Trillions of dollars in investment, thousands of new players—AI data centers have become a "new gold rush" that concerns the fate of the global economy.

A gold rush in infrastructure driven by artificial intelligence (AI) is reshaping the global data center market. The main players are no longer just tech giants, but thousands of new entrants from diverse backgrounds. This wave is attracting trillions of dollars in investment and could profoundly impact the global economy. According to Bloomberg-compiled data, the scale of this building frenzy is unprecedented. In 2025 alone, U.S. data center credit transactions have already reached at least $178.5 billion. Tech giants like Oracle, Meta, and Alphabet have helped push this year’s global bond issuance past $6.57 trillion. JPMorgan Chase predicts that issuers will tap virtually all of the world’s major debt markets to fund this worldwide construction. However, the most significant change is the composition of market participants. Data shows that most of the planned future data center capacity will be built by “new players” and other industry participants outside of large tech companies. While this structural shift disperses risk, it also raises new concerns: if the commercial prospects for AI are invalidated, its collapse could ripple through all corners of the global economy, touching every facet of the equity and debt markets. Large tech companies are also adjusting their strategies, increasingly opting to rent rather than build their own data centers, thereby shifting huge debt loads and construction risks onto developers. Microsoft CEO Satya Nadella, who has warned of “overbuilding” computing capacity, candidly said one reason he is “happy to be a tenant” is precisely this. This shift in strategy is concentrating risk on emerging developers who are relatively inexperienced. **New Entrants: From Energy Firms to Bitcoin Miners** Participants in the AI infrastructure race come from diverse backgrounds, highlighting a fundamental change in the market. In Puglia, southern Italy, Lorenzo Avello, previously involved in renewable energy projects, plans to create a “Mediterranean AI hub” through his new company, Adriatic DC. The project includes a massive campus with a power consumption as high as 1.5 gigawatts, aiming to become one of Europe’s largest computing centers. Avello himself has never built a data center before. Similar ambitions are unfolding in North America. Kevin O’Leary, host of the TV show “Shark Tank,” plans to develop the “world’s largest AI data center industrial park” in Alberta, Canada, leveraging abundant local natural gas and geothermal resources. He’s confident: “I’ll be one of five to succeed in building superscale facilities, and you’ll see hundreds fail.” Meanwhile, cryptocurrency companies are making big moves. Bitcoin mining firm Bitdeer Technologies Group plans to invest “hundreds of millions, even billions of dollars” to expand its AI cloud business. Chief Strategy Officer Haris Basit says the company plans to launch a 570-megawatt AI data center campus in Clarington, Ohio in the second half of 2027. Bitdeer is not alone—former crypto mining company CoreWeave Inc. has become a leader in leasing Nvidia chips to firms like Microsoft. **Trillions in Capital Flood In and Bubble Concerns Rise** The vast inflow of capital has accompanied growing worries about an AI bubble. OpenAI, Oracle, and SoftBank Group announced this year the Stargate infrastructure project, promising up to $500 billion in investments across the rural United States. Lorenzo Avello’s Italian project is expected to cost more than €5 billion ($5.9 billion). However, Wall Street veteran and contrarian investor Michael Burry has warned of an AI bubble. The recursive nature of recent AI sector deals—where Nvidia, Microsoft, and others invest in startups that buy their products—has also triggered doubts about the real market demand for AI products. “There simply aren’t that many customers,” said Charles Fitzgerald, a former Microsoft manager and longtime data center observer. He believes many projects under development will never materialize. Howard Marks, co-founder of Oaktree Capital Management LP, also warned that the risks of overbuilding are looming. He noted that savvy tech companies are adding flexibility clauses to their computing power leasing contracts—if they pull out, data center owners will be left in a bind. **Tech Giants Shift Strategies: From Self-Building to Renting** Facing massive capital expenditures and uncertain long-term demand, tech giants are cleverly shifting risks. Take Meta, for example. The company received around $60 billion in October for building data centers, but half of it ($30 billion) won’t appear on its balance sheet. This deal, arranged by Morgan Stanley, channels financing into a special purpose vehicle (SPV) tied to asset management firm Blue Owl Capital Inc. Microsoft has pledged more than $60 billion to rental agreements with “neoclouds,” companies that build data centers and equip them with advanced AI chips. This strategy enables tech giants to reap the benefits of expanding computing capacity while avoiding the heavy asset risk of direct construction and operation—leaving the pressure on developers. **Execution Risks: Project Delays and Contract Terminations** For new players bearing these risks, the path from blueprint to reality is full of challenges. Former crypto mining firm CoreWeave recently cut its annual revenue forecast due to “temporary delays involving a third-party data center developer behind schedule.” Bitdeer’s Bitcoin mining facility in Ohio caught fire last month. Data center power developer Fermi Inc. saw its share price plunge 46% in a single day after disclosing that an investment-grade tenant had terminated a $150 million contract. These incidents reveal the severe tests new entrants face in project management, construction safety, and contract stability. Despite Bitdeer’s Haris Basit claiming the company’s contract terms are “rock solid”—and even expressing hope for customer defaults so they can collect exit fees and re-lease at higher prices—systemic risks remain. As Basit admitted, while the company seeks diversification across Bitcoin and AI, self-mining and hosting, Europe and the U.S., “you can imagine a scenario where everything is affected. You may not be able to guard against that.” This is the core uncertainty investors must confront when evaluating the AI data center gold rush. Risk Warning and Disclaimer The market has risks; investment requires caution. This article does not constitute personal investment advice and has not considered the specific investment objectives, financial situations, or needs of individual users. Users should assess whether any opinions, viewpoints, or conclusions in this article fit their particular circumstances. Investments made on this basis are at your own risk.