--- title: "Time horizon: what it means · The AI Compute Race · DRYBURGH" source: https://dryburgh.com/vol1/glossary/time-horizon description: "Time horizon: The length of task a model completes at a given reliability, counted in the time a skilled human would need, as METR measures it." --- # Time horizon What does “Time horizon” mean? Time horizonThe length of task a model completes at a given reliability, counted in the time a skilled human would need, as METR measures it. The Scorecard reads it at 50 and 80 percent reliability. *Chapter 3.* From *The AI Compute Race*. [Read the book](https://dryburgh.com/vol1#buy) Defined in [Chapter 3 — The Missing Researcher](https://dryburgh.com/vol1/notes/chapter-3). ## Where the book uses it - [Chapter 3 — The Missing Researcher](https://dryburgh.com/vol1/notes/chapter-3)The dated promise of an AI that does its own research, what the laboratories have published, and what the public record can and cannot yet check. ## Where the record uses it - Scorecard indicator [Post-2024 METR horizon doubling time](https://dryburgh.com/vol1/scorecard#row-3) ## Also discussed in - Dated test [Chapter 3, “attempt” and “leave with it”](https://dryburgh.com/vol1/tests#2027-6) ## Sources that use it - METR, “Time horizons” dashboard (Time Horizon 1.1), last updated 8 May 2026 (Claude Mythos Preview: 17.41 hours at 50 percent, 3.10 hours at 80 percent; “Measurements above 16 hrs are unreliable with our current task suite”; the dashboard’s post-2023 doubling-time estimate 128.7 days; the 130.8-day figure is from METR’s Time Horizon 1.1 post of 29 January 2026; accessed 16 September 2026, still last updated 8 May 2026, with no entry for GPT-5.6 Sol or for GPT-6 Astra, released 3 September; the dashboard’s Updates block: “March 3rd, 2026: Corrected a regularization mistake that affected our measurements”; Opus 4.5 at about 293 minutes and no since-2024 doubling figure; of what a time horizon does not measure, “Most jobs are not composed of well-specified, algorithmic tasks. Instead, they tend to require interacting with other people and involve success metrics that cannot be algorithmically scored”; read again on 23 September 2026, the newest update still that of 8 May 2026; as served on 13 October 2026 the page says “This page is no longer actively updated.”, shows LAST UPDATED September 8, 2026, and its Updates block opens “September 8th, 2026: Clarified that this page is no longer actively updated.”, while its newest model entry is still “May 8th, 2026: Added Claude Mythos Preview (early)”; the copies the house took on 23 September and 4 October 2026 carry no such notice, so when it first appeared is not established; the page carried the notice “This page is no longer actively updated.” with its last update dated 8 September 2026, read on 29 October 2026). [In the Sources](https://dryburgh.com/vol1/sources/research#metr-dashboard) ## Related terms - [METR](https://dryburgh.com/vol1/glossary/metr) - [Adoption clock](https://dryburgh.com/vol1/glossary/adoption-clock) - [Agent](https://dryburgh.com/vol1/glossary/agent) - [AGI](https://dryburgh.com/vol1/glossary/agi) - [Alignment](https://dryburgh.com/vol1/glossary/alignment) - [ARC-AGI-3](https://dryburgh.com/vol1/glossary/arc-agi-3) - [Astra](https://dryburgh.com/vol1/glossary/astra) - [Automated AI researcher](https://dryburgh.com/vol1/glossary/automated-ai-researcher) - [Benchmark](https://dryburgh.com/vol1/glossary/benchmark) - [Brake, the](https://dryburgh.com/vol1/glossary/brake-the) - [Capability clock](https://dryburgh.com/vol1/glossary/capability-clock) - [Chain of thought](https://dryburgh.com/vol1/glossary/chain-of-thought) From the glossary of The AI Compute Race. - [Campuses](https://dryburgh.com/vol1/ledger) - [Tests](https://dryburgh.com/vol1/tests) - [Scorecard](https://dryburgh.com/vol1/scorecard) - [Data](https://dryburgh.com/vol1/data/releases) - [Chapter by chapter](https://dryburgh.com/vol1/companion) - [The record at a glance](https://dryburgh.com/vol1/observatory) - [The machine](https://dryburgh.com/vol1/machine) - [One campus, four futures](https://dryburgh.com/vol1/hyperion) - [Glossary](https://dryburgh.com/vol1/glossary) - [People](https://dryburgh.com/vol1/people) - [Sources](https://dryburgh.com/vol1/sources) - [Notes](https://dryburgh.com/vol1/notes) - [Where the book was wrong](https://dryburgh.com/vol1/corrections) - [Where its sources were wrong](https://dryburgh.com/corrections?book=vol1) - [The Paper](https://dryburgh.com/vol1/paper) - [Consent](https://dryburgh.com/vol1/consent) - [Who builds and pays](https://dryburgh.com/vol1/parties) *The AI Compute Race: Five Races, One Machine* is a book by Lee S. Dryburgh, the first volume of THE COMPUTE ECONOMY, published by Dryburgh, an imprint of Contestra OÜ, Tallinn, Estonia; its record closes on 8 November 2026. [About the book](https://dryburgh.com/vol1)[Tell me when it is out](https://dryburgh.com/vol1#notify)