OpenAI Launches the First Phase of Embedded Evaluations

Written by

in

OpenAI has launched the first phase of embedded evaluations, collaborating independently with Apollo Research, METR, Transluce, and Goodfire to evaluate model alignment, safeguards, and monitoring systems. The current materials do not provide the evaluation methodology, timeline, or specific results.

Source:OpenAI 新闻 · Read the original article

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *