AI Safety Gatekeepers Step Into Spotlight

Safety teams that quietly test AI models for risks are gaining more public attention. As model capabilities grow, these gatekeepers play a more visible role in assessing potential harms before release.