Exams / AIF-C01 / Evaluating foundation models / Question 3
AIF-C01 practice question: Evaluating foundation models (3 of 3)
Applications of Foundation Models A free question from our original AIF-C01 practice exam. Try it, then open the answer to see why each option is right or wrong.
An online retailer launched a generative AI shopping assistant. The board asks whether it is delivering business value. Which measure BEST answers that question?
Choose one answer.
- AA. The number of parameters in the underlying model
- BB. The temperature setting used for the assistant
- CC. The change in conversion rate among assistant users
- DD. The number of tokens in the system prompt
Show the answer and explanation
Answer: C
- A. Wrong: Model size is a technical property. It says nothing about whether the business gained anything.
- B. Wrong: Temperature is a configuration choice. It is not a measure of business outcome.
- C. Right: Correct. Conversion rate ties the assistant directly to revenue, which is a business metric.
- D. Wrong: Prompt length affects cost and latency, but it does not show value to the business.
Explanation
Technical metrics such as ROUGE or latency show how well a model performs, but business metrics show whether it is worth it. For generative AI these include conversion rate, average revenue per user, customer lifetime value, efficiency gains, cost savings and return on investment.
Study
Keep practising
Answer questions with a score at the end, or practise just this topic with the explanation after each one. No sign-up needed.