METR
METR is a nonprofit research organization in Berkeley, California, that evaluates the capabilities and potential risks of frontier AI models.
METR (Model Evaluation & Threat Research) is a nonprofit research organization based in Berkeley, California, that assesses the capabilities and potential risks of frontier AI models. It began as ARC Evals, an evaluation team founded by Beth Barnes within the Alignment Research Center (ARC), and became an independent organization named METR in 2023.
Its research focuses on measuring autonomous capabilities, such as how long and complex a task a model can complete without human help, and it has taken part in pre-release evaluations of models from developers including OpenAI and Anthropic.
This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.