Personal data must not enter a knowledge base. Which service can detect it during ingestion?
Why C is the answer
Amazon Comprehend
Amazon Comprehend can detect personal data in text, so it can catch it during ingestion, before indexing.
In detail
- Comprehend can detect personal data in text.
- Macie can also find personal data in S3.
Use case
An illustrative scenario, not a real incident.
A pipeline runs new documents through Comprehend, finds names and account numbers, and removes them before indexing.
See it in different ways
flowchart LR A["App in a private subnet"] --> B["VPC endpoint"] B --> C["Amazon Bedrock"] D["IAM policy"] --> C E["KMS key"] --> F["Data at rest"]
mindmap
root((RAG))
Ingest
Sync data source
Detect personal data
Chunk
By structure
Hierarchical
Search
Vector
Keyword
Rank
Rerankerflowchart LR S0["New document"] --> S1["Comprehend detects PII"] --> S2["Redact"] --> S3["Add to knowledge base"]
Key point: Detect personal data during ingestion.
Why the other options are wrong
- A. AWS Step Functions alone, with no detection service in the workflow
- Step Functions needs a detection service.
- B. Amazon Translate
- Translate changes language.
- D. Amazon Textract
- Textract reads documents, and does not detect PII by itself.