Analysis of the Impact of AI Enterprise Data Center Asset Equity Transactions on Industry Investment Valuation and Competitive Landscape

I will now provide you with a comprehensive and in-depth analysis report based on the collected data.

Sharon AI Holdings Inc. (an Australian “new cloud” concept company) announced on December 23, 2025 that it had signed a binding term sheet to sell its 50% stake in Texas Critical Data Centers LLC (TCDC) to joint venture partner New Era Energy & Digital Inc. for USD 70 million [1][2].

This transaction adopts a complex mixed consideration structure, reflecting the new trends in current data center asset transactions [1][2]:

TCDC was originally a joint venture equally owned (50% each) by Sharon AI and New Era. Through this transaction, Sharon AI achieves a complete exit while retaining a small equity exposure to New Era, leaving upside potential for future valuation increases [1].

The valuation paradigm of data centers in the AI era is undergoing profound changes. Traditional valuation methods are centered on “bricks and mortar”, referencing industrial real estate valuation models, focusing on rent per square foot, stable net operating income (NOI), and capitalization rates [3]. However, the AI revolution has subverted this paradigm:

• Investors have shifted to using “megawatts (MW)” as the core underwriting unit
• Revenue models have been adjusted around power to a pricing model of “USD per kilowatt per month”
• Power security capabilities have become the primary value driver

This shift reflects the extreme dependence of AI workloads on power resources, while also revealing that the value anchor of data center assets has shifted from physical space to energy supply capacity.

Data center asset transactions show strong growth momentum [3][4]:

• North America: 50% of total transactions (dominant position)
• Europe: 26%
• Asia: 12%
• Other Regions: 12%

This distribution highly aligns with the global AI computing power demand pattern, reflecting North America’s leading position in AI infrastructure.

Data center REITs performed strongly in 2025 [4][5]:

• Funds from Operations (FFO) in Q1 increased by over 21% year-over-year
• Net Operating Income (NOI) increased by 7.2%
• Over 50% of quarterly new bookings were driven by AI-related demand

Valuation indicators of representative enterprises:

• Digital Realty (DLR)
  : Q3 2025 core FFO reached a record USD 1.89 per share, up 13% year-over-year, with a backlog of USD 852 million [5]
• Equinix (EQIX)
  : Maintains a valuation premium through its interconnection ecosystem
• Iron Mountain (IRM)
  : Annual revenue from its data center business is close to USD 800 million, and is expected to exceed USD 1 billion in 2026 [5]

The release of the DeepSeek R1 model in early 2025 had a significant impact on the valuation logic of the data center industry [6]:

• DeepSeek used NVIDIA’s older H800 chips, with a training cost of only USD 5.576 million, which is just 1/10 of the cost of Meta’s Llama model [6]
• Challenged the industry consensus that “hyperscale computing providers must invest billions of dollars in advanced AI chips”
• NVIDIA’s market capitalization evaporated USD 59 billion, falling 16% in one week [6]

1. Adjustment to Capital Expenditure Expectations
   : In a pessimistic scenario, the capital expenditure intensity of hyperscale computing providers may fall back to 2015-2019 levels, reducing capital expenditure by approximately 24% [6]
2. Revision to Power Demand Expectations
   : Previously, the U.S. Department of Energy assumed that AI workloads would drive data center power demand to grow at a compound annual growth rate of 26%. After the DeepSeek incident, in a pessimistic scenario, this growth rate may fall back to 13% [6]
3. Industry Revenue Risks
   : Analysts predict that by 2028, chip manufacturers’ data center revenue may decrease by approximately 7 percentage points due to changes in capital expenditure plans, involving revenue risks of approximately USD 37.5 billion [6]

The AI data center industry is forming a multi-tiered competitive landscape:

The funding sources for data center assets show a trend of diversification [4]:

• Private Equity and Alternative Investors
  : Led the largest single transaction in 2025, with SoftBank leading a USD 40 billion investment in OpenAI [4]
• Sovereign Wealth Funds
  : Increased allocation to data center assets
• REITs
  : Unlock value through asset securitization
• Hyperscale Enterprises
  : Parallel strategy of self-construction and leasing

Principal Financial raised USD 3.6 billion for its Data Center Growth & Income Fund in February 2025, planning to invest in U.S. hyperscale data center facilities [3], reflecting institutional investors’ continued enthusiasm for this sector.

