AML.CS0012
Face Identification System Evasion via Physical Countermeasures
- Type
- Exercise
- When
- 2020-01-01
- Actor, as ATLAS names it
- Not stated
- Target
- Commercial Face Identification Service
- ATLAS attributes it to MITRE AI Red Team (security evaluation).
Techniques, in order
- 1Search Open Technical DatabasesReconnaissance · AML.T0000
- 2Valid AccountsInitial Access · AML.T0012
- 3AI Model Inference API AccessAI Model Access · AML.T0040
- 4Discover AI Model OntologyDiscovery · AML.T0013
- 5Acquire Public AI Artifacts: DatasetsResource Development · AML.T0002.000
- 6Replicate AI Model BehaviorAI Attack Adaptation · AML.T0005
- 7Craft Adversarial Data: White-Box OptimizationAI Attack Adaptation · AML.T0043.000
- 8Acquire Infrastructure: Physical CountermeasuresResource Development · AML.T0008.003
- 9Physical Environment AccessAI Model Access · AML.T0041
- 10Evade AI ModelImpact · AML.T0015
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.