H3XAOperational intelligence
Live analytical pictureEvidence before assessment
Threat intelligence / Technique

Manual Modification

Adversaries may manually modify the input data to craft adversarial data. They may use their knowledge of the target model to modify parts of the data they suspect helps the model in performing its task. The adversary may use trial and error until they are able to verify they have a working adversarial input.

Technique
TypeTechnique
External ID—
First seen2026-08-14
Last seen2026-08-14

Confidence

—

Adversaries may manually modify the input data to craft adversarial data. They may use their knowledge of the target model to modify parts of the data they suspect helps the model in performing its task. The adversary may use trial and error until they are able to verify they have a working adversarial input.

9 recorded

Relationships

mitigates
from Limit AI Service Query Volume and Rate
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
subtechnique-of
to Craft Adversarial Data
Technique
uses
from Attempted Evasion of ML Phishing Webpage Detection System
Campaign
uses
from Bypassing Cylance's AI Malware Detection
Campaign
uses
from Evasion of Deep Learning Detector for Malware C&C Traffic
Campaign
1 external

References

mitre-atlas
AML.T0043.003
open
Continue into the dataIntelligence explorerHow to read the graphEvidence & provenance