AI-900 Practice Exam 58 — 20 Questions

Question 1

A healthcare organization plans to use AI to analyze patient records for medical research. They must ensure that patient data is protected from unauthorized access during storage and processing. Which Microsoft responsible AI principle is most directly relevant to this requirement?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Privacy and security. The requirement to protect patient data from unauthorized access during storage and processing directly aligns with the Privacy and security principle. This principle mandates that AI systems implement robust data protection mechanisms, such as encryption at rest (e.g., AES-256) and in transit (e.g., TLS 1.2/1.3), access controls (e.g., Azure RBAC), and compliance with regulations like HIPAA. It ensures that sensitive healthcare data remains confidential and secure throughout its lifecycle.

Answer

Fairness

Answer

Inclusiveness

Answer

Reliability and safety

Question 2

What is 'extractive vs abstractive summarisation' and which does Azure AI Language's document summarisation feature support?

Accepted Answer

Azure AI Language supports both extractive (key sentences) and abstractive (generated synthesis) summarisation. Azure AI Language's document summarization feature supports both extractive summarization (selecting key sentences from the original text) and abstractive summarization (generating a new, condensed summary that rephrases the content). Option B is correct because the service provides both capabilities, allowing users to choose based on their needs.

Answer

Azure AI Language only supports extractive summarisation — abstractive requires Azure OpenAI

Answer

Azure AI Language only supports abstractive summarisation because it is more advanced

Answer

Extractive is for short texts; abstractive is required for documents longer than 10,000 words

Question 3

What is 'reinforcement learning from human feedback' (RLHF) and how is it used in training AI models?

Accepted Answer

Using human preference ratings to train a reward model that guides language model optimisation. Reinforcement learning from human feedback (RLHF) is a technique where human evaluators rank or rate model outputs (e.g., text completions) to create a reward model. This reward model then guides the fine-tuning of a language model using reinforcement learning, typically with Proximal Policy Optimization (PPO), to align outputs with human preferences.

Answer

A technique where humans physically assist robots in learning physical tasks

Answer

Having human reviewers manually rewrite model outputs to improve training data

Answer

Allowing end users to flag incorrect answers to automatically retrain the model in real time

Question 4

A bank deploys an AI system to automatically approve or reject loan applications. After six months, an audit reveals that the system approves loans at a significantly lower rate for applicants from a specific ethnic group compared to other groups with similar financial profiles. Which Microsoft responsible AI principle is most directly violated by this outcome?

Accepted Answer

Fairness. The AI system's approval rate disparity for a specific ethnic group, despite similar financial profiles, directly violates the Fairness principle. Fairness requires that AI systems treat all groups equitably and avoid discrimination based on sensitive attributes like ethnicity. This outcome demonstrates a lack of fairness in the model's decision-making process.

Answer

Transparency

Answer

Privacy

Answer

Reliability

Question 5

A data scientist trains a classification model to predict whether an email is spam or not. The model achieves 98% accuracy on the test set, but upon inspection, it classifies all emails as 'not spam' because the dataset has 95% non-spam emails. What is the most likely issue?

Accepted Answer

Data imbalance. The model achieves 98% accuracy by simply predicting all emails as 'not spam', which reflects the 95% majority class in the dataset. This is a classic symptom of class imbalance, where the model learns to exploit the skewed distribution rather than learning meaningful patterns to distinguish spam from non-spam. In Azure Machine Learning, techniques like SMOTE or stratified sampling are used to mitigate this issue.

Answer

Overfitting

Answer

Underfitting

Answer

Feature scaling error

Question 6

What is 'retrieval augmented generation' (RAG) and which Azure services typically implement it?

Accepted Answer

Combining Azure AI Search (retrieval) with Azure OpenAI (generation) to ground LLM responses in a knowledge base. Retrieval Augmented Generation (RAG) is a pattern that combines a retrieval step with a generative step. In Azure, this is typically implemented by using Azure AI Search to retrieve relevant documents or chunks from a knowledge base, then passing those results as context to an Azure OpenAI model (e.g., GPT-4) to generate a grounded, fact-based response. This approach reduces hallucinations and ensures the output is based on authoritative data rather than the model's training data alone.

Answer

Using Azure Storage to retrieve training data for model fine-tuning

Answer

Using Azure CDN to deliver AI-generated content faster globally

Answer

A method of compressing large datasets before training language models

Question 7

A hotel chain wants to automatically analyze guest reviews to identify the most frequently mentioned aspects of their stay, such as 'cleanliness', 'staff friendliness', or 'location'. They want to use a prebuilt Azure AI Language feature without custom training. Which feature should they use?

