Jim Chanos: The Math Ain’t Mathing for the AI Data Center Build

The current AI infrastructure boom mirrors the speculative excesses of the late 1990s, characterized by massive capital expenditure on data centers that may lack long-term economic viability. Jim Chanos, founder of Chanos & Co., highlights that companies are committing hundreds of billions to projects based on near-term spot pricing rather than sustainable demand. This capital-intensive build-out is often masked by aggressive accounting practices, such as capitalizing costs for assets that remain unused or under-depreciated. While hyperscalers currently drive this spending, the return on incremental invested capital is declining rapidly. The market’s willingness to value these projects as if they are guaranteed successes ignores the risks of technological obsolescence and the potential for a severe correction when reality fails to meet the current, highly speculative expectations.