AB-410 Practice Question: Using an AI Builder prompt to summarize customer feedback
You are using an AI Builder prompt to summarize customer feedback. The prompt is producing output in a format that is not consistent for your downstream database. What should you use to enforce a structure?
⚠ Common exam trap
AB-410 often tests the misconception that lowering temperature or post-processing with regex can enforce output structure, when the correct approach is to specify the desired format (e.g., JSON schema) directly in the prompt instructions.
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
✓
Provide a JSON schema definition within the prompt instructions.
Providing a JSON schema definition within the prompt instructions is the correct approach because it explicitly tells the AI model the exact structure, fields, and data types required for the output. This leverages the model's ability to follow structured formatting instructions, ensuring consistent, machine-readable JSON that can be directly ingested by downstream databases. Unlike temperature adjustments or post-processing, schema-based prompting enforces structure at generation time, reducing parsing errors and manual cleanup.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Configure the 'Temperature' setting to 0.
Why it's wrong here
Temperature 0 makes sampling deterministic, reducing randomness but not guaranteeing a fixed schema; the model can still emit varying field names or prose. It is tempting for consistency, and would be correct when you need repeatable, less creative wording rather than enforced JSON structure.
- ✗
Retrain the model.
Why it's wrong here
AI Builder prompts are not retrained by users; retraining applies to custom models, so it cannot enforce output structure. It is tempting because retraining fixes poor model accuracy, and would be correct when a trained model's predictions are wrong rather than its output format.
- ✓
Provide a JSON schema definition within the prompt instructions.
Why this is correct
Embedding a JSON schema in the prompt instructions constrains the model's decoding to emit only fields and types matching that schema, so every summary arrives in a predictable shape for the downstream database. This directly satisfies the stem's requirement to enforce consistent structure, unlike free-text formatting guidance.
- ✗
Use a regex action after the prompt.
Why it's wrong here
Regex parses or extracts from text after generation; it cannot force the prompt to emit a defined schema and breaks when wording varies. It is tempting for post-processing, and would be correct when the output format is already predictable and you only need to extract specific values.
About these practice questions
Courseiva writes every AB-410 question from scratch — 278 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
You might also like
Similar AB-410 questions
- Which THREE of the following can cause an AI model to fail in production? (Choose THREE)
- You want to automate the extraction of data from invoices using AI Builder. Which model type should you choose?
- What is the primary function of an AI Builder 'Prebuilt' model?
- You are building a Microsoft Copilot Studio agent that must answer questions using an internal SharePoint site. The site…
- You are building a Power Automate cloud flow that processes incoming invoices. The flow uses an AI Builder custom docume…
Written and reviewed by Johnson Ajibi, MSc IT Security
Security Systems Architect · founder of Courseiva
Last reviewed September 2026 · checked against the official Microsoft exam blueprint
This AB-410 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AB-410 exam.