How to meet it
OWASP AISVS v1.01
Training Data Integrity & Traceability · Training Data Origin & Data Security
Train only on the data you need
Verify that training data includes only features, attributes, and fields required for the model's stated purpose.
OWASP AISVS v1.01, requirement 1.1.1
In plain English
Train only on the fields the model needs for its stated purpose. Every extra column is something that can leak, be memorised or be poisoned, and it buys you nothing.
What good looks like
- The model's purpose is written down, and each field in the training set is justified against it.
- Identifiers and sensitive fields that the purpose does not need are dropped before training.
- The field list is reviewed when the purpose or the data source changes.
How teams usually do it
- A data contract or schema for each training set, with a short reason next to every field.
- An extraction step that selects named columns, so new upstream fields are left out by default.
Evidence an auditor would accept
- The purpose statement and the approved field list for each dataset.
- The extraction code or query showing only those fields are selected.
Common mistakes
- Copying whole tables into the training store "in case they are useful later".
- Removing names but keeping fields that identify people when combined, such as postcode, date of birth and job title.
Not sure which requirements apply to your system? Take the assessment.
Requirement text from the OWASP AI Security Verification Standard, licensed under CC BY-SA 4.0. This guidance is shared under the same licence.