MICROSOFT · AB-731

AI Transformation Leader practice questions

Leaders planning AI adoption

Pass mark700 / 1000
Practice questions here40

What the AB-731 exam covers

Skills as of 22 Jul 2026
Identify the business value of generative AI solutions35–40%
Benefits and opportunities of Microsoft’s AI apps and services35–40%
Implementation and adoption strategy20–25%

Sample AB-731 questions

Five of the 40 practice questions, each with its answer and explanation.

Identify the business value of generative AI solutions

What is the main driver of cost in usage-based generative AI services?

  1. The number of tokens processed in prompts and responses
  2. The number of employees in the company
  3. The colour scheme of the application
  4. The number of meetings held
Show the answer

A. The number of tokens processed in prompts and responses

Consumption pricing is based on tokens, so both prompt length and response length affect cost.

Identify the business value of generative AI solutions

An assistant must answer from the company's current product catalogue without retraining a model. Which approach fits?

  1. Retrieval-augmented generation
  2. Training a new model from scratch
  3. Raising the temperature
  4. Removing the system prompt
Show the answer

A. Retrieval-augmented generation

Retrieval-augmented generation grounds answers in up-to-date company data at the time of the question.

Identify the business value of generative AI solutions

A model states a plausible but false fact. What is this called, and what reduces it?

  1. Bias, reduced by longer prompts
  2. Fabrication, reduced by grounding in trusted data and human review
  3. Latency, reduced by caching
  4. Drift, reduced by more licences
Show the answer

B. Fabrication, reduced by grounding in trusted data and human review

Fabrications are confident but incorrect outputs. Grounding and review are the main mitigations.

Identify the business value of generative AI solutions

A retailer wants to forecast next month's demand from five years of structured sales data. Which approach is usually the better fit?

  1. Generative AI text generation
  2. Traditional machine learning forecasting
  3. Image generation
  4. Speech synthesis
Show the answer

B. Traditional machine learning forecasting

Numeric prediction from historical structured data is a classic machine learning task.

Benefits and opportunities of Microsoft’s AI apps and services

Employees need AI help inside Word, Excel, Outlook and Teams, grounded in their own work content. Which product fits?

  1. Microsoft 365 Copilot
  2. Azure Vision
  3. Microsoft Defender
  4. Power BI
Show the answer

A. Microsoft 365 Copilot

Microsoft 365 Copilot works inside the Microsoft 365 apps and grounds answers in the user's work data.