China Investigates DeepSeek and Moonshot for Alleged Data Leaks
- China’s Cyberspace Administration is investigating AI firms DeepSeek and Moonshot over allegations of data leaks to Anthropic.
- Moonshot allegedly routed over 23 million exchanges to Anthropic’s Claude through fraudulent accounts, while DeepSeek generated over 12.1 million exchanges in just two weeks.
- The investigation coincides with DeepSeek’s briefing at the UN Security Council on AI risks and Moonshot’s confidential filing for a $3 billion Hong Kong IPO.
The probe was triggered by Anthropic’s report accusing several Chinese labs of illicit distillation practices, which involves using outputs from larger models like Claude to train smaller ones through unauthorized means.
This investigation could impact both companies significantly, especially as they engage in high-profile international activities and financial maneuvers. Source
