The $650 billion warning that never tested its own thesis

Bridgewater called the AI boom's "most dangerous phase," but a four-model panel found the load-bearing assumptions were asserted, not modeled. For allocators positioning against the call, that's the difference between a testable thesis and a narrative.