Home / AWS Certified AI Practitioner (AIF-C01) / 1 Practice / AWS-AIF-518
Records reach their retention deadline.
Prevent them entering training.
Choose two practices.
Hint 1: Follow the data beyond the source file Apply retention and legal-hold rules to each dataset version and job, then trace indexes, derived data, and model impact after removal.
Hint 2: Key terms Review the lifecycle, retention, derivative-data, and audit concepts behind expiring training data.
Data lifecycleThe stages through which data moves from collection and use to retention, update, and deletion.
Retention policyA rule that defines how long data is kept and when it may be deleted.
Legal holdA controlled exception that pauses normal deletion while a legal or audit need exists.
Derived dataAn index, feature set, or training artifact produced from source data.
Model impact analysisAn assessment of which models or outputs depend on data that is changed or removed.
Audit trailA traceable record of exclusions, deletions, approvals, and resulting checks. × Incorrect Apply collection date, retention, legal hold, and consent scope to dataset versions and jobs, then verify removal from indexes and model impact.
Detailed explanation A Incorrect Data in a trained model is never subject to retention considerations
Impact still needs assessment.
B Correct Verify exclusion, deletion, index update, model impact, and audit results
Derivative use is checked.
C Incorrect Continue reusing expired datasets
Purpose is exceeded.
D Correct Apply collection, retention, legal hold, and consent conditions to dataset versions and jobs
Expired records are excluded.
Check official documentation Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 1.3・5.1のデータライフサイクル、保持、モデル影響を確認する。
Official references Official documentation 1 Official documentation 2 What to confirm データ期限を原本だけでなく索引・学習・モデル影響まで追跡できる。
Key points Expiration Retention Model impact Notes Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認 Check whether indexes, feature sets, caches, training outputs, and backups retain copies after source deletion. Record the rule, approval, deletion result, and model-impact decision so the lifecycle can be audited later. Foundation review Retention scope A deletion rule must identify the data version, derived artifacts, and processing jobs it covers.
Legal hold A legal hold is a limited, approved exception to normal deletion, not a reason to keep every record forever.
Model impact Removing training data can require a lineage review, retraining decision, or disclosure of residual influence.
Path & Pass is an independent learning site and is not affiliated with any certification body. Questions are original practice content based on published objectives.
