factual

GPU memory-to-compute ratio limits where large models can run

Because leading GPUs have very high floating-point operations per second but limited high-bandwidth memory (a ratio up to ~1000:1, e.g. hundreds of teraflops with only 80-160GB), a model whose weights take 80TB cannot fit on one GPU—making the earliest-phase AI relatively identifiable and vulnerable, though distillation can later shrink specialized capabilities into smaller models.

factualpending

Speaker

Carl Shulman

Evidence Quote

“if you're trying to fit a model whose weights take 80TBs then with those chips you'd have to have a large number of the chips”

Source

Carl Shulman (Pt 2) — AI Takeover, bio & cyber attacks, detecting deception, & humanity's far future— Dwarkesh Patel Podcast
Created: 6/13/2026, 3:35:43 AM

Your notes

Loading notes...