information presented is unaudited and may be adjusted or presented differently in future SEC filings.
Future Outlook
WhiteFiber anticipates significant growth in data center capacity, targeting 115 gross MW within four years, driven by its retrofit strategy and expansion of key facilities like NC-1. The company expects to roll out new cloud technology solutions and cross-data center workload capabilities in Q1 2026, enhancing its AI infrastructure offerings. Initial deployment for NC-1 is expected by mid-2026, with potential for further expansion towards double its initial capacity by the end of 2027.
Management Comments
- "We believe the information contained herein is accurate in all material respects."
- "We have based the forward-looking statements contained in this presentation primarily on our current expectations, estimates, forecasts, and projections about future events and trends that we believe may affect our business, financial condition, results of operations, and prospects."
- "Management believes NC-1 may receive up to 200.0 gross MW of power supply over time subject to infrastructure upgrades and other conditions."
Industry Context
WhiteFiber positions itself as a key player in the rapidly expanding AI infrastructure market, addressing the growing demand for high-performance computing (HPC) data centers and cloud solutions. Its retrofit strategy offers a competitive advantage by providing faster deployment and lower costs compared to traditional greenfield developments, aligning with the urgent need for scalable and energy-dense solutions for AI workloads. The company's focus on Tier-3 data centers and specialized cooling systems caters to the stringent requirements of AI customers, differentiating it from general-purpose data center providers.
Comparison to Industry Standards
- Buildout cost of $7M-$9M per gross MW is up to 40% lower than greenfield solutions, based on company estimates for industry average.
- MTL-3 site delivered in 6 months, significantly faster than the 18+ months typical for greenfield builds.
- MTL-3 achieved a Power Usage Effectiveness (PUE) of 1.3, which is better than the industry standard of 1.45.
- Data centers are designed for 99.982% uptime, equivalent to no more than 1.6 hours of downtime annually, meeting Tier-3 concurrently maintainable design standards.
Stakeholder Impact
- Shareholders: Potential for long-term value creation through growth in AI infrastructure, but also exposure to risks associated with rapid expansion and competitive markets. Recent financial performance shows a decline, which could impact short-term sentiment.
- Customers (Cerebras Systems, Nscale, 20+ enterprises): Benefit from high-performance, energy-dense, and reliable data center and cloud solutions tailored for AI/HPC workloads, with faster deployment times for retrofit projects.
- Employees: Opportunities in a growing company focused on cutting-edge AI infrastructure.
- Creditors/Lenders: Potential for new project-level financing for NC-1, indicating ongoing capital needs and opportunities for lenders.
- Suppliers: Opportunities for partnerships in specialized procurement ecosystems for modular data center builds and direct-to-chip liquid cooling.
Next Steps
- NC-1 initial deployment expected to be operational by mid-2026.
- Project-level financing for NC-1 expected in 1Q 2026.
- WhiteFiber Cloud Technology and cross-data center workload technology expected to be rolled out/commercially available in Q1 2026.
- Potential expansion of NC-1 toward ~2x initial deployment by end of 2027.
- Achieve 44.0 MW of service for NC-1 by April 2026 and 99.0 MW within four years, with potential for up to 200.0 gross MW over time.
Key Dates
| Date | Description |
|---|---|
| 2021-01-01 | MTL-1 data center became operational. |
| 2024-10-01 | MTL-1 4.0 MW data center platform acquired and Enovum acquired. |
| 2024-12-01 | MTL-2 5.0 MW colocation agreement established (development put on hold). |
| 2025-01-01 | Cloud Services launched. |
| 2025-04-01 | NC-1 initial power availability of 24 MW. |
| 2025-08-01 | Company IPO. |
| 2025-08-08 | Prospectus filed pursuant to Rule 424(b)(4) of the Securities Act of 1933. |
| 2025-09-30 | End of Q3 2025 financial reporting period. |
| 2025-10-01 | MTL-3 fully operational. |
| 2025-11-13 | Quarterly Report on Form 10-Q for the quarter ended September 30, 2025, filed with the SEC. |
| 2025-12-01 | Colocation agreement for NC-1 with $865M TCV signed. |
| 2026-01-15 | FactSet market cap reference date. |
| 2026-01-26 | Date of earliest event reported and date of Form 8-K filing. |
| 2026-03-31 | Expected rollout of WhiteFiber Cloud Technology and cross-data center workload technology commercially available. |
| 2026-04-01 | Expected 20 MW IT Load for NC-1 Phase 1 & 2 and Duke Energy to achieve 44.0 MW service for NC-1. |
| 2026-05-01 | Expected 20 MW IT Load for NC-1 Phase 1 & 2. |
| 2026-06-30 | NC-1 initial deployment expected to be operational by mid-2026. |
| 2026-12-31 | Expected 76 MW online by year-end 2026. |
| 2027-12-31 | Potential expansion of NC-1 toward ~2x initial deployment by end of 2027. |
| 2030-12-31 | Estimated global data center capacity demand of ~298 GW. |
| 2031-12-31 | Estimated global Cloud AI market of ~$454B. |
Recommendation
holdThe company operates in a high-growth sector with a compelling strategy and significant project wins, suggesting long-term potential. However, the recent decline in net income, EBITDA, and Adjusted EBITDA for 3Q 2025, coupled with the delay of the MTL-2 project, introduces short-term uncertainty. While the long-term outlook is positive due to market demand and strategic execution, the recent financial performance warrants a cautious "hold" recommendation until there is clearer evidence of improved profitability and successful execution of the NC-1 project and cloud technology rollout.
Keywords
AI infrastructure, data center, cloud solutions, HPC, high-performance computing, retrofit, colocation, NVIDIA GPUs, AI workloads, SEC filing, investor presentation, WhiteFiber
Disclaimer:The information provided here is for general informational purposes only and does not constitute financial advice, recommendation, or endorsement of any kind. It may contain errors or omissions. You should not rely on this information to make financial decisions. Always seek the advice of a qualified financial professional before making any investment or financial decisions. Use of this information is at your own risk.