US AI leaders are using Chinese open models for cybersecurity. Here’s why

US AI leaders are using Chinese open models for cybersecurity. Here’s why

Andrew Ng and Hugging Face use Chinese AI models (Kimi K3, GLM-5.2) for security review because OpenAI and Anthropic refuse—their safeguards block legitimate cybersecurity tasks. Reliance on Chinese open-weight alternatives signals a gap in US model vendor strategy, pushing IT leaders toward new procurement choices in compliance-heavy security operations.