genai / news / / CNBC TV18
OpenAI, Google and Anthropic are sharing data via the Frontier Model Forum to detect Chinese adversarial distillation, targeting firms like DeepSeek over security and profit.
OpenAI, Google, and Anthropic are sharing data to detect Chinese adversarial distillation via the Frontier Model Forum.
KEY POINTS
- US AI firms specifically target Chinese companies like DeepSeek, which allegedly use distillation to replicate proprietary models.
- Anthropic identified DeepSeek, Moonshot, and MiniMax as illicitly extracting capabilities from its Claude chatbot via distillation.
- Information sharing on adversarial distillation is limited by antitrust concerns, and firms seek clearer US government guidance.
- US officials estimate unauthorized distillation costs Silicon Valley AI labs billions of dollars in annual profit.
COMPANIES
Summarized by Newsio from CNBC TV18. How we summarize →