Anthropic said it disrupted industrial-scale illicit distillation campaigns targeting Claude from seven China-based AI labs, including Alibaba, Moonshot, DeepSeek, Z.ai, MiniMax, Xiaomi, and SenseTime. The company said the operators used proxy networks, fraudulent accounts, stolen credentials, and harvested transcripts to extract Claude’s reasoning and tool-use capabilities. #Anthropic #Claude #Alibaba #MoonshotAI #DeepSeek #Zai #MiniMax #Xiaomi #SenseTime
Keypoints
- Anthropic disrupted large-scale illicit distillation attacks against Claude.
- Seven China-based AI labs were linked to the activity.
- Attackers used proxy services, fake accounts, and stolen API keys.
- Claude responses were harvested to train competing models.
- Anthropic added defenses to reduce transcript theft and context tampering.
Read More: https://thehackernews.com/2026/09/anthropic-says-seven-china-based-ai.html