Microsoft Azure AI Engineer Associate AI-102 AI-102 Questions 376–450 | Page 6/11

376

MCQmedium

You are building an agent with Azure AI Agent Service. The agent must consult a product catalog stored in an Azure AI Search index before answering questions, and you want the retrieval to happen automatically on every run without writing any orchestration code. You have already created the index and a project connection to it. Which configuration should you apply to the agent?

A.Enable grounding by uploading the catalog documents to the agent's file storage and referencing them in the system message.

B.Attach the Azure AI Search index as a knowledge tool on the agent and let the service invoke it during runs.

C.Create a function tool that calls the Azure AI Search REST API and register it as a tool on every run.

D.Add the search index endpoint and admin key to the agent's instructions so the model can call it directly.

AnswerB

Knowledge tools are the built-in retrieval mechanism for Azure AI Agent Service. Once the search index is attached as a knowledge tool with a valid project connection, the service decides when to query it and injects the retrieved passages into the model context, so no application-side orchestration code is required for the automatic retrieval behavior described.

Why this answer

Attaching the Azure AI Search index as a knowledge tool is the native way Azure AI Agent Service performs retrieval-augmented generation. The service manages query formulation, execution against the index through the project connection, and injection of results into the model context, satisfying the no-orchestration-code requirement while reusing the existing index.

Exam trap

The trap here is assuming the model can reach an external search endpoint from instructions alone, when retrieval must be exposed as a tool backed by a project connection.

Full explanation →

377

MCQeasy

A developer is using Azure Cognitive Service for Language to perform sentiment analysis on customer reviews. The service returns sentiment labels (positive, negative, neutral) and confidence scores. For a particular review, the service returns 'positive' with a confidence score of 0.55. The developer wants to ensure that only high-confidence results are used. What should the developer do?

A.Use the Text Analytics for Health API instead.

B.Retrain the sentiment analysis model with additional labeled data.

C.Configure a minimum confidence threshold of 0.75 in the application logic.

D.Adjust the input text by removing ambiguous phrases.

AnswerC

A minimum confidence threshold of 0.75 in application logic discards the 0.55 result, ensuring only high-confidence sentiment labels are used. The service itself always returns a label with a score; filtering must therefore happen client-side, which this configuration achieves.

Why this answer

The developer must implement a confidence threshold in the application logic to filter out low-confidence results. The Azure Cognitive Service for Language returns confidence scores between 0 and 1 for each sentiment label, and the developer can set a minimum threshold (e.g., 0.75) to ensure only high-confidence predictions are used. This approach does not require retraining the model or modifying the input text, as the threshold is applied post-inference in the client code.

Exam trap

The trap here is that candidates may assume they can retrain the prebuilt sentiment model (Option B) or use a different API (Option A) to solve the confidence issue, when in fact the correct solution is a simple application-level threshold check.

How to eliminate wrong answers

Option A is wrong because the Text Analytics for Health API is a specialized domain-specific API for extracting medical entities and relationships, not for general sentiment analysis or confidence thresholding. Option B is wrong because the prebuilt sentiment analysis model in Azure Cognitive Service for Language cannot be retrained with custom labeled data; custom model training is only available for custom text classification or custom named entity recognition, not for the built-in sentiment analysis. Option D is wrong because removing ambiguous phrases from input text does not guarantee higher confidence scores; the model's confidence is a function of its internal weights and the entire input, and manually editing text may introduce bias or reduce the sample size without addressing the underlying confidence threshold requirement.

Full explanation →

378

MCQmedium

You are building a generative AI solution with Azure OpenAI Service. Prompts must be assembled from a system instruction, retrieved document chunks, and the user's question. You need a mechanism that automatically inserts the retrieved chunks into a designated placeholder in a prompt template before the request is sent to the model. What should you use?

A.A content filter configured on the Azure OpenAI deployment

B.A system-assigned managed identity on the Azure OpenAI resource

C.The max_tokens parameter on the chat completions request

D.Prompt flow with a Jinja prompt template node

AnswerD

Jinja templating in prompt flow renders placeholders such as {{context}} and {{question}} by substituting runtime inputs, so retrieved document chunks are injected into the template before the chat call executes. This provides deterministic prompt assembly, supports conditional logic, and keeps the orchestration inside a deployable flow, which matches the requirement to fill a designated placeholder automatically.

Why this answer

Prompt flow's Jinja prompt template node is the supported way to author a reusable template whose placeholders are filled from flow inputs at runtime, which is exactly what injecting retrieved chunks requires. The other choices address safety filtering, authentication, or output length, none of which assemble prompt text. Using a template also keeps prompt logic versioned and testable within the flow.

Exam trap

The trap here is assuming that any Azure OpenAI feature which touches prompts, such as content filtering, performs prompt assembly.

