DeepSeek and Moonshot AI face Beijing's probe over potential data leaks to Anthropic

China's internet regulator has investigated DeepSeek and Kimi chatbot creator Moonshot AI over accusations that the artificial intelligence companies are routing user prompts and data to Anthropic's Claude, according to The Information.

The Cyberspace Administration of China has sent officials to the offices of the two companies to interview senior executives and staff, aiming to gauge the severity of any potential violations, the report said.

Beijing fears the two AI labs' routing of data through Anthropic may have leaked sensitive military, policing and corporate information to the US, potentially breaching the country's data security laws, the report added.

Earlier, Anthropic said in a report that Moonshot and DeepSeek allegedly routed live customer conversations, which sometimes included sensitive information, through Claude and used its responses as training data.

The report also unveiled that Anthropic had seen misuse from seven Chinese AI labs, including Alibaba (9988), Moonshot, DeepSeek, Z.AI (2513), MiniMax (0100), SenseTime (0020), and Xiaomi (1810).

𝗗𝗼𝘄𝗻𝗹𝗼𝗮𝗱 𝗧𝗵𝗲 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱 𝗔𝗽𝗽 ↓