The current $660 billion capital expenditure by hyperscalers represents the largest infrastructure build-out in human history, driven by a fundamental shift from pre-compiled software to real-time generative intelligence. Brad Gerstner, founder of Altimeter Capital, argues that this spending is sustainable because AI has reached an inflection point where "tokens" are now profitable and software has evolved from a static tool into an agentic system capable of reasoning and research. Using Meta as a primary example, the discussion highlights how AI-driven recommender systems are already significantly boosting corporate earnings. Unlike the fiber optic glut of the dot-com era, there are no "dark GPUs"; demand remains so high that even six-year-old hardware is appreciating in value. This transition mirrors Amazon’s early investment in AWS, suggesting that those fixated on short-term cash flow are overlooking a generational shift where machines augmenting human thought will drive all future economic progress.
Sign in to continue reading, translating and more.
Continue
