3.4Standard
An internal-document answer is fluent but sometimes wrong.
Measure unsupported claims.
Choose two evaluations.
× Incorrect
Break answers into claims and check evidence, omissions, and contradictions, including ambiguous, unsupported, and updated-document cases.
Detailed explanation
CorrectInclude ambiguous questions, missing evidence, opposing evidence, and updated documents
Failure-prone conditions are covered.
CorrectDecompose answers into claims and compare each with source evidence
Factuality is measured at claim level.
IncorrectIgnore severe examples because the average is high
High-impact individual failures can be hidden.
IncorrectAccept fluent text as factual
Fluency does not prove truth.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 3.4の生成品質、グラウンディング、幻覚評価を確認する。
Official references
What to confirm
流暢さと事実性を分け、根拠のない主張をケース別に評価できる。
Key points
- Hallucination
- Factuality
- Claims
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
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Hallucination
A plausible-sounding generated statement that is unsupported or factually wrong.
Factuality
The degree to which output statements agree with reliable facts or evidence.
Question IDAWS-AIF-342
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue