AI-900 Practice Test 28 — 15 Questions

Question 1

What is 'structured output' (JSON mode) in Azure OpenAI?

Accepted Answer

Constraining model responses to valid JSON conforming to a specified schema for application integration. Structured output (JSON mode) in Azure OpenAI constrains the model to generate responses that are valid JSON objects conforming to a user-defined schema. This is achieved by setting the `response_format` parameter to `{ "type": "json_object" }` and optionally providing a JSON schema via the `json_schema` parameter, ensuring the output can be directly parsed and integrated into applications without additional formatting logic.

Answer

Formatting the model's text response with numbered sections and bullet points

Answer

Saving model responses to a structured database table automatically

Answer

Generating output in multiple languages simultaneously in a structured format

Question 2

What is 'overfitting' in machine learning and how does Azure ML help prevent it?

Accepted Answer

When a model learns training data too specifically and fails to generalise to new data. Overfitting occurs when a machine learning model learns the training data too precisely, including noise and outliers, resulting in poor performance on unseen data. Azure ML helps prevent overfitting through automated machine learning (AutoML) which applies regularization, cross-validation, and early stopping techniques, as well as by enabling easy configuration of train/test splits and hyperparameter tuning.

Answer

When a model is trained on too much data and becomes too accurate

Answer

When a model's predictions exceed the acceptable numerical range

Answer

When Azure ML runs training for longer than the allocated compute budget

Question 3

What industries benefit most from Azure AI Document Intelligence's capabilities?

Accepted Answer

Finance, healthcare, legal, government, and any industry processing high volumes of documents. Azure AI Document Intelligence (formerly Form Recognizer) is designed to extract, analyze, and structure data from documents at scale using prebuilt and custom models. Industries like finance, healthcare, legal, and government process massive volumes of forms, invoices, medical records, and contracts, making them the primary beneficiaries of automated document processing.

Answer

Only the entertainment industry for processing movie scripts

Answer

Only manufacturing for quality control inspection

Answer

Only retail for product catalog management

Question 4

What is Azure AI Content Safety used for in computer vision scenarios?

Accepted Answer

Detecting harmful or inappropriate content in images for content moderation. Azure AI Content Safety is designed to detect harmful or inappropriate content in images, such as violence, hate speech, self-harm, or sexually explicit material. In computer vision scenarios, it analyzes visual features to classify content into severity levels, enabling automated content moderation. This directly supports safe user-generated content platforms by flagging or blocking prohibited imagery.

Answer

Compressing images to reduce storage costs

Answer

Enhancing image quality and resolution

Answer

Converting images to text descriptions for accessibility

Question 5

A company uses Azure OpenAI to generate marketing copy. They want to ensure that the generated text does not contain inappropriate or harmful content before it is published. Which Azure OpenAI feature is specifically designed for this purpose?

Accepted Answer

Content filters. Content filters are the Azure OpenAI feature specifically designed to detect and block inappropriate or harmful content in generated text. They apply configurable severity levels across categories like hate, violence, self-harm, and sexual content, ensuring outputs meet safety policies before publication.

Answer

Temperature

Answer

Top-p (nucleus sampling)

Answer

System message

Question 6

What is a neural network?

Accepted Answer

A machine learning model architecture with layers of interconnected nodes that learn representations. A neural network is a machine learning model architecture composed of layers of interconnected nodes (neurons) that process input data through weighted connections and activation functions. These layers learn hierarchical representations of data, enabling the model to capture complex patterns and relationships without explicit rule-based programming. This aligns with option B as the correct definition.

Answer

A computer network for distributed AI training across multiple servers

Answer

A database for storing trained ML models

Answer

A rule-based expert system for decision making

Question 7

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 8

A financial institution uses an AI system to recommend credit limits for new customers. When a customer is declined for a credit limit increase, the customer asks why, but the institution cannot provide any explanation because the model is a complex deep neural network and the decision-making process is opaque. Which Microsoft responsible AI principle is most directly violated?

Accepted Answer

Transparency. Transparency. The scenario describes a deep neural network that cannot explain its decision to deny a credit limit increase, which directly violates the transparency principle. Microsoft's responsible AI principle of transparency requires that AI systems be understandable and that their decisions can be explained to users, especially when those decisions have significant impact on individuals.

Answer

Reliability and safety

Answer

Fairness

Answer

Privacy and security

Question 9

What types of documents does Azure AI Document Intelligence's prebuilt 'receipt' model extract data from?

Accepted Answer

Sales receipts from stores and restaurants, extracting merchant details, items, and totals. Azure AI Document Intelligence's prebuilt 'receipt' model is specifically designed to extract key information from sales receipts, such as merchant details, transaction items, and totals. It uses optical character recognition (OCR) and deep learning models to parse both printed and handwritten receipts from stores and restaurants, handling various formats and layouts.

Answer

Only digital PDF receipts with standardized formatting

Answer

Medical receipts and prescription records only

Answer

Electronic bank transfer receipts for financial transactions

Question 10

A company needs to automatically extract text from scanned invoices that contain both printed text and handwritten notes. Which Azure AI service is specifically designed to handle this type of document?

Accepted Answer

Azure AI Document Intelligence (formerly Form Recognizer). Azure AI Document Intelligence (formerly Form Recognizer) is specifically designed to extract text, key-value pairs, and tables from scanned documents, including invoices with both printed text and handwritten notes. It uses optical character recognition (OCR) combined with deep learning models to handle mixed content, making it the correct choice for this scenario.

Answer

Azure Face API

Answer

Azure Custom Vision

Answer

Azure Video Indexer

Question 11

What is 'real-time speech translation' in Azure AI Speech?

Accepted Answer

Converting spoken words in one language to text or speech in another language instantly. Real-time speech translation in Azure AI Speech is designed to translate spoken language into another language with minimal latency, enabling live conversations. Option B correctly describes this capability, as it converts spoken words in one language to text or speech in another language instantly, leveraging the Speech Translation API with streaming audio input.

Answer

Translating pre-recorded audio files overnight in a batch processing job

Answer

Generating subtitles for pre-existing videos stored in Azure Media Services

Answer

Converting text written in one language into spoken audio in the same language

Question 12

What is 'pronunciation assessment' in Azure AI Speech?

Accepted Answer

Scoring a speaker's accuracy, fluency, and completeness against native pronunciation norms. Pronunciation assessment in Azure AI Speech evaluates a speaker's spoken audio against native pronunciation norms, providing scores for accuracy, fluency, and completeness. This is a feature of the Speech-to-Text API that uses a reference script and phoneme-level comparison to generate detailed feedback, making it ideal for language learning and accent reduction applications.

Answer

Checking whether a text-to-speech voice pronounces technical terms correctly

Answer

Detecting regional accents to route calls to the appropriate customer service team

Answer

Generating a list of commonly mispronounced words in a specific language

Question 13

What is 'max_tokens' parameter in Azure OpenAI and how does it affect responses?

Accepted Answer

A limit on the model's generated response length — stopping output at the specified token count. The 'max_tokens' parameter in Azure OpenAI sets a hard limit on the number of tokens (words or subwords) the model can generate in its response. Once this token count is reached, the model stops producing further output, effectively controlling response length. This is distinct from input processing limits, as 'max_tokens' applies solely to the generated completion.

Answer

The maximum number of tokens in the input prompt the model can process

Answer

The total number of API calls allowed per Azure subscription per hour

Answer

The maximum number of conversation turns before the session resets

Question 14

What is 'AI transparency' in Microsoft's Responsible AI principles?

Accepted Answer

Ensuring people understand when they're interacting with AI, how it works, and what its limitations are. AI transparency, as defined in Microsoft's Responsible AI principles, is about ensuring that users understand when they are interacting with an AI system, how the system makes decisions, and what its limitations are. This principle focuses on clear communication and documentation, not on open-sourcing code or financial reporting.

Answer

Making AI model source code publicly available as open source

Answer

Reporting all AI project costs transparently in financial statements

Answer

Making all training data publicly available for independent researchers to audit

Question 15

What is 'named entity recognition' (NER) in Azure AI Language?

Accepted Answer

Identifying and classifying real-world entities (people, organisations, locations) mentioned in text. Named entity recognition (NER) is a feature of Azure AI Language that identifies and categorizes real-world entities such as people, organizations, locations, dates, and quantities within unstructured text. It uses pre-trained machine learning models to extract these entities, enabling downstream tasks like information retrieval and content summarization. Option B correctly describes this core functionality.

Answer

Renaming database fields to follow a consistent naming convention

Answer

Recognising the named author of a document for copyright purposes

Answer

Detecting when a user provides their name in a chatbot conversation