Toby Ord argues benchmark odds give theoretical basis for exponentially growing task horizons
tobyordoxford · x · 2026-11-14
Toby Ord offers a theoretical rationale for why AI task horizons might scale exponentially with effective compute (ECI). Benchmark score, expressed as a percent, equals the probability of succeeding at a random benchmark task; reformulating it as odds of success makes the scaling structure clearer. Measuring the exponent of this power law, he argues, is the key open question, since it governs how horizon length scales with RL compute.
Related event: Oxford's Toby Ord: RL Compute Needs 10,000x Boost to Lift Benchmark Scores(17 posts)→
More from AGI Musings
- Interconnects: Jensen's resignation is a major setback for US open-model ecosystem — Interconnects (Nathan Lambert) · 2026-11-14
- AGI endgame: either a balanced multipolar world or a centralized panopticon — CedricMakes · 2026-11-14
- "60% of people around you have some form of AI psychosis," says Alsop III — StewartalsopIII · 2026-11-14
- AI won't make everyone equal—it will split the smart from the rest worse than money — prasenx · 2026-11-14
- Toddlers learn new verbs only from responsive partners; a call to rerun the test with AI — AnnaCiaunica · 2026-11-14
- Developer muses: in Murphy's world, billions of interdependent events make rare failures daily — bingxu_ · 2026-11-14