US Agencies Warn China Is Systematically Extracting Frontier AI Capabilities
NSA, CISA and FBI warn Chinese AI firms including DeepSeek and Moonshot systematically extracted billions of tokens from US frontier models since late 2024.
The NSA, CISA, and FBI report that China-based AI companies including DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI extracted billions of tokens from US frontier models such as Claude, GPT-4/GPT-5, Gemini, and Grok 4 since late 2024. The distillation trained DeepSeek's R1 and V3 and Moonshot's Kimi-K2/K3 models, and the agencies mapped the tactics to MITRE ATLAS while noting additional novel techniques like subscription exploitation and request metadata sanitization. They describe the activity as a strategic economic threat to US technological leadership and recommend behavioral detection, differential privacy, and targeted cost-imposing responses.
- Agencies name DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI as extracting data since late 2024
- Distilled capabilities from Claude, GPT-4/5, Gemini, and Grok 4 trained DeepSeek R1/V3 and Kimi models
- Distillation TTPs mapped to MITRE ATLAS, plus novel techniques like subscription exploitation and metadata sanitization
- Recommendations include behavioral monitoring, differential privacy, and targeted responses to high-confidence distillation
- ATT&CK
- T1078 Valid Accounts
- Threat actors
- China
- Countries
- ChinaUnited States
From the original article
3 min read at securityweek.com
Distillation is an ‘attack’ against an AI model designed to capture outputs, understand reasoning processes, and subsequently train a different model. The post US Agencies Warn China Is Systematically Extracting Frontier AI Capabilities appeared first on SecurityWeek.
By Kevin Townsend · securityweek.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.