METR
An organization that measures AI capabilities, notably the time horizon of tasks AIs can complete (referenced as averaging about one hour).
As defined in AI 2027: month-by-month model of intelligence explosion — Scott Alexander & Daniel Kokotajlo
Where it comes up · 3
- AI Genius Predicts the Next 3 Years2 claims
- AI 2027: month-by-month model of intelligence explosion — Scott Alexander & Daniel Kokotajlo1 claim
- Dwarkesh Patel Podcast - The avalanche that hit math is coming for AI development1 claim
Most important claims · 4 of 4Search all claims for “METR” →
- METR, after three months embedded inside Anthropic, concluded that Anthropic's overall rate of progress on well-scoped tasks is 2x what it was a year ago, and that a majority of this acceleration is likely attributable to Anthropic's extensive use of AI to do R&D.Dwarkesh Patel Podcast - The avalanche that hit math is coming for AI development
- Just as AI capabilities are now measured by 'time horizon' (tasks lasting minutes vs hours), human writers may have differing time horizons: many can write a great comment or three-paragraph Tumblr post but can't sustain coherence across a full blog post, and current AI's ~1-hour task horizon likely explains why it can't yet plan and execute a research-heavy blog post despite strong sentence-level style.AI 2027: month-by-month model of intelligence explosion — Scott Alexander & Daniel Kokotajlo
- The METR autonomy metric, where the 18-hour figure represents the equivalent human time to complete the work (not the actual AI execution time), has been doubling approximately every 2-3 months for the past few years, with actual AI execution time being 100 times faster than the human equivalent (e.g., AI completes an 18-hour human task in 1-2 hours).AI Genius Predicts the Next 3 Years
- The METR (Methodology for Evaluating Transformative Robotics) autonomy metric measures how much human work an AI model can perform automatically before requiring human correction, and this metric has grown from approximately 10 seconds two to three years ago to 18-20 hours in the latest models like Claude Opus 4.6.AI Genius Predicts the Next 3 Years