Toby Ord: share of impossible tasks could skew METR's AI time horizon
tobyordoxford · x · 2026-11-09
Toby Ord replies to Basil Halperin's summary of METR time-horizon analysis, raising a key question: if a substantial share of tasks (say 25%) are ones agents simply cannot do well, those tasks shorten the 80%-accuracy time horizon far more than tasks the agent can partially do. The thread continues the debate over OpenAI's lower-than-headline AI R&D time horizons.
Related event: Toby Ord Questions METR Time-Horizon Findings(2 posts)→
More from AGI Musings
- METR says it has 'convincing evidence' AI speeds up research — but the details are redacted — peterwildeford · 2026-11-09
- OpenAI co-founder Greg Brockman: frontier models with test-time compute are becoming the engine of scientific discovery — gdb · 2026-11-09
- Long-term memory vs. search is like L1 cache vs. RAM, argues zetalyrae — zetalyrae · 2026-11-09
- AI is eroding traditional moats: years of advantage now replicated in days — DeryaTR_ · 2026-11-09
- Epoch AI: 2027 chip supply could run 30–170M frontier AI agents, or 1.9B cheaper ones — Dr_Singularity · 2026-11-09
- White paper: moving beyond the AI compute arms race as big tech faces historic spending losses — ReportProfessional57 · 2026-11-09