AI-900 exam dumps

AI-900 practice question 54 of 286

Microsoft Azure AI Fundamentals. Free level, Microsoft. Free question with the correct answer and a full explanation.

AI-900 Question 54

Select 3

A financial institution is deploying an AI-powered fraud detection system. To ensure reliability and safety, which considerations should they prioritize in their solution?

  1. A

    Implementing bias detection and mitigation strategies in the AI model

  2. B

    Ensuring the model provides explanations for its predictions

  3. C

    Maximizing the speed of predictions at the expense of model accuracy

  4. D

    Establishing monitoring systems for model performance over time

  5. E

    Disabling user feedback mechanisms to avoid unnecessary complexity

Show answer and explanation

Correct answers: A, B, D

Explanation

When developing AI solutions, particularly in critical areas like fraud detection, considerations for reliability and safety include addressing bias, ensuring transparency, and monitoring performance over time. These steps help reduce risks and improve the effectiveness of the system, fostering trust and accountability. Speed should not come at the cost of accuracy, and user feedback mechanisms are an essential component of maintaining and improving AI reliability.

  • A. Correct.

    Bias detection and mitigation are crucial for ensuring the AI system is reliable, fair, and safe, especially in sensitive domains like financial fraud detection.

  • B. Correct.

    Ensuring the model provides explanations for its predictions builds trust and transparency, which are essential for safety and accountability in AI systems.

  • C. Incorrect.

    While speed is important, sacrificing accuracy for speed can lead to unreliable predictions, which would compromise the safety and reliability of the fraud detection system.

  • D. Correct.

    Monitoring systems allow the institution to track the model's performance and address potential issues like data drift or reduced accuracy over time, ensuring the solution remains reliable.

  • E. Incorrect.

    Disabling user feedback mechanisms removes an important avenue for improving the model and addressing potential errors, which could harm reliability and safety.

Timed practice exam

Take a AI-900 practice test under exam conditions

60 questions in 30 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam