AML.CS0056
Model Distillation Campaigns Targeting Anthropic Claude
- Type
- Incident
- When
- 2026
- Actor, as ATLAS names it
- Not stated
- Target
- Anthropic Claude
- ATLAS attributes it to DeepSeek (adversarial operation).
- ATLAS attributes it to Moonshot AI (adversarial operation).
- ATLAS attributes it to MiniMax (adversarial operation).
- Reported by Anthropic.
Techniques, in order
- 1Acquire Infrastructure: AI Service ProxiesResource Development · AML.T0008.005
- 2Valid AccountsInitial Access · AML.T0012
- 3LLM Prompt CraftingAI Attack Adaptation · AML.T0065
- 4Centralized AI Service Request RoutingDefense Evasion · AML.T0140
- 5Exfiltration via AI Inference APIExfiltration · AML.T0024
- 6Outputs from AI ModelsCollection · AML.T0135
- 7Replicate AI Model Behavior: Replicate Behavior from Model OutputsResource Development · AML.T0005.001
- 8External Harms: AI Intellectual Property TheftImpact · AML.T0048.004
- 9External Harms: Societal HarmImpact · AML.T0048.002
- 10External Harms: User HarmImpact · AML.T0048.003
From MITRE ATLAS 2026.10 (Apache-2.0). ShieldMap lists who and what was involved and the techniques used, and links to ATLAS for the write-up; it does not copy the steps. Attribution is ATLAS's, not ShieldPi's.