AI and the Material World

The current AI-driven capital expenditure cycle exhibits characteristics of a fundamental bubble, where massive, price-agnostic spending by US hyperscalers creates a pervasive, singular thematic across global markets. AI-related capital investment now accounts for a significant portion of US economic growth, while non-AI sectors face negative year-on-year growth. In emerging markets, the weighting of Korea and Taiwan has surged to 45% of the MSCI index, driven by semiconductor and hardware demand. This concentration creates extreme correlation risks for asset allocators. Drawing parallels to the 2003–2013 mining super cycle, the current environment relies on unsustainable earnings growth rather than sound valuation metrics. As the capital cycle nears a potential turning point, investors should pivot toward neglected, undervalued sectors—such as Chinese real estate and the Philippines—where management teams are actively rationalizing capital allocation in response to market derating.