Newsletter Subscribe
Enter your email address below and subscribe to our newsletter
[forminator_form id="25163"]

bloomberg+1bloombergreutersAnthropic on Thursday published a new report accusing Chinese AI company Moonshot AI of covertly funneling nearly 300,000 user requests to its Claude models — particularly the Opus model — without informing users, then using the responses to train its own Kimi chatbot in what the U.S. firm described as a systematic distillation campaign.bloomberg+1
The report adds granular detail to allegations Anthropic first raised in February, when it disclosed that Chinese AI labs had collectively generated more than 16 million exchanges with Claude through roughly 24,000 fraudulent accounts.mashable+1
According to Anthropic, Moonshot used 5,380 fraudulent accounts — most appearing to be located in Singapore and Japan — to circumvent the company's policy barring access from inside China. The requests were routed primarily to Anthropic's Opus model, and Anthropic said Moonshot redirected extraction traffic to newly released Claude models within 24 hours of launch, suggesting a standing, pre-built operation rather than an ad hoc effort.wsj+1
Anthropic determined the activity was part of a distillation process, in which a developer harvests outputs from a more capable AI model to improve a weaker one. In this case, Anthropic said it believes the Claude responses were used to train Moonshot's Kimi models. The report also named DeepSeek and Xiaomi as having engaged in similar practices.bloomberg
The report lands two days after the U.S. government issued a joint advisory through CISA accusing six Chinese AI companies — including DeepSeek, Moonshot AI, Alibaba , MiniMax, StepFun, and Z.AI — of extracting "billions of tokens" from American AI systems. Reuters reported that U.S. officials characterized the activity as "aggressive, malicious and targeted distillation activities at an industrial scale," a move that could further strain relations ahead of a planned meeting between leaders of the two countries later this month.cisa+1
Representatives for Moonshot, DeepSeek, and Xiaomi did not immediately respond to requests for comment. China's Commerce Ministry has previously said there is "no factual or legal basis" for accusations of industrial-scale distillation by Chinese AI companies and threatened countermeasures if the U.S. "takes action to contain and suppress" Chinese AI firms.bloomberg