AML.CS0065
Model Namespace Reuse Supply Chain Attack
- Type
- Exercise
- When
- 2025-09-03
- Actor, as ATLAS names it
- Not stated
- Target
- Users of Hugging Face-backed model catalogs, code pipelines, and cloud integrations
- ATLAS attributes it to Unit 42 (security evaluation).
Techniques, in order
- 1Search Open Websites/DomainsReconnaissance · AML.T0095
- 2Establish AccountsResource Development · AML.T0021
- 3MasqueradingDefense Evasion · AML.T0074
- 4Manipulate AI Model: Embed MalwareAI Attack Adaptation · AML.T0018.002
- 5Publish Poisoned AI Artifacts: ModelsResource Development · AML.T0115.001
- 6Supply Chain Compromise: Model RepositoryInitial Access · AML.T0010.003
- 7User Execution: Unsafe AI ArtifactsExecution · AML.T0011.000
- 8Cyber Communication ChannelCommand and Control · AML.T0072
How these attacks work
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.