ATLASAML.T0015
ATLAS index
AML.T0015

Evade AI Model

Adversaries can Craft Adversarial Data that prevents an AI model from correctly identifying the contents of the data or Generate Deepfakes that fools an AI model expecting authentic data. This technique can be used to evade a downstream task where AI is utilized. The adversary may evade AI based virus/malware detection

Framework
MITRE ATLAS
Maturity
Realized
Platforms
Predictive AI
Release
2026.05

Overview

Adversaries can Craft Adversarial Data that prevents an AI model from correctly identifying the contents of the data or Generate Deepfakes that fools an AI model expecting authentic data.

This technique can be used to evade a downstream task where AI is utilized. The adversary may evade AI-based virus/malware detection or network scanning towards the goal of a traditional cyber attack. AI model evasion through deepfake generation may also provide initial access to systems that use AI-based biometric authentication.

Sources

  1. MITRE ATLAS AML.T0015: Evade AI Model — MITRE