GPUs already within 2x of brain efficiency, and still beat human workers on energy
MikePFrank · x · 2026-10-09
Chip expert MikePFrank argues data-center power debates ignore compute output: measured per synaptic event in FLOPS, batched GPU inference is already within 2x of the human brain's efficiency (citing Nvidia's sparse 4-bit Rubin figures). The brain uses 20W, but an average human's share of global primary energy is 10kW (heating, travel, goods) — a GPU never flies to Hawaii. Even a human-level AI at hundreds of watts would out-employ humans on energy efficiency. His company is exploring resonant energy recovery to surpass the brain.
Related event: GPU Inference Energy Efficiency Within 2x of Human Brain(2 posts)→
More from Infra
- Open-source sfFFT delivers 3.8-6.1x speedup for FFT convolutions on DGX Spark GB10 — IgorCarron · 2026-10-09
- Do AI agents still need Kubernetes? Berlin event says yes, with agent-on-K8s cases — Al_Grigor · 2026-10-09
- Cloud Backlogs Hit $1.69T, CoreWeave Posts $2.58B Quarter as Inference Becomes the Battleground — FinanceYF5 · 2026-10-09
- NVIDIA Is AI's Central Bank: A100 Paper Citations Still Beat H100+H200 Combined — FinanceYF5 · 2026-10-09
- Browser-Based Calculator Crunches the Real Power Cost of Self-Hosted LLMs vs Cloud APIs — paq85 · 2026-10-09
- PartyKit shuts down free hosted platform 2.5 years after Cloudflare acquisition — threepointone · 2026-10-09