Anthropic has accused Alibaba of using thousands of fake accounts to access Claude AI, alleging millions of interactions were used to extract model capabilities. The dispute highlights growing concerns over AI security, model distillation, and intensifying competition in the global AI industry.