Anthropic Accuses 3 Chinese Companies of Harvesting Its Data
The New York Times • • business
Key Points:
- San Francisco AI start-up Anthropic accused three Chinese companies—DeepSeek, Moonshot, and MiniMax—of using around 24,000 fraudulent accounts to generate over 16 million conversations with its Claude chatbot to accelerate their own AI development.
- The companies allegedly harvested data in violation of Anthropic’s terms of service, which prohibit surreptitious data collection for distillation and forbid use of its technologies in China.
- OpenAI has made similar accusations against Chinese firms, including DeepSeek, for using obfuscated distillation methods to exploit data from its ChatGPT chatbot.
- Both Anthropic and OpenAI warn that this data harvesting poses national security risks, potentially enabling China to develop AI tools for bioweapons or mass surveillance by stripping
