AI Investment Bubble Concerns Rise as Costs and Valuations Spur Investor Caution

Analysts and investors warn that the rapid surge in artificial‑intelligence spending may be forming a new financial bubble. Economist Eric Winograd argues that AI‑related capital investment is losing growth momentum, with the 2024 boom flattening after an 85 % jump and practical constraints such as scarce land, power and chip prices threatening profitability. Veteran investor Jeremy Grantham likens the AI hype to the dot‑com, 2008‑housing and 19th‑century railroad bubbles, contending that current valuations are unsustainable and urging a shift toward non‑US assets, bonds and gold. Taiwan’s central‑bank governor echoed similar worries, flagging an AI‑bubble risk. In parallel, banks report exploding token‑based fees from services like ChatGPT, forcing them to cap AI budgets and seek cost‑efficient deployment, while regulators such as the ECB demand AI risk‑management plans. Some market participants, including asset‑manager Jennison, argue that AI valuations remain below the dot‑com peak and that demand is genuine, but the debate underscores growing investor fatigue and a shift from hype to scrutiny of unit economics.

The convergence of slowing investment growth, rising operational costs and divergent analyst views suggests that the AI sector may soon face a reality check, impacting technology firms, financial markets and corporate budgeting worldwide.