Trillions in Investment, Thousands of New Players—AI Data Centers Become a "New Gold Rush" Shaping Global Economic Fate

A gold rush in artificial intelligence (AI)-driven infrastructure is reshaping the global data center market, with thousands of new entrants from diverse backgrounds joining tech giants in a wave attracting trillions in investment—potentially with profound implications for the global economy.

According to compiled data, the scale of this construction boom is unprecedented. In 2025 alone, U.S. data center credit transactions reached at least $178.5 billion. Tech giants like Oracle, Meta, and Alphabet have helped push global bond issuance this year to over $6.57 trillion. JPMorgan estimates issuers will tap nearly all major debt markets to fund this global build-out.

The core shift, however, lies in market participants. Data shows most planned data center capacity will be built by "new players" and industry entrants beyond big tech—a structural change dispersing risk but raising concerns: if AI’s commercial promise falters, its collapse could ripple through these newcomers, impacting equity and debt markets worldwide.

Big tech is also adjusting strategy, increasingly leasing rather than building data centers, offloading massive debt and construction risks to developers. Microsoft CEO Satya Nadella cited anticipated "overbuilding" of computing power as a reason he’s "happy to be a tenant." This shift concentrates risk on less-experienced developers.

**New Players: From Energy Firms to Bitcoin Miners** The AI infrastructure race features remarkably diverse entrants, signaling a market transformation.

In Italy’s Puglia region, Lorenzo Avello—previously in renewables—plans to build a "Mediterranean AI hub" via his new firm Adriatic DC, a 1.5-gigawatt campus aiming to be among Europe’s largest computing centers. Avello has no prior data center experience.

Similar ambitions emerge in North America. Kevin O’Leary of *Shark Tank* plans Alberta’s "world’s largest AI data center park," leveraging local gas and geothermal resources. "I’ll be one of five who succeed at hyperscale—you’ll see hundreds fail," he asserts.

Crypto firms are also diving in. Bitcoin miner Bitdeer Technologies Group plans to invest "billions" expanding AI cloud services, with a 570-megawatt Ohio campus slated for 2027. Ex-crypto miner CoreWeave Inc. now leads in leasing NVIDIA chips to firms like Microsoft.

**Trillion-Dollar Inflows and Bubble Fears** Massive capital inflows coincide with growing AI bubble concerns.

OpenAI, Oracle, and SoftBank’s $500 billion Stargate project and Avello’s €50 billion+ Italian venture highlight the scale. But skeptics like Michael Burry warn of AI hype. Circular deals—where NVIDIA and Microsoft fund startups buying their products—raise doubts about real demand.

"There simply aren’t enough customers," says ex-Microsoft manager Charles Fitzgerald, predicting many projects will never materialize. Oaktree’s Howard Marks warns overbuilding risks loom, noting tech firms’ flexible lease terms could leave data center owners stranded.

**Big Tech’s Pivot: From Builders to Tenants** Facing massive capex and uncertain demand, tech giants are deftly transferring risk.

Meta secured ~$60 billion for data centers in October, but half ($30B) stays off its balance sheet via a Blue Owl Capital SPV. Microsoft committed over $60B to lease from "neocloud" firms building AI-chip-equipped centers—avoiding direct operational risks while pressuring developers.

**Execution Risks: Delays and Terminations** For new players, turning plans into reality proves fraught.

CoreWeave cut revenue forecasts due to delays with a lagging developer. Bitdeer’s Ohio Bitcoin facility caught fire last month. Fermi Inc.’s stock plunged 46% after a $150M tenant cancellation.

These incidents highlight newcomers’ struggles with project management, safety, and contract stability. Bitdeer’s Haris Basit calls terms "ironclad," hoping defaults trigger exit fees—but systemic risks remain.

As Basit admits, despite diversification across Bitcoin/AI, owned/hosted, and geographies, "you can imagine a scenario where everything’s affected. You can’t hedge against that." This uncertainty lies at the heart of the AI data center gold rush.