China Investigates DeepSeek and Moonshot Over Alleged Data Leaks to Anthropic
China's CAC is probing DeepSeek and Moonshot over allegations that sensitive state, police and military data was routed to Anthropic's Claude models.
China has launched an investigation into two of its prominent artificial intelligence firms, DeepSeek and Moonshot, over potential data security breaches. The probe centres on allegations that sensitive information was routed to models operated by the US company Anthropic.
The concerns stem from claims that DeepSeek and Moonshot directed sensitive data to Anthropic's Claude models. The information involved is reported to include data linked to Chinese state, police and military entities.
The investigation is being conducted by China's Cyberspace Administration (CAC), which is examining the two companies for possible data security violations.
As part of the same inquiry, the regulator summoned representatives of several other Chinese technology companies for questioning. Those called in include Alibaba, Zhipu, SenseTime, MiniMax and Xiaomi.
The development highlights growing scrutiny of cross-border data flows involving artificial intelligence systems, particularly where national security interests are concerned.