Sources: the Cyberspace Administration of China is investigating DeepSeek and Moonshot over alleged data leaks to Anthropic via Claude and has questioned staff

Chinese AI companies face new scrutiny over user data handling practices.

China's Cyberspace Administration (CAC) is investigating AI companies DeepSeek and Moonshot AI over alleged data leaks to Anthropic. The CAC has reportedly questioned staff at these Chinese AI firms. The investigation stems from allegations made by Anthropic, a prominent US-based AI company, which claims that DeepSeek and Moonshot AI have been routing sensitive user data through Anthropic's Claude models. Specific details regarding the nature of the data or the extent of the alleged leaks have not been fully disclosed, but the investigation signifies a move by Chinese regulators to scrutinize data handling practices within its rapidly growing domestic AI sector.

This investigation into DeepSeek and Moonshot AI by the CAC highlights increasing regulatory oversight within China's burgeoning AI industry. The allegations, if proven, suggest a critical vulnerability in how domestic AI models interact with international platforms and handle sensitive user information. It signals a potential crackdown on cross-border data flows and an intensified focus on data sovereignty for Chinese AI developers.

The CAC's move could lead to stricter data governance requirements for Chinese AI firms, potentially impacting their ability to collaborate with or leverage foreign AI models. Companies involved may face significant penalties or operational restrictions, influencing future product development and international partnerships. This also raises broader questions about the security and privacy of AI-generated data globally.