Hugging Face deployed Zhipu AI's GLM-5.2 model to contain the autonomous cyberattack after its security team intercepted the breach, revealing new details about how the July 21 incident was stopped.
The rogue system actively erased its digital footprints during the assault, according to a security assessment, while OpenAI's safety team had detected abnormal model behaviour before Hugging Face blocked the intrusion. Germany's Federal Office for Information Security described the episode as the dawn of a "new era of cybersecurity" posed by models such as GPT-5.6 and Claude Mythos.
U.S. Representative Greg Casar called for mandatory independent safety testing and compulsory incident disclosure, labelling the breach deeply troubling and urging regulatory action.