AI Business Value and Boundaries: 10 practice questions

Question 1 of 10

Which value category is demonstrated when a risk score is presented to an investigator who makes the final decision?

1. AAssisting human decision making with predictive evidence
2. BScaling identical manual reviews without a learned pattern
3. CGuaranteeing a prescribed outcome through fixed logic
4. DReplacing the investigator with an exact calculation

Question 2 of 10

What AI/ML value is most directly associated with applying the same learned capability across a large stream of inputs?

1. ASolution scalability across many inputs
2. BA guaranteed answer for every possible input
3. CA fixed rule that does not learn from data
4. DA smaller manual queue with no model output

Question 3 of 10

Which example represents automation by AI/ML?

1. AShowing every transaction to an analyst without a prediction
2. BReturning a contractually fixed response for each status code
3. CCalculating tax from a complete table of statutory rates
4. DRouting a transaction after a model flags suspicious activity

Question 4 of 10

How do decision assistance and automation differ?

1. AAssistance handles scale; automation reduces the number of inputs
2. BAssistance informs a person; automation lets a workflow act on output
3. CAssistance guarantees an answer; automation creates uncertain data
4. DAssistance uses fixed rules; automation cannot use predictions

Question 5 of 10

Which workload characteristic points toward deterministic application logic rather than ML?

1. AInput data changes and model accuracy needs monitoring
2. BHistorical data contains patterns useful for prediction
3. CStable steps completely specify the required result
4. DA forecast helps a planner choose an inventory level

Question 6 of 10

What belongs in a cost-benefit assessment for a proposed ML workload?

1. ATraining-record count compared with the number of deployment Regions
2. BPrediction volume compared with the number of deterministic policy rules
3. CExpected value compared with build, operation, monitoring, and improvement costs
4. DModel feature count compared with the number of available class labels

Question 7 of 10

Why can a technically feasible prediction project still be inappropriate?

1. AIts expected benefit can be lower than its lifecycle cost and risk
2. BA feasible model removes the need for data quality and monitoring
3. CPrediction feasibility means deterministic rules are unavailable in every case
4. DTechnical feasibility proves that its predictions are prescribed outcomes

Question 8 of 10

A payroll system must compute a legally prescribed deduction from a complete rate table. Why is ML not the best core approach?

1. AThe requirement primarily calls for forecasting a future deduction from history
2. BThe requirement primarily calls for discovering unknown groups in payroll records
3. CThe requirement primarily calls for a recommendation that a reviewer may reject
4. DThe requirement calls for an exact prescribed result from deterministic rules

Question 9 of 10

A rare decision yields little value, but its model needs costly data preparation, monitoring, and retraining. Which test rejects the proposal?

1. AThe automation test, because model outputs cannot trigger workflow steps
2. BThe assistance test, because people cannot review predictive evidence
3. CThe scalability test, because every high-volume workload is inexpensive
4. DThe cost-benefit test, because lifecycle burden exceeds expected value

Question 10 of 10

A fraud model scores millions of payments, sends uncertain cases to investigators, and opens review tickets for high-risk scores. Which interpretation is sound?

1. AIt removes scalability because some cases receive human review
2. BIt demonstrates deterministic calculation but no learned prediction
3. CIt combines scalability, decision assistance, and workflow automation
4. DIt provides only a prescribed outcome because investigators review cases