PrivacyMedium
Anthropic Said DeepSeek, Moonshot, and MiniMax Used Fraudulent Accounts and Proxies to Illicitly Distill Claude Capabilities at Scale
- Occurred
- Feb 23, 2026
- Published
- Feb 23, 2026
- Source
- AI Incident Database
Summary
Anthropic said it identified large-scale campaigns that used fraudulent accounts and proxy services to generate high volumes of Claude interactions to extract model capabilities for competitor training ("distillation"). Anthropic attributed the activity to DeepSeek, Moonshot, and MiniMax and said it involved millions of exchanges across thousands of accounts, violating its terms and access restrictions. Anthropic described detection measures, account controls, and indicator-sharing in response.
Developers
- Anthropic
Deployers
- DeepSeek
- Moonshot AI
- MiniMax
- Proxy reseller services
Harmed parties
- Anthropic
- Claude users
- Anthropic customers
- National security and intelligence stakeholders
Tags
- response
Classification
Category confidence: 85%
Severity confidence: 80%
Rationale: Fraudulent account usage and unauthorized data extraction violate terms and access controls, fitting privacy/security breach category.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8