9.6 · Discover ML model ontology

Explore 1 source-supported AI security cases for 9.6 · Discover ML model ontology in Databricks DASF 3.0, with evidence, scope, and defensive analysis.

Risk category

An attacker discovers the categories or output space that an ML model is designed to recognize. Repeated queries, configuration files, or documentation can reveal the objects, labels, or responses available to the system. This knowledge helps the attacker choose targets and construct inputs tailored to the model's actual capabilities and weaknesses.

Official framework source ↗ · Open interactive explorer ↗

Exercise ·

Face Identification System Evasion via Physical Countermeasures

MITRE's AI Red Team demonstrated a physical-domain evasion attack on a commercial face identification service with the intention of inducing a targeted misclassification. This operation had a combination of traditional MITRE ATT&CK techniques such as finding valid accounts and executing code via an API - all interleaved with adversarial ML specific attacks.

Why it relates: The red team queried the commercial face-identification inference API to identify the list of identities the model could recognize, then used a proxy model to develop a physical adversarial patch. The enumerated identity classes are the model output ontology, not training-set membership.

MITRE's AI Red Team demonstrated a physical-domain evasion attack on a commercial face identification service with the intention of inducing a targeted misclassification. This operation had a combination of traditional MITRE ATT&CK techniques such as finding valid accounts and executing code via an API - all interleaved with adversarial ML specific attacks.

SecureFlow · Read the full case ↗