Chart shows hyperscalers went on a CAPEX spree after DeepSeek — and Kimi broke the logic
StewartalsopIII · x · 2026-09-14
Citing a chart and analysis from Izabella Kaminska, the thread argues hyperscalers accelerated CAPEX right after the DeepSeek moment: Chinese open-source AI proved near-frontier performance at commodity prices, convincing US labs that scale was their last moat and triggering an unsustainable spending war.
- Kimi 3 matching near-SOTA at a fraction of the compute broke the linear CAPEX-race logic, making the burn-war pointless
- The quoted take: labs need a credible off-ramp from the hyper-scaling narrative, and safety concerns are the obvious justification for a pivot toward "pacing the frontier" or even a government-funded, closed Manhattan-project-style setup
- Prediction: the scale narrative has run its course, with significant liquidations ahead
More from Companies & People
- Dario Amodei says government and public should have a stake in AI, mocked as doomer — whurley · 2026-09-14
- King Charles III to convene Nvidia, Google, DeepMind, OpenAI, Anthropic leaders in Scotland — Polymarket · 2026-09-14
- OpenAI quotes Terence Tao on AI freeing researchers, netizens say it aged like milk — chaumian · 2026-09-14
- Hugo Bowne launches Master Agentic Data Science workshop with PyMC's Thomas Wiecki — hugobowne · 2026-09-14
- Anthropic CEO Dario Amodei: the industry 'lied for too long' about AI risks — Polymarket · 2026-09-14
- CS enrollment trend sharply reverses, per FT — CSProfKGD · 2026-09-14