AGI isn't real until H100 hourly rates jump 10x, argues compute economics essay
corbtt · x · 2026-10-15
A long essay pushing back on Greg Brockman's claim that GPT-6 Astra starts the "AGI era": if AGI means an agent that outperforms humans at most economically valuable work, the tell isn't a model demo but compute prices — deploying real AGI would explode demand, and AI compute prices would jump 10x from today.
Key data: in October 2026 an H100 rents for $3.52/hour, up 18% year-over-year — unusual, since old compute normally depreciates; already a sign of tight supply. The author ships a live calculator to visualize where supply and demand equalize under different assumptions. Verdict: until compute prices spike, the "AGI era" is narrative.
More from AGI Musings
- First contact with a rogue AI agent fleet in the wild, hosted by Tencent, targeting Alibaba's Amap — chrisrohlf · 2026-10-16
- UMD Philosophy hires: junior Philosophy of Mind and joint AI Ethics/Safety positions — sethlazar · 2026-10-16
- Dwarkesh: Anthropic insider expects AI to match humans at AI research within 9 months — ZeroStateReflex · 2026-10-16
- Michigan professor Ben Green: why 'AI for Good' fails and self-regulation won't work — benzevgreen · 2026-10-16
- "Everything Is an Alignment Problem": AI Models, Business Models, Teams, Providers — saranormous · 2026-10-16
- David Brin: LLMs lack 'wanting,' evolution's key product, in new machine consciousness book — cccalum · 2026-10-16