Brookfield Estimates $7 Trillion Needed for Global AI Buildout

Brookfield Asset Management estimates that global artificial intelligence growth will require approximately $7 trillion in capital investment over the next decade. This projected spend includes $4 trillion for computing infrastructure, $2 trillion for AI factories, and $1 trillion split between power transmission and strategic adjacencies. CFO Hadley Peer Marshall expects these investments to generate $10 trillion in annual productivity.

Despite the optimism, analysts warn of a growing economic bubble. Tech companies project infrastructure spending to reach $400 billion this year, with U.S. expenditures likely exceeding $500 billion in 2026 and 2027. However, Bain & Company reports that $2 trillion in new revenue is required to justify current scaling trends, noting a potential $800 billion revenue shortfall by 2030 as computing costs outpace income from services.

Corporate behavior reflects a preference for aggressive spending over caution. Mark Zuckerberg of Meta indicated he would rather overspend billions than risk a late entry into the expansion. Similarly, Oracle saw a 36% stock surge following cloud contracts with OpenAI, despite facing negative free cash flow and a need for $100 billion in additional capital.

Critics like Paul Kedrosky argue this capital concentration is starving other sectors, such as domestic manufacturing, of investment. Physical expansion has also sparked local opposition in Northern Virginia over noise and land use, while reports from the Harvard Business Review and MIT highlight productivity losses from AI-generated "workslop" and the risks of overtrusting inaccurate medical advice.