Anthropic accuses Chinese labs of trying to illicitly take Claude’s capabilities

Anthropic alleges that three Chinese AI labs ran industrial-scale distillation campaigns against its Claude model—using roughly 24,000 fraudulent accounts and some 16 million queries—to siphon capabilities for their own models. The startup warns that illicitly distilled models lack safety guardrails and could be fed into military, intelligence, or surveillance systems to enable offensive cyber operations, disinformation, and mass surveillance, and it reiterated calls for stronger export controls.