OpenAI and Anthropic take a third of incremental world compute, Dylan Patel says
bookwormengr · x · 2026-08-27
Dwarkesh relays Dylan Patel's view: OpenAI and Anthropic currently take about a third of the incremental world compute supply, likely rising to half next year.
At the current rate of physical compute scaling and algorithmic progress, we are a single-digit number of years away from individual AI companies with a greater effective workforce than there are humans on Earth.
Multiple forces push toward centralization: economies of scale in training (learnings amortized across billions of sessions), and being slightly ahead lets you economize on ever-more-expensive compute — with more factors unaccounted for.
More from AGI Musings
- False Collatz Proof via a Lean Bug: Talk Probes Proof Assistants as LLM-Found Bug Targets — marc_lelarge · 2026-10-16
- Pennsylvania residents reject $10,000-per-household offer for massive Amazon data center — Polymarket · 2026-10-16
- A 1968 newspaper predicted today's AI art debate almost word for word — mtizard · 2026-10-16
- Curing AI-era FOMO: define what 'fulfilling your potential' means to you — dejavucoder · 2026-10-16
- "An LLM doesn't understand anything" gets the perfect parody reversal: neither do you — VoidStateKate · 2026-10-16
- Nina Schick: Europe's Policy Circle Still Doesn't Believe FSD, Orbital Compute Are Real — NinaDSchick · 2026-10-16