AI-900 Practice Question: Describe Artificial Intelligence workloads and considerations
Which of the following is an example of 'anomaly detection' as an AI workload?
⚠ Common exam trap
Watch out — candidates often confuse anomaly detection with classification (Option D) because both involve identifying unusual items, but classification requires labeled training data for known categories, whereas anomaly detection focuses on deviations from a learned norm without predefined labels for anomalies.
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
✓
Automatically identifying fraudulent credit card transactions that deviate from a customer's normal patterns
Anomaly detection identifies data points that deviate significantly from the norm. In this case, fraudulent credit card transactions are detected because they do not match the customer's typical spending patterns, which is a classic use case for anomaly detection in AI workloads.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Translating customer support emails from Spanish to English
Why it's wrong here
Translating customer support emails from Spanish to English is a machine translation task, a sequence-to-sequence NLP workload that maps text between languages while preserving meaning. The model's objective is to generate fluent, semantically equivalent target-language output, not to detect atypical data points or unusual patterns. Hence this is a Translator/Generative AI scenario, not anomaly detection.
- ✓
Automatically identifying fraudulent credit card transactions that deviate from a customer's normal patterns
Why this is correct
Anomaly detection models learn a statistical baseline of a customer's typical spending behaviour — such as amounts, merchant categories, times, and locations — and flag transactions that are highly improbable under that baseline. Because fraud patterns are often novel and unlabelled, this is a classic unsupervised anomaly-detection use case, not a supervised classification task. That is why identifying credit card transactions that deviate from a customer's normal patterns is the correct answer.
- ✗
Generating product descriptions from a list of specifications
Why it's wrong here
Generating product descriptions from a specification list is a natural language generation (NLG) workload, where the model consumes structured attributes (color, size, material, etc.) and produces original, coherent text. It does not model a baseline of normal data and then flag outliers; instead, it conditions on given inputs to create new content. Therefore this option tests generative AI, not anomaly detection.
- ✗
Classifying customer reviews as positive or negative
Why it's wrong here
Classifying customer reviews as positive or negative is a supervised text classification task in natural language processing: the model learns patterns from labelled examples and assigns each review to a discrete sentiment category. The goal is category assignment, not the identification of statistical outliers or deviations from a learned normal profile. Such a task falls under Azure Text Analytics, not the Anomaly Detector service.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI-900 question from scratch — 985 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 →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 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 AI-900 exam.