AIF-C01 Applications of Foundation Models Practice Question

Exhibit

Refer to the exhibit.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": "bedrock:InvokeModel",
            "Resource": "arn:aws:bedrock:us-east-1:123456789012:model/amazon.titan-text-express-v1"
        }
    ]
}

A security engineer creates the above IAM policy to allow a user to invoke an Amazon Bedrock model. However, invocation fails. What is the issue?

⚠ Common exam trap

AWS often tests the distinction between 'model' and 'foundation-model' in Bedrock ARNs, as candidates may assume all Bedrock models use the same resource type, overlooking that foundation models require a specific path.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

The ARN should use "foundation-model" instead of "model".

The IAM policy's resource ARN incorrectly uses 'model' in the path, but Amazon Bedrock requires 'foundation-model' to reference foundation models. The correct ARN format for invoking a Bedrock foundation model is 'arn:aws:bedrock:region::foundation-model/model-id'. Using 'model' instead of 'foundation-model' causes the policy to not match any valid Bedrock resource, resulting in an invocation failure.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
The action should be "bedrock:InvokeModelWithResponseStream".
Why it's wrong here
InvokeModelWithResponseStream is a separate action for streaming responses; a policy granting it does not authorise standard InvokeModel calls. The failure stems from the resource ARN or principal configuration, not the action name. Streaming is correct only when the application requires incremental token delivery.
✗
The resource ARN is missing the account ID.
Why it's wrong here
Bedrock model resource ARNs use the format foundation-model/ or inference-profile/ without an account ID segment, so adding one would make the ARN invalid. Account IDs appear in ARNs for account-scoped resources. The actual defect lies in the action or resource pattern, not the account identifier.
✓
The ARN should use "foundation-model" instead of "model".
Why this is correct
Bedrock foundation models are addressed with the resource ARN segment "foundation-model", not "model". The policy's incorrect ARN means the Allow statement matches no real resource, so the invocation is implicitly denied despite otherwise valid permissions.
✗
The statement is missing a condition for the model ID.
Why it's wrong here
A condition restricting the model ID is optional; omitting it broadens permissions rather than denying invocation, so it cannot cause the failure. Conditions are correct when you must limit access to specific models or regions, but their absence does not block an otherwise valid allow statement.

Go deeper

AIF-C01 Applications of Foundation Models Practice Question

Exhibit

Why each option matters

Option-by-option breakdown

Related to this question