OpenAI Accuses Moonshot of 16K-Request Distillation Attack
Summary
OpenAI says Chinese startup Moonshot ran a coordinated campaign to extract protected reasoning from its models, peaking at 16,000 requests from 4,000 users in two days. The activity began in July and involved 15,000 accounts by July 28. OpenAI claims no encryption was broken, but the scale and alleged link to Moonshot's Kimi chatbot raise IP and competitive risks. This follows Anthropic's similar accusations against Moonshot and others, and Treasury Secretary Bessent's July warning about sanctions on Chinese models using distilled American AI. Watch for any U.S. government response or further evidence linking Moonshot to the activity.
This news item was assessed with negative market sentiment and an importance score of 7 out of 10. Source: Seeking Alpha.