U.S. Accuses China of Industrial-Scale AI Model Distillation, Beijing Rejects Claims and Threatens Retaliation

U.S. intelligence agencies have accused six Chinese artificial intelligence companies of conducting systematic, large-scale theft of American AI trade secrets using knowledge distillation—a technique that trains less-advanced models on proprietary U.S. systems to extract restricted capabilities. The NSA, FBI, and CISA jointly alleged in September 2026 that companies including DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI have extracted data from leading U.S. models including OpenAI's ChatGPT and GPT-5 variants, Anthropic's Claude Opus and Sonnet models, Google's Gemini, and xAI's Grok.