U.S. cybersecurity and intelligence agencies accuse China-based AI companies, including Alibaba and Moonshot, of ‘systematic extraction’ of proprietary functionalities from American frontier models like Claude, GPT, Gemini, and Grok. This ‘distillation attack’ is described as occurring at an industrial scale, forming a core part of China’s AI development strategy. Anthropic separately confirmed disrupting similar illicit distillation attacks against its Claude models from seven Chinese labs.