CISA, NSA and FBI warn of China-based AI firms' distillation campaigns against US AI models
AI-generated. This page was written and translated automatically by AI and checked by software against the cited sources. No human editor reviewed it. Not legal advice.
Summary
CISA, working with the NSA and the FBI, released a joint cybersecurity advisory warning that China-based AI companies are targeting U.S. AI companies to extract proprietary models through knowledge distillation. The advisory says that, likely with Chinese government awareness, DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun and Z.AI extracted billions of tokens across millions of requests from U.S. frontier models since at least late 2024. It recommends that U.S. frontier companies implement detection and mitigation, deploy targeted response changes for suspected distillation attempts, and establish cross-organization intelligence sharing.
Why it matters
The advisory addresses U.S. frontier AI developers, urging immediate steps to protect their platforms from distillation campaigns. It characterizes the activity as violating terms of service and undermining fair competition, and it recommends three actions covering detection, altered responses and information sharing across model providers, cloud platforms and API aggregators.
- Parties involved
- CISA; NSA; FBI
- Primary source
- CISA (cisa.gov)
Source retrieved: . Sources are often in other languages; links open the original. - Topics
- Frontier models & safety · Competition & antitrust · General AI governance