Threat intelligence / Technique

Evade AI Model

Adversaries can [Craft Adversarial Data](/techniques/AML.T0043) that prevents an AI model from correctly identifying the contents of the data or [Generate Deepfakes](/techniques/AML.T0088) 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.

Technique

TypeTechnique

External ID—

First seen2026-08-14

Last seen2026-08-14

Confidence

—

Adversaries can [Craft Adversarial Data](/techniques/AML.T0043) that prevents an AI model from correctly identifying the contents of the data or [Generate Deepfakes](/techniques/AML.T0088) 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.

25 recorded

Relationships

mitigates

from AI Red Team

Mitigation

mitigates

from Deepfake Detection

Mitigation

mitigates

from Predictive AI Adversarial Input Detection

Mitigation

mitigates

from Predictive AI Ensembles

Mitigation

mitigates

from Predictive AI Input Restoration

Mitigation

mitigates

from Predictive AI Model Hardening

Mitigation

mitigates

from Predictive AI Multi-Sensor Fusion

Mitigation

uses

from AI Model Tampering via Supply Chain Attack

Campaign

uses

from Attack on Machine Translation Services

Campaign

uses

from Attempted Evasion of ML Phishing Webpage Detection System

Campaign

uses

from Backdoor Attack on Deep Learning Models in Mobile Apps

Campaign

uses

from Botnet Domain Generation Algorithm (DGA) Detection Evasion

Campaign

uses

from Bypassing Cylance's AI Malware Detection

Campaign

uses

from Bypassing ID.me Identity Verification

Campaign

uses

from Camera Hijack Attack on Facial Recognition System

Campaign

uses

from Confusing Antimalware Neural Networks

Campaign

uses

from Evasion of Deep Learning Detector for Malware C&C Traffic

Campaign

uses

from Face Identification System Evasion via Physical Countermeasures

Campaign

uses

from Google Photos AI Model Extraction

Campaign

uses

from Live Deepfake Image Injection to Evade Mobile KYC Verification

Campaign

uses

from Malware Prototype with Embedded Prompt Injection

Campaign

uses

from Microsoft Azure Service Disruption

Campaign

uses

from Microsoft Edge AI Evasion

Campaign

uses

from ProKYC: Deepfake Tool for Account Fraud Attacks

Campaign

uses

from ProofPoint Evasion

Campaign

1 external

References

mitre-atlas

AML.T0015