AML.CS0076
China-Based Artificial Intelligence Companies Conducting Industrial-Scale Distillation Campaigns Against U.S. AI Companies
- Type
- Incident
- When
- 2024
- Actor, as ATLAS names it
- Not stated
- Target
- Models from Anthropic, OpenAI, Google, and XAI
- ATLAS attributes it to DeepSeek (adversarial operation).
- ATLAS attributes it to Moonshot AI (adversarial operation).
- ATLAS attributes it to MiniMax (adversarial operation).
- ATLAS attributes it to Alibaba Group (adversarial operation).
- ATLAS attributes it to StepFun (adversarial operation).
- ATLAS attributes it to Z.AI (adversarial operation).
Techniques, in order
- 1Establish AccountsResource Development · AML.T0021
- 2Acquire Infrastructure: AI Service ProxiesResource Development · AML.T0008.005
- 3Valid AccountsInitial Access · AML.T0012
- 4Centralized AI Service Request RoutingDefense Evasion · AML.T0140
- 5AI Service Request Metadata SanitizationDefense Evasion · AML.T0141
- 6Discover AI Model OutputsDiscovery · AML.T0063
- 7LLM Prompt CraftingAI Attack Adaptation · AML.T0065
- 8LLM Prompt Injection: DirectExecution · AML.T0051.000
- 9LLM JailbreakDefense Evasion · AML.T0054
- 10Exfiltration via AI Inference APIExfiltration · AML.T0024
- 11Outputs from AI ModelsCollection · AML.T0135
- 12Verify AttackAI Attack Adaptation · AML.T0042
- 13Replicate AI Model Behavior: Replicate Behavior from Model OutputsResource Development · AML.T0005.001
- 14External Harms: AI Intellectual Property TheftImpact · AML.T0048.004
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.