Topic lessonTopic: Applied container, data, cost and AI labs
Exercise Bedrock input classification and log access
Inference processing and configurable invocation logging have separate scope.
Use synthetic input to evaluate collection and reader scope without sensitive values.
The cases use fictional inputs.
What to predict
Inference processing and configurable invocation logging have separate scope.
- Input classification → Invocation loggingAlign conditions
- Invocation logging → Destination readersCompare outcomes
- Destination readers → DecisionVerify the decision
- Input classification
- Synthetic / Sensitive input proposed
- Invocation logging
- Enabled for selected path / Off
- Destination readers
- Scoped reviewers / Unneeded readers included
- Decision
- Use synthetic input to evaluate collection and reader scope without sensitive values.
Compare the cases
| Case | Conditions | Decision |
|---|---|---|
| Trial 1 | Input classification Synthetic, Invocation logging Enabled for selected path, Destination readers Scoped reviewers | Modeled prerequisites pass. Verify fields and retention with a synthetic record. |
| Trial 2 | Input classification Sensitive input proposed, Invocation logging Enabled for selected path, Destination readers Scoped reviewers | Hold the input. Prepare a synthetic substitute without confidential values. |
| Trial 3 | Input classification Synthetic, Invocation logging Off, Destination readers Scoped reviewers | Invocation logs cannot provide evidence in this setup. Check approved alternatives. |
| Trial 4 | Input classification Synthetic, Invocation logging Enabled for selected path, Destination readers Unneeded readers included | Review reader scope, then test intended access and out-of-scope denial. |
Acceptance and limits
Enter only a classification. The exercise accepts no prompts and leaves endpoint-specific coverage to the official guide.
Read the initial conditions and operational goal.
Change one condition and compare the decision.
Record the model boundary and outstanding checks.
Change a condition
Enter only a classification. The exercise accepts no prompts and leaves endpoint-specific coverage to the official guide.
Modeled prerequisites pass. Verify fields and retention with a synthetic record.
{"reject":"Hold the input. Prepare a synthetic substitute without confidential values.","off":"Invocation logs cannot provide evidence in this setup. Check approved alternatives.","restrict":"Review reader scope, then test intended access and out-of-scope denial.","checked":"Modeled prerequisites pass. Verify fields and retention with a synthetic record."}Check your understanding
Choose an option to see the result and explanation.
Check, then explain in your own words
- Read the initial conditions and operational goal.
- Change one condition and compare the decision.
- Record the model boundary and outstanding checks.
Check your understanding: Use synthetic input to evaluate collection and reader scope without sensitive values.
Enter only a classification. The exercise accepts no prompts and leaves endpoint-specific coverage to the official guide.
Record the model boundary and outstanding checks.
Sources and conditions checked
The official references used in this lesson are collected here with their scope, conditions and corresponding sections. Independent expert review and live AWS comparison have not been performed.
Scope and conditionsInference processing and configurable invocation logging have separate scope.
Used in this lessonWhat to predict Acceptance and limits