A county deed-vault office is about to index a shared bucket into a knowledge base and is afraid deed scans still hold Social Security numbers. Which service should find those files first?
Select an answer to reveal the explanation.
Short Explanation
Think of deed scans about to enter a knowledge base, still possibly holding Social Security numbers. Amazon Macie finds those files first. Skip discovery and the numbers ride in.
Full Explanation
The county deed-vault office is about to index a shared bucket into a knowledge base and fears deed scans still contain Social Security numbers. Amazon Macie is the service aimed at discovering sensitive data in data stores that feed AI systems. Finding and classifying those files before indexing prevents personal identifiers from entering embeddings, chunks, and retrieval results that staff and the public might later surface.
Amazon Polly fails because speech synthesis does not scan buckets for Social Security numbers or classify deed objects. Amazon QuickSight-style dashboards fail because a director chart is not a discovery and classification scan of object contents. Skipping discovery and indexing immediately fails because unchecked numbers ride into the knowledge pile and become hard to purge from vectors, caches, and prompt logs later.
Exam caveat: discovery finds candidates; teams still need redaction, exclusion, or tokenization workflows before re-indexing—Macie alone does not sanitize the corpus or rebuild the index. Operational check: run Macie (or equivalent sensitive-data discovery) on the shared bucket, quarantine findings, remove or redact SSN-bearing objects from the AI ingest path, and only then rebuild the knowledge base from the cleaned set.