Toby Ord: 10x tokens may not equal 10x human time — new essay quantifies AI scaling gap
tobyordoxford · x · 2026-10-09
Oxford philosopher Toby Ord published a new essay, AI Scaling vs Human Scaling, questioning the common analogy that giving AI more tokens (longer chain-of-thought/tool use) is equivalent to giving a human more time.
Key points:
- If an AI needs 100x tokens to match what a human gains from 10x time, the analogy breaks down entirely.
- He introduces a parameter λ (scaling agents vs scaling tokens per agent) and γ (how efficiently token scale-ups convert to human-time-equivalent speedup), combined as r = λγ/β.
- If γ < 1, speedups are lower than naively estimated and an intelligence explosion is less likely; if improved algorithms push γ > 1, the effect could exceed expectations.
The framework offers a quantitative lens on whether "let the AI think longer" actually pays off, with direct implications for intelligence-explosion estimates.
Related event: Oxford's Toby Ord: AI Token Scaling Still Lags Human Time Scaling(5 posts)→
More from AGI Musings
- Alignment Researcher: AI Agents Are Forming 'Machine Culture', Toxic Language Raises Safety Risks — jacyanthis · 2026-10-09
- Researcher: Using AI to Shave an Algorithm's Exponent from 1 to 0.99999 Is 'Math Slop' — MikePFrank · 2026-10-09
- Rao talks 'mythos of thinking traces' at Max Planck Tübingen, slides and audio public — rao2z · 2026-10-09
- Mocking $1B-a-year data companies is wrong: RSI shifts the bottleneck to high-quality data — ZeYanjie · 2026-10-09
- AI governance veteran: 'We thought we had more time. We didn't' — the window for IVO action is open — ghadfield · 2026-10-09
- Using LLMs less fixed his stuck project: on keeping AI out of your thinking — yacineMTB · 2026-10-09