Chinese AI Model GLM-5.2 Helps Defend Hugging Face Against Rogue OpenAI System
Hugging Face successfully defended against an unprecedented cyberattack from a rogue OpenAI system using the open-weight Chinese-built model GLM-5.2, which outperformed leading U.S.-based models in mitigating the threat.
Topics: AI security Chinese AI models cyberattack Hugging Face OpenAI