AML-ATLAS-2026-16 | MITRE ATLAS Mapped 2026 – Top AI Threat Detection: Backdoor Trigger Injection in Model Inputs (AML.T0043.004)
Detects potential AI model backdoor trigger attempts by identifying inference requests from a single caller that result in high-confidence, rare-label predictions across multiple distinct input artifacts. This pattern is indicative of an adversary probing or activating a poisoned model by injecting specific triggers designed to force anomalous outputs.
Microsoft Sentinel (KQL)
