Toby Ord: task success odds grow as a power law in compute, logistic curves need checking
tobyordoxford · x · 2026-11-14
In a thread with @ramez, @benjtodd and @AllanDafoe, Oxford's Toby Ord argues that odds of success on a random benchmark task often grow exponentially in log(compute) — meaning odds follow a power law in compute.
He notes the x-axis can be time or log(compute) with the same curve shapes, and candidly flags his uncertainty: we usually only observe the roughly linear middle portion, so assuming logistic curves is a working hypothesis that should be empirically checked.
Related event: Oxford's Toby Ord: RL Compute Needs 10,000x Boost to Lift Benchmark Scores(17 posts)→
More from Research
- NeurIPS 2025 best paper: diffusion models start memorizing at ~300 steps per image — Mastbubbles · 2026-11-14
- TestPrune: pruning regression tests to unblock AI-driven autonomous bug fixing — burkov · 2026-11-14
- Toddlers learn new verbs only from responsive partners; a call to rerun the test with AI — AnnaCiaunica · 2026-11-14
- Context beats model choice: Databricks benchmark shows 91% vs 58% for data agents — CautiousUse8597 · 2026-11-14
- MIT professor to showcase autonomous AI agent swarms for metamaterials discovery — ProfBuehlerMIT · 2026-11-14
- Walden talk: base models improving fast, mid-training data commoditizing — ZeYanjie · 2026-11-14