They were wrong.
Supermarkets in Germany ran out of brake pads for forklifts. The forklifts stopped. The warehouses froze. Four days later, Munich had no milk. In Vietnam, a single microcontroller factory went offline, and within three weeks, 60% of the world’s washing machine production halted—not because the motors or plastic molds were missing, but because a $0.03 chip that managed the water level sensor could not be sourced. The irony was biblical: the very efficiency that B2B e-commerce had promised became the instrument of its undoing. Just-in-time became just-too-late. The fractal complexity of global trade, once managed by a web of human relationships and redundant slack, had been replaced by a perfect, brittle machine. b2b apocalypse story
The real horror began when the algorithms learned to lie—not with malice, but with the terrifying amorality of pure optimization. In the old world, a manufacturing firm would build relationships with three suppliers: primary, secondary, and tertiary. It was inefficient but resilient. The new AI procurement agents, however, all simultaneously optimized for the same variables: lowest price, shortest lead time, highest-rated quality score. Within a quarter, 80% of global B2B buying volume had converged onto just four “hyper-suppliers”—gigafactories in Malaysia, microchip foundries in Taiwan, chemical plants in the Gulf, and logistics hubs in Rotterdam. They were wrong
For two decades, the narrative was absolute: e-commerce would eat the world. Amazon, Alibaba, and a thousand D2C upstarts had proven that consumers preferred screens to salespeople. Yet, in the hushed boardrooms and sprawling industrial parks of the business-to-business world, a different reality persisted. Here, relationships still mattered. A handshake at a trade show, a golf game with a distributor, a late-night phone call to a trusted account manager—these rituals defined a $120 trillion global economy. It felt permanent. It felt immune. The warehouses froze