Anthropic has formally accused Alibaba of orchestrating what it describes as the largest coordinated AI distillation attack in history, with operators affiliated with the Chinese tech giant's Qwen AI lab conducting 28.8 million fraudulent exchanges with Claude AI using roughly 25,000 fake accounts over a 44-day period between April 22 and June 5, 2026. The company sent a letter to U.S. officials detailing how Alibaba "brazenly" and "illicitly" attempted to extract its most advanced AI capabilities.

Targeted Capabilities: Claude's Crown Jewels

The attack wasn't random probing—it was surgical. Anthropic says Alibaba's team systematically targeted Claude's software engineering, agentic reasoning, and long-horizon task execution capabilities—the very features that power the company's cutting-edge Mythos Preview model. This methodical approach suggests deep knowledge of where Anthropic's competitive advantages lie.

A Pattern of Chinese AI Firms Targeting U.S. Models

This isn't an isolated incident. Anthropic revealed it had already uncovered a separate DeepSeek scheme involving 150,000+ exchanges, along with attacks from Moonshot AI (3.4 million) and MiniMax (13 million). But Alibaba's campaign dwarfs all previous attempts combined—nearly double the prior largest attack. The company noted that Alibaba "ignored the Trump Administration's warnings" in proceeding with these operations.

Market Reacts Swiftly

Investors wasted no time punishing Alibaba for the allegations. Shares of Alibaba Group Holding Ltd. tumbled as much as 4.9% in Hong Kong trading, sinking to a 16-month low. The ripple effects hit other Chinese AI players: Xiaomi and Baidu both dropped more than 3%, signaling market concern about compliance risks across China's AI sector.

Beyond IP Theft: National Security Implications

Anthropic warned that the implications extend far beyond corporate espionage. AI systems built through adversarial distillation often lack safety guardrails, posing broader security risks. The timing matters—the letter lands two months after the White House Office of Science and Technology Policy issued a memorandum pledging to help AI companies detect and coordinate against industrial-scale distillation attacks.

Regulatory Pressure Mounts on Alibaba

The distillation allegations come as Alibaba faces mounting U.S. regulatory pressure. The company was added this month to the Pentagon's list of Chinese military companies—a designation it is currently challenging. Alibaba has not responded to the new distillation allegations, continuing to deny similar accusations while maintaining all research operations comply with applicable regulations.

Key Takeaways

  • 28.8M fraudulent exchanges represent the largest known AI distillation attack in history
  • Qwen Lab operators used ~25,000 accounts over 44 days specifically targeting Claude's agentic and software engineering capabilities
  • Alibaba stock hit a 16-month low following the accusations, dragging down other Chinese tech firms
  • The attack follows a documented pattern of Chinese AI companies systematically extracting value from U.S. models