Media report · AI-ASSISTED BRIEF

OpenAI Embeds Outside Safety Evaluators From Apollo, Goodfire, METR and Transluce

Report publication · YYSI publication

According to TechCrunch, OpenAI is embedding safety evaluators from Apollo Research, Goodfire, METR and Transluce in a pilot, TechCrunch reports. Apollo's work, starting this week, examines whether models attempt to subvert their own or successors' alignment controls; Goodfire monitors training runs, METR runs independent evaluations, and Transluce audits post-training safety tests. Apollo and Goodfire are compensated, and access is limited to information needed for subversion assessment.

Source & context

techcrunch.com · Read the original source ↗

This AI-assisted brief attributes information to the linked media report. It is not an official announcement or independent YYSI reporting. Consult the original report and its evidence.

Report a correction · Editorial standards