Accepted Answer

Key Phrase Extraction. Key Phrase Extraction is the correct choice because it is a prebuilt Azure AI Language feature designed to automatically identify and return the most salient words or phrases from unstructured text, such as 'cleanliness', 'staff friendliness', or 'location'. This feature requires no custom training and directly addresses the hotel chain's need to surface frequently mentioned aspects of guest reviews.

Answer

Sentiment Analysis

Answer

Entity Recognition

Answer

Language Detection

Question 8

What is the Responsible AI principle most relevant to Azure AI Face's attribute prediction features?

Accepted Answer

Fairness and privacy — preventing bias across demographic groups and avoiding surveillance misuse. Azure AI Face's attribute prediction features (e.g., age, emotion, hair color) have been restricted or retired due to concerns about demographic bias and potential misuse for surveillance. The Responsible AI principle of Fairness and privacy directly addresses these issues by requiring that AI systems avoid bias across demographic groups and prevent applications like unauthorized tracking or profiling, which is why this principle is most relevant.

Answer

Reliability — ensuring Face API returns consistent results across all images

Answer

Inclusiveness — ensuring Face API works for users of all abilities

Answer

Transparency — documenting how Face API determines attribute values

Question 9

A data scientist has trained a binary classification model to predict whether an email is spam (positive) or not spam (negative). On a test set, the model correctly identifies 90 out of 100 actual spam emails and 80 out of 100 actual non-spam emails. Which metric shows the proportion of actual spam emails that the model correctly predicted?

Accepted Answer

B. Recall. Recall (also known as sensitivity or true positive rate) measures the proportion of actual positive cases that were correctly predicted by the model. In this scenario, the model correctly identified 90 out of 100 actual spam emails, so the recall is 90/100 = 0.9 (90%). This metric directly answers the question about how well the model captures actual spam emails.

Answer

A. Precision

Answer

C. F1 Score

Answer

D. Accuracy

Question 10

What does Azure AI Speech service's text-to-speech (TTS) feature do?

Accepted Answer

Converts written text into natural-sounding spoken audio. Azure AI Speech service's text-to-speech (TTS) feature converts written text into natural-sounding spoken audio using neural voice models. It synthesizes speech from input text, enabling applications like voice assistants, audiobooks, and accessibility tools. This is the core function of TTS, distinct from speech-to-text or translation capabilities.

Answer

Converts spoken audio into written text

Answer

Identifies the language of spoken audio

Answer

Translates spoken words from one language to another

Question 11

What is 'cross-lingual transfer learning' in multilingual NLP models?

Accepted Answer

Using shared multilingual representations so knowledge learned in one language transfers to others. Cross-lingual transfer learning leverages shared multilingual representations (e.g., from models like multilingual BERT or XLM-R) that encode multiple languages into a common semantic space. This allows knowledge learned from training data in one language (e.g., English) to improve performance on tasks in other languages without requiring labeled data for each target language. The model transfers understanding of syntax, semantics, and context across languages because it was pre-trained on a diverse corpus of many languages simultaneously.

Answer

Automatically translating training data from English to other languages before fine-tuning

Answer

Using the same model for both NLP and computer vision tasks

Answer

Transferring a model trained in Azure to run on another cloud provider

Question 12

A company implements an AI system to monitor employee productivity by tracking keystrokes and mouse movements. Employees are not informed that this monitoring is taking place, nor did they consent to it. Which Microsoft responsible AI principle is most directly violated?

Accepted Answer

Privacy & Security. The scenario describes monitoring employee keystrokes and mouse movements without their knowledge or consent. This directly violates the Privacy & Security principle, which requires that individuals have control over their personal data and that data collection is transparent and consensual. Microsoft's responsible AI framework mandates that AI systems must respect privacy and obtain informed consent before collecting or using personal data.

Answer

Fairness

Answer

Reliability & Safety

Answer

Inclusiveness

Question 13

What is a 'bot' in Microsoft Azure Bot Service?

Accepted Answer

A conversational AI application that interacts with users through natural language in messaging channels. A 'bot' in Microsoft Azure Bot Service is a conversational AI application designed to interact with users through natural language across various messaging channels like Microsoft Teams, Slack, or Facebook Messenger. It leverages the Bot Framework SDK and Azure AI services (e.g., LUIS, QnA Maker) to understand user intent and provide responses, enabling human-like dialogue without requiring a custom interface for each channel.

Answer

An automated script that performs repetitive tasks in the Azure portal

Answer

A security tool that detects automated bot traffic attempting to misuse an API

Answer

An autonomous AI agent that independently manages Azure cloud infrastructure

Question 14

A company wants to implement an AI solution that treats all users fairly regardless of their background. Which Microsoft responsible AI principle does this requirement primarily address?

