Has γ Plateaued at ~0.66? Toby Ord Says We Can't Tell Yet
tobyordoxford · x · 2026-10-09
Toby Ord adds a caveat to his γ scaling analysis: it's currently impossible to tell whether γ has plateaued at roughly 0.66 or is still increasing.
In theory γ could even exceed 1 — meaning the AI would gain more performance from 10x the tokens than a human gains from 10x the time. The open question bears directly on how far AI time-horizon scaling can ultimately go.
Related event: Oxford's Toby Ord Estimates AI Capability Scaling Factor γ at 0.66(3 posts)→
More from Models
- Arena weekly: Nano Banana 2.1 top-6 on three image boards, Mistral Large 4 climbs to #43 in Agent Arena — arena · 2026-10-09
- Yann LeCun doubles down: LLMs have nothing to do with intelligence — mikeflache · 2026-10-09
- Liquid AI d1 3B decides in 8ms on a 3090, crushes OpenAI API (345ms) at Snake — victor_explore · 2026-10-09
- Unsloth adds fine-tuning to turn Qwen, Llama and Gemma into small local decision models — blaizedsouza · 2026-10-09
- Meta details Muse Glimmer: open-weight model for local agents, from pre-training to quantization — LotfiSanae · 2026-10-09
- NeurIPS paper: LLMs keep thinking even when explicitly told not to — james_y_zou · 2026-10-09