Sanitize Training Data - AI Mitigation
AI MitigationDetect and remove or remediate poisoned training data. Training data should be sanitized prior to model training and recurrently for an active learning model. Implement a filter to limit ingested training data. Establish a content policy that would remove unwanted content such as certain explicit or offensive language from being used.
Overview
A source-backed snapshot of this defense.
Safeguard details
Where this defense applies and how the source classifies it.
- ATLAS ID
- AML.M0007
- Priority score
- 20
Covered techniques
Attacks this defense is designed to help with.
AML.T0010.002 - Data
Detect and remove or remediate poisoned data to avoid adversarial model drift or backdoor attacks.
AML.T0059 - Erode Dataset Integrity
Remediating poisoned data can re-establish dataset integrity.
AML.T0018.000 - Poison AI Model
Prevent attackers from leveraging poisoned datasets to launch backdoor attacks against a model.
AML.T0020 - Poison Training Data
Detect modification of data and labels which may cause adversarial model drift or backdoor attacks.
Source evidence
Original public records and references for this page.
Original source
Original source links
Open the public records and source datasets used for this page.