CCAO-F Claude Certified Associate – Foundations

A senior manager is preparing an AI-assisted work product for external delivery and applying the Diligence competency. Which action best reflects this principle?

• A. Deliver without independent verification because AI assistance shifts accountability away from the human.
• B. Assign accountability to Claude instead of the responsible professional.
• C. Assume fluent output is accurate and skip verification.
• D. Identify which elements need independent checking or correction, complete those checks, and take ownership of the final work.

Which two statements accurately describe effective prompt iteration? (Choose two.)

• A. Repeating the same prompt is the most reliable improvement method.
• B. Compare the output with the original criteria to check whether the iteration helped.
• C. Make each revision target a specific diagnosed weakness.
• D. Continue until every possible phrasing has been tried.
• E. More iterations always improve an output.

A Claude-generated customer-segment description makes a broad claim about a demographic group’s buying behavior that does not apply to individuals. What is the most appropriate response?

• A. Replace the generalization with observed behavior and validate the revised description against actual customer data.
• B. Remove every description involving demographic characteristics.
• C. Keep the statement because the brief is for internal use.
• D. Add a disclaimer but leave the unsupported generalization unchanged.

Put these recurring-workflow optimization steps in the correct order:

1. Apply a targeted improvement to the bottleneck.
2. Roll out the optimized workflow and document the change.
3. Use time per step and queue length to identify the bottleneck.
4. Map the current steps, owners and decision points.
5. Pilot the change and compare its effect with the baseline.

• A. 4, 1, 3, 5, 2
• B. 4, 3, 1, 5, 2
• C. 3, 4, 1, 5, 2
• D. 1, 4, 3, 5, 2

A customer-success workflow needs fast turn-by-turn responses with light reasoning at scale. Which model selection best fits?

• A. Any model, because latency is the same for every model.
• B. Haiku, for low latency and efficient handling of light reasoning tasks.
• C. Opus, to maximize reasoning depth despite higher latency.
• D. Select alphabetically by model name.