The Sharon AI transaction case reflects the consolidation trends in the AI data center industry:

1. Specialized Division of Labor
   : Sharon AI focuses on core operations and sells stakes in non-core assets to its joint venture partner [1]
2. Capital Efficiency Optimization
   : Unlock funds through asset allocation adjustments to support core business development
3. Deepening of Strategic Alliances
   : New Era obtains full control of TCDC through the acquisition, enabling resource integration to achieve scale effects

• The U.S. accounts for over 40% of global data center capacity, and its share is expected to continue to increase
• Vacancy rates in primary markets (Northern Virginia, Dallas, Phoenix, Chicago, Silicon Valley) are extremely low
• Pre-leasing has become the norm, and supply constraints stem mainly from available power capacity rather than land or capital

• The TCDC involved in the Sharon AI transaction is located in Texas, which has become a hot spot for data center investment
• Benefiting from relatively low energy costs, a relaxed regulatory environment, and renewable energy supply capabilities

Investors need to adopt an updated valuation framework to evaluate AI data center assets:

1. Power Security Capabilities
   : Power capacity locked in through long-term power purchase agreements (PPAs)
2. Technology Life Cycle
   : Key facilities need to be replaced and upgraded every 8-15 years
3. Tenant Quality
   : Credit qualifications of hyperscale tenants
4. Modular Capability
   : Rapid deployment and expansion capabilities

Based on the above analysis, different types of investors can adopt differentiated strategies:

1. AI Commercialization Progress Falls Short of Expectations
   : May lead to capital expenditure contraction
2. Fluctuations in Power Costs
   : Impact operating profit margins
3. Technology Iteration Risk
   : Existing facilities may depreciate rapidly
4. Changes in Regulatory Policies
   : Power supply and environmental compliance requirements

Sharon AI’s sale of a 50% stake in its Texas data center is a typical case of optimized data center asset allocation in the AI era. This transaction reflects the following industry trends:

1. Valuation Paradigm Shift
   : From “bricks and mortar” to “power-driven”, megawatts (MW) become the core valuation unit
2. Transaction Structure Innovation
   : Mixed consideration (cash + equity + convertible notes) becomes the mainstream
3. Accelerated Industry Consolidation
   : Specialized division of labor and deepening of strategic alliances
4. Reshaped Competitive Landscape
   : Hyperscale enterprises and specialized REITs form differentiated competition
5. Revision of Valuation Assumptions
   : The DeepSeek incident reminds investors to re-evaluate growth expectations for AI chip demand and power demand

For investors, understanding these structural changes and adjusting their investment frameworks accordingly will be the key to seizing long-term opportunities in the AI data center industry.

[1] Investing.com - “Sharon AI to sell 50% stake in Texas data center JV for $70m” (https://ca.investing.com/news/company-news/sharon-ai-to-sell-50-stake-in-texas-data-center-jv-for-70m-93CH-4377206)

[2] Yahoo Finance - “Sharon AI to Divest 50% Interest in Texas Critical Data Centers LLC for US$70m” (https://finance.yahoo.com/news/sharon-ai-divest-50-interest-130000961.html)

[3] SitusAMC - “Data Centers in an AI-Driven Era: Key Trends Reshaping Valuation” (https://www.situsamc.com/resources-insights/articles/data-centers-ai-driven-era-key-trends-reshaping-valuation)

[4] Crunchbase - “6 Charts That Show The Big AI Funding Trends Of 2025” (https://news.crunchbase.com/ai/big-funding-trends-charts-eoy-2025/)

[5] Intellectia AI - “How to Profit from the Best Data Center REITs in 2026” (https://intellectia.ai/blog/best-data-center-reits)

[6] J.P. Morgan Private Bank - “Will the DeepSeek Incident Change the Landscape of AI Transactions?” (https://privatebank.jpmorgan.com/apac/cn/insights/markets-and-investing/tmt/is-the-deepseek-drama-a-gamechanger-for-the-ai-trade)

[7] U.S. News - “7 Best Data Center Stocks, ETFs and REITs to Buy” (https://money.usnews.com/investing/articles/best-data-center-stocks)

[8] Data Center Knowledge - “Data Centers Are Commercial Real Estate’s Jewel for 2026” (https://www.datacenterknowledge.com/data-center-site-selection/with-all-eyes-on-ai-data-centers-are-commercial-real-estate-s-jewel-for-2026)