Wharton finance professor Jessica Wachter and a coauthor have built a stark estimate around confirmed hyperscaler capital outlays, projecting nearly $1.1 trillion in data center spending through 2027.

That projection spans five of the largest technology companies: Alphabet (NASDAQ: GOOGL), Microsoft (NASDAQ: MSFT), Amazon (NASDAQ: AMZN), Meta (NASDAQ: META), and Oracle (NYSE: ORCL).

Wachter, who previously served as chief economist at the Securities and Exchange Commission, found that hyperscaler productivity must grow 2.7 times over to break even by 2030.

That calculation accounts for the cost of capital, a required 15% return on investment, and the depreciation of the physical assets being built out at scale.

Alphabet posted a $5.9 billion free cash flow deficit last quarter, its first since going public in 2004, underscoring the financial pressure these companies are beginning to absorb.

Morgan Stanley calculates that hyperscalers will finance over half of their planned $2.9 trillion in data center spending through 2028, with that money increasingly drawn from external capital rather than internal cash reserves.

The financing structures behind these projects are growing more complex, as demonstrated by Meta transferring an 80% stake in its Hyperion data center to private-credit firm Blue Owl Capital in Louisiana.

Columbia Business School’s Stijn Van Nieuwerburgh warns that debt of this kind increasingly flows through pension funds and private credit vehicles, exposing everyday savers in ways they may not recognize.

Van Nieuwerburgh says few people realize how deeply that exposure has spread into their own retirement and insurance savings, raising systemic concerns well beyond Silicon Valley boardrooms.

Gary Gensler, former SEC chair and current MIT Sloan School professor, has described the situation as “a parlay bet by the capital markets and the economy,” and expects a retrenchment eventually, though its timing remains uncertain.

Crypto strategist Arthur Hayes has floated a related scenario, arguing that an AI credit bust could force the Federal Reserve to print money, potentially pushing bitcoin toward $1 million.

MIT Technology Review’s broader examination of the bubble makes clear that whether a correction arrives gradually or abruptly may determine how much of the trillion-dollar infrastructure bet becomes a permanent loss.