Toby Ord finds most measured reasoning gains came from longer inference
Toby Ord estimated that 63% to 92% of measured reasoning gains on three benchmarks came from using more inference tokens rather than improved performance at a fixed token budget. He argued that this raises deployment cost and moderates conclusions about gains in underlying intelligence.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why it moved the index
The dated first-person analysis adds a distinct quantitative mechanism for separating underlying capability gains from spending more compute at inference. It is material to projections of rapid autonomy, but confidence is moderated because the calculation uses three selected benchmarks and limited public reporting.
Assessment history
-
R1
Away 30 · confidence 72
New dated tracked-person analysis found in the October 2025 backfill.
13 Aug 2026
Share this page
-
DoomBench assesses “Toby Ord finds most measured reasoning gains came from longer inference” as evidence moving away from doom, with magnitude 30 and confidence 72 out of 100 in the human resilience category.
-
The DoomBench assessment of “Toby Ord finds most measured reasoning gains came from longer inference” is based on reporting from Toby Ord and records the editorial rationale, source quality, attribution, and revision history.
-
DoomBench summarizes “Toby Ord finds most measured reasoning gains came from longer inference” as follows: Toby Ord estimated that 63% to 92% of measured reasoning gains on three benchmarks came from using more inference tokens rather...
https://www.doombench.com/news/toby-ord-finds-most-measured-reasoning-gains-came-from-longer-inference-2025-10-03