Accepted Answer

Fairness. The requirement to treat all users fairly regardless of background directly aligns with the Fairness principle, which mandates that AI systems should allocate outcomes, opportunities, or resources equitably and avoid discrimination based on sensitive attributes such as race, gender, or age. In Azure AI, this is operationalized through tools like Fairlearn and the Responsible AI dashboard, which assess and mitigate bias in model predictions. The other principles address different concerns: Privacy focuses on data protection, Inclusiveness on accessibility for diverse abilities, and Transparency on explainability of decisions.

Answer

Privacy

Answer

Inclusiveness

Answer

Transparency

Question 15

What is 'model confidence score' in Azure Custom Vision predictions?

Accepted Answer

A per-prediction certainty measure indicating how sure the model is about a specific classification. In Azure Custom Vision, the model confidence score is a per-prediction value (ranging from 0 to 1) that quantifies the model's certainty that a given input image belongs to a specific class. It is computed during inference based on the probability distribution output by the trained classifier, not during training. This score helps users decide whether to accept or reject a prediction based on a custom threshold.

Answer

The percentage of training images the model correctly labelled during training

Answer

A rating of the training data quality provided by the annotation team

Answer

Microsoft's certification level for how well a Custom Vision model meets enterprise standards

Question 16

A data scientist is training a regression model to predict energy consumption. The dataset includes features like temperature, humidity, time of day, and day of week. After training, the model performs well on the training set but poorly on new data. Which approach would most likely help reduce this problem?

Accepted Answer

Use a simpler model with fewer parameters.. The model performs well on the training set but poorly on new data, which is classic overfitting. Using a simpler model with fewer parameters reduces the model's capacity to memorize noise and irrelevant patterns, forcing it to learn the underlying generalizable relationships. This directly addresses the variance problem without requiring additional data or computational resources.

Answer

Add more features to the model.

Answer

Increase the number of training epochs.

Answer

Use a more complex model to capture more patterns.

Question 17

A legal firm wants to use Azure OpenAI to generate summaries of lengthy contracts. The firm requires that the generated summaries are strictly based on the provided contract text and do not include any external knowledge or hallucinated facts. Which Azure OpenAI feature should the firm configure to meet this requirement?

Accepted Answer

Azure OpenAI on Your Data (data grounding). Azure OpenAI on Your Data (data grounding) restricts the model's responses to the content of the provided contract documents, preventing the generation of information not present in the source text. This feature uses a retrieval-augmented generation (RAG) approach, where the model only references the indexed contract data, effectively eliminating external knowledge or hallucinated facts.

Answer

Content filtering

Answer

Prompt engineering with system messages

Answer

Fine-tuning the model on legal texts

Question 18

A company uses Azure OpenAI Service to generate marketing copy for a new product. They have a strict brand voice that requires formal, technical language and explicitly prohibits any humorous or informal phrases. They want to enforce these constraints without retraining the model. Which technique should they use?

Accepted Answer

B) Prompt engineering. Prompt engineering is correct because it allows the user to craft system messages or user prompts that explicitly instruct the model to use formal, technical language and avoid humor, all without modifying the underlying model weights. This technique leverages the model's instruction-following capability to enforce constraints at inference time, making it ideal for brand voice enforcement without retraining.

Answer

A) Fine-tuning

Answer

C) Reinforcement learning

Answer

D) Transfer learning

Question 19

What is 'chain-of-thought prompting' and when is it most effective?

Accepted Answer

Prompting the model to show explicit reasoning steps before giving a final answer. Chain-of-thought prompting instructs the model to break down a complex problem into intermediate reasoning steps before producing the final answer. This technique improves accuracy on tasks requiring multi-step logic, such as arithmetic, commonsense reasoning, or symbolic manipulation, by making the model's internal reasoning explicit and reducing errors from shortcut answers.

Answer

Linking multiple AI models in a pipeline where each model's output feeds the next

Answer

Training a model on a sequence of related documents to build contextual knowledge

Answer

A method for connecting chatbot conversation turns to maintain long-term memory

Question 20

A city government deploys an AI system that automatically detects traffic violations (e.g., running red lights) from traffic camera footage. The system triggers fines without immediate human review. According to Microsoft's responsible AI principles, which principle is most directly concerned with ensuring there is human oversight and that the organization can be held liable for the system's decisions?

Accepted Answer

Accountability. The Accountability principle in Microsoft's responsible AI framework requires that organizations take ownership of AI system outcomes and ensure human oversight. In this scenario, the system automatically issues fines without human review, which directly challenges accountability because there is no mechanism for human intervention or liability assignment. This principle mandates that the organization must be able to answer for the system's decisions, including errors or biases.

Answer

Transparency

Answer

Reliability and Safety

Answer

Privacy and Security