‍Anthropic Accuses Chinese Firms of Massive AI Distillation Attacks

The AI laboratory Anthropic has accused three China-based AI developers—DeepSeek, Moonshot, and MiniMax—of conducting extensive "distillation" attacks against its proprietary Claude models. Anthropic alleges that these firms used millions of automated exchanges via approximately 24,000 fraudulent accounts to siphon intellectual property and enhance their own AI architectures.

The targeted capabilities included agentic reasoning, coding and data analysis, computer vision, and tool use. Anthropic warns that such unauthorized scraping circumvents R&D costs and could enable authoritarian governments to deploy advanced AI for offensive cyber operations, disinformation, and surveillance. This incident highlights growing concerns over data security and geopolitical competition in the global AI landscape, drawing parallels to threats faced by the cryptocurrency and blockchain sectors.

https://cryptovka.com/news/anthropic-accuses-chinese-firms-of-massive-ai-distillation-attacks