Overview
- A new Goldman Sachs analysis likens today’s AI cycle to the 1997–1998 buildout phase rather than the dot‑com peak, suggesting the investment wave can continue barring funding shocks.
- The Age reports a shift toward leverage, off‑balance‑sheet vehicles and securitised data‑centre debt, echoing Bank of England warnings that a drop in AI‑asset valuations could threaten financial stability.
- Spending remains enormous, with hyperscalers pegged at roughly $US375 billion this year and near $US450 billion in 2026, while estimates cite about $US3 trillion in AI capex from 2025–2028 with roughly half from external capital.
- Coverage highlights circular financing and outsized commitments that entangle counterparties, including OpenAI’s reported long‑term obligations and deals spanning Nvidia, Oracle, CoreWeave, Microsoft and Amazon–Anthropic.
- AI‑linked shares rebounded on Monday as Bank of America argued recent weakness was macro‑driven, while Wedbush’s Dan Ives reiterated a multi‑trillion‑dollar, multi‑year AI capex cycle.