AIF-C01 Question #146: Real Exam Question with Answer & Explanation

Question

An AI practitioner is developing a prompt for an Amazon Titan model. The model is hosted on Amazon Bedrock. The AI practitioner is using the model to solve numerical reasoning challenges. The AI practitioner adds the following phrase to the end of the prompt: "Ask the model to show its work by explaining its reasoning step by step." Which prompt engineering technique is the AI practitioner using?

Options

• AChain-of-thought prompting
• BPrompt injection
• CFew-shot prompting
• DPrompt templating

Explanation

Chain-of-thought prompting (A) is correct because the practitioner is explicitly instructing the model to show its reasoning step by step, which is the defining characteristic of this technique - it encourages the model to break down complex problems (especially numerical reasoning) into intermediate logical steps before arriving at a final answer.

Option B (Prompt injection) is wrong because that refers to a security attack where malicious instructions are embedded in a prompt to manipulate or hijack the model's behavior - it has nothing to do with improving reasoning quality.

Option C (Few-shot prompting) is incorrect because that technique involves providing example input-output pairs within the prompt to guide the model, whereas this prompt contains no examples at all.

Option D (Prompt templating) is wrong because that refers to using reusable, structured prompt formats with placeholders or variables - not a technique for guiding reasoning style.

Memory tip: Think of "chain-of-thought" like showing your work on a math test - you're creating a chain of logical thoughts that connect the problem to the answer step by step. Whenever you see phrases like "explain your reasoning" or "think step by step," that's your cue for chain-of-thought prompting.

No community discussion yet for this question.