Domingos: AI Computing Is So Inefficient Nvidia Won't Keep Its Moat
pmddomingos · x · 2026-09-15
ML professor Pedro Domingos argues that current AI computing is so extraordinarily inefficient that Nvidia won't have much of a moat when a better solution appears — implying the GPU monopoly rests on wasted compute, and a more efficient architecture or paradigm could upend the landscape.
More from Infra
- Voodoo Dynamic Quant Goes MIT: Gradient Descent Picks Per-Tensor Quant Levels — 1ncehost · 2026-09-15
- Europe needs compute sovereignty at massive scale, not just AI regulation — VraserX · 2026-09-15
- EU compute report reportedly cites 45GW, making the case even weaker, critic says — eliebakouch · 2026-09-15
- Palantir and NVIDIA launch Sovereign AI OS Reference Architecture for enterprises — arieljalali · 2026-09-15
- Third-generation rancher risks losing 700-acre legacy to Bloom Data Center — SnoozeDoggyDog · 2026-09-15
- Ex-DeepMind engineer's deep dive dissects vLLM's inference architecture — solyarisoftware · 2026-09-15