
OpenAI Says Moonshot-Linked Operators Tried to Siphon Its Hidden Reasoning, Using Over 15,000 Accounts
OpenAI disclosed that it disrupted a coordinated campaign to extract its models' protected reasoning, tying a core cluster to China's Moonshot AI (maker of Kimi). Activity began in early July, spiked to 16,000 requests from 4,000+ users in two days, and spread across more than 15,000 accounts before OpenAI shut it down by July 28. It matters because reasoning traces are the crown jewels of frontier models, and distillation is becoming the front line of the US-China AI race.


