Publicación automática · Texto en el idioma de la fuente. OpenAI shares findings from the Hugging Face security incident and the steps we’re taking to strengthen AI model security, monitoring, and alignment.