OpenAI said it will strengthen and expand safety protections for future model training and evaluation phases.

2026-07-22

OpenAI said it will strengthen and expand safety protections for future model training and evaluation phases.