SecurityRegulation

Anthropic accuses Alibaba of largest Claude cloning attack

Anthropic alleges that Alibaba generated over 28.8 million queries to Claude through approximately 25,000 fraudulent accounts between April and June 2026 to clone the model.

In detail

  • Alibaba operators generated 28.8 million exchanges with Claude between April 22 and June 5, 2026 using ~25,000 fake accounts.
  • Targets included high-value capabilities such as agentic reasoning, software engineering, and long-horizon tasks.
  • Alibaba used obfuscation techniques and proxy networks; Anthropic warns of growing "circumvention economy" enabling future distillation attacks.

Why it matters

This signals escalating geopolitical tensions over AI technology. For German companies using US models, this means data protection and access control are becoming critical competitive factors.

For you Review how your use of US AI models is protected against unauthorized extraction—especially if sensitive business data is involved.

← All news

Summaries are generated automatically and link to the original source.