Training Data Poisoning
From Cognitive Attack Taxonomy
Training Data Poisoning
Short Description: Injecting false data into training dataset to induce poor model performance.
CAT ID: CAT-2023-005
Layer: 7
Operational Scale: Operational
Level of Maturity: Proof of Concept
Category: Exploit
Subcategory:
Also Known As:
Description:
Brief Description:
Closely Related Concepts:
Mechanism:
Multipliers:
Detailed Description: Data poisoning exploits AI/ML model vulnerability to fales or misleading training data, which leads to maladjusted predictions and/or model performance.
INTERACTIONS [VETs]:
Examples:
Use Case Example(s):
Example(s) From The Wild:
