PulseAugur
EN
LIVE 02:56:45

METR updates AI time horizon estimates with new evaluation framework

By PulseAugur Editorial · [1 sources] ·

METR has released Time Horizon 1.1 (TH1.1), an updated evaluation of AI model capabilities. This new version expands the task suite from 170 to 228 tasks, incorporating more long-duration tasks and utilizing the Inspect framework from the UK AI Security Institute. While most new estimates fall within previous confidence intervals, the trend in time horizon growth shows some differences, with tighter confidence intervals for recent models due to the enhanced task suite. AI

IMPACT Provides updated benchmarks for AI model capabilities, influencing future research and development directions.

RANK_REASON The item describes an updated methodology and evaluation of AI models, which falls under research. [lever_c_demoted from research: ic=1 ai=1.0]

Read on METR (Model Evaluation & Threat Research) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

METR updates AI time horizon estimates with new evaluation framework

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The item describes an updated methodology and evaluation of AI models, which falls under research. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
273 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. METR (Model Evaluation & Threat Research) TIER_1 (SQ) ·

    Time Horizon 1.1

    <p><strong>We’re releasing a new version of our time horizon estimates (TH1.1), using more tasks and a new eval infrastructure.</strong></p> <p>Our estimates of time horizons for many models have been updated. The new estimates generally fall within our existing confidence interv…

RELATED ENTITIES

RELATED TOPICS