External Resource
Measuring AI Ability to Complete Long Tasks
METR (Model Evaluation & Threat Research)
METR's finding that the length of tasks frontier AI can complete autonomously (at 50% reliability) has doubled roughly every 7 months — empirical evidence of rapidly growing AI capability.
Theses this informs
- SuperintelligenceDiscusses
- LimitationsDiscusses
- AI ConstitutionDiscusses