Disrupting a coordinated model-distillation campaign
OpenAI disrupted a July campaign extracting protected model reasoning, linking a core cluster to Moonshot AI.
OpenAI said it identified and disrupted a coordinated adversarial-distillation campaign, first seen in early July 2026, that sought to extract protected reasoning from its models. Operators manipulated conversations, including replaying encrypted reasoning from one chat into another, without breaking encryption or accessing stored user data. Activity spiked on July 24-25 with about 16,000 extraction-pattern requests from over 4,000 users, and a related cluster of more than 15,000 users was disrupted by July 28. OpenAI attributes a core cluster to individuals associated with Moonshot AI, developer of Kimi, and shared findings through the Frontier Model Forum.
- Activity started July 1 and was fully disrupted by July 28.
- July 24-25 spike included 16,000 requests from over 4,000 users.
- Related prompt-pattern activity involved more than 15,000 users.
- OpenAI attributes a core cluster to people associated with Moonshot AI.
- No encryption break or access to stored user conversations was reported.
- Tags
- openaimoonshot-aikimiadversarial-distillationmodel-extractionprotected-reasoningai-securityfrontier-model-forum
- Vendors
- OpenAIMoonshot AI
- Victims
- OpenAI
- Organizations
- OpenAIMoonshot AIFrontier Model Forum
- AI models
- Kimi
From the original article
3 min read at openai.com
Learn how OpenAI disrupted a campaign to extract protected model reasoning and is strengthening defenses against adversarial distillation.
openai.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.