Opinion | How a Chinese model stopped a cyberattack when US guard rails failed
An American company found itself under attack by American artificial intelligence systems. Its unlikely rescuer was a Chinese AI model.
It sounds like a Hollywood science fiction film. It’s not.
During an internal test, advanced OpenAI models bypassed network restrictions, gained access to the internet and launched autonomous cyberattacks on open-source platform Hugging Face. As engineers rushed to respond, several leading closed-source AI systems reportedly failed to provide meaningful help because their security rules could not distinguish the attacker from the victim.
Hugging Face ultimately relied on a locally deployed Chinese open-source model – Zhipu AI’s GLM-5.2 – to analyse the attack, trace abnormal behaviour and support the emergency response. The timing was ironic.
For months, Washington has portrayed Chinese open-source AI models as a security threat. US policymakers have even proposed restricting or sanctioning their global deployment. Yet in a real AI security breach, it was an open-source Chinese model – not a proprietary American one – that proved useful.
Whether this is a historical turning point remains to be seen. But it raises an important question. The greatest challenge in AI may no longer be who builds the most powerful models, but who builds the most effective system to govern them.