White House Accuses China of Industrial-Scale AI Model Distillation

The White House says foreign actors, mainly China, are running an “industrial-scale” AI model distillation campaign to steal American frontier AI capabilities and proprietary data. In a memo titled “Adversarial Distillation of American AI Models,” Michael Kratsios (Assistant to the President for Science and Technology, OSTP Director) cited evidence of coordinated efforts using tens of thousands of proxy accounts and jailbreak techniques. The administration warns that unauthorized AI model distillation can make copied models appear similar on some benchmarks at lower cost, while also potentially removing safety safeguards designed to keep AI “ideologically neutral and truth-seeking.” It also distinguishes lawful distillation for efficient open-source/open-weight models from “systematically extracting and copying” industry innovations. Next steps: federal agencies will work with U.S. AI companies to strengthen protections around frontier models, coordinate defenses with the private sector, and explore ways to hold foreign actors accountable. The concern is primarily about AI security and IP risk rather than immediate financial policy, but it could affect AI-sector sentiment and related funding decisions.