Full explanation →

379

MCQeasy

A developer wants to integrate a pre-built AI model that can extract key information from invoices, such as vendor name, invoice date, and total amount. Which Azure AI service should they use?

A.Azure AI Language

B.Azure OpenAI Service

C.Azure AI Document Intelligence

D.Azure Cognitive Search

AnswerC

Azure AI Document Intelligence provides pre-built invoice models that extract structured fields such as vendor name, invoice date and total amount from documents, requiring no custom training. This directly matches the requirement for a pre-built extraction model.

Why this answer

Azure AI Document Intelligence (formerly Form Recognizer) is the correct service because it is specifically designed to extract structured data from documents like invoices, including fields such as vendor name, invoice date, and total amount. It uses pre-built models trained on common document types, making it ideal for this use case without requiring custom training.

Exam trap

The trap here is that candidates may confuse Azure AI Language's entity extraction capabilities with document-specific extraction, not realizing that Azure AI Document Intelligence is purpose-built for semi-structured documents like invoices and forms.

How to eliminate wrong answers

Option A is wrong because Azure AI Language focuses on text analytics, sentiment analysis, and entity recognition from unstructured text, not on extracting structured fields from semi-structured documents like invoices. Option B is wrong because Azure OpenAI Service provides generative AI models (e.g., GPT-4) for text generation and conversation, not a pre-built model optimized for invoice data extraction. Option D is wrong because Azure Cognitive Search is a search and indexing service for building search experiences over data, not a document processing service for extracting key information from invoices.

Full explanation →

380

MCQhard

Refer to the exhibit. You are deploying a generative AI model as an online endpoint in Azure Machine Learning. You receive complaints that the endpoint returns 503 errors during peak hours. What is the most likely cause?

A.The manual scale setting with 2 instances may be insufficient for peak traffic.

B.The request timeout of 30 seconds is too short.

C.The environment variable MODEL_CACHE_SIZE is set too low.

D.The model version is not specified correctly.

AnswerA

A manual scale setting pinned at two instances cannot absorb peak-hour request volume, so the endpoint's instances saturate and Azure Machine Learning returns 503 errors. Configuring autoscale, or raising the instance count, matches capacity to the traffic constraint described in the stem.

Why this answer

503 errors during peak hours indicate that the endpoint is overwhelmed by the request volume. With manual scaling set to only 2 instances, the compute capacity is insufficient to handle the increased traffic, causing the service to reject requests. Azure Machine Learning online endpoints require sufficient instance count or autoscaling to absorb traffic spikes.

Exam trap

Microsoft often tests the distinction between HTTP status codes (503 vs. 408/504) to mislead candidates into confusing timeout-related errors with capacity-related errors.

How to eliminate wrong answers

Option B is wrong because a 30-second request timeout would cause timeout errors (e.g., 408 or 504), not 503 Service Unavailable errors; 503 specifically signals resource exhaustion, not slow processing. Option C is wrong because the MODEL_CACHE_SIZE environment variable affects model loading performance and caching, not the endpoint's ability to handle concurrent requests; a low cache size might cause slower inference but not 503 errors. Option D is wrong because an incorrect model version would result in deployment failures (e.g., 404 or 500 errors) during endpoint creation or update, not intermittent 503 errors during peak traffic.

Full explanation →

381

MCQhard

You are developing a conversational language understanding (CLU) project in Azure AI Language. You have defined intents and entities for a banking bot. During testing, you notice that utterances containing the phrase 'transfer funds' are sometimes predicted as the 'CheckBalance' intent instead of 'TransferMoney'. You have already added 20 labeled examples for 'TransferMoney' and 20 for 'CheckBalance'. What should you do to improve the model's ability to distinguish between these intents?

A.Add a prebuilt entity for 'money' to the project and map it to the 'TransferMoney' intent.

B.Add more labeled utterances for 'TransferMoney' that include variations of the phrase 'transfer funds' and similar expressions, ensuring diversity in phrasing.

C.Enable the 'Use multiple languages' option in the project settings to improve model understanding.

D.Increase the number of training epochs for the CLU model to allow it to learn more from the existing data.

AnswerB

This is correct because adding more diverse labeled examples for the confusing intent helps the model learn the distinguishing features. The issue is likely due to insufficient or non-diverse training data for 'TransferMoney', causing the model to misclassify ambiguous utterances. By providing varied examples that include the problematic phrase and its synonyms, the model can better generalize.

Why this answer

The model misclassifies utterances because the training data for 'TransferMoney' may not include enough variations of the phrase 'transfer funds'. Adding more diverse labeled examples for that intent helps the model learn the distinguishing features. Increasing epochs or adding entities does not directly address intent confusion.

Multilingual settings are irrelevant here.

Exam trap

The trap here is thinking that more training epochs or additional entities will resolve intent confusion, but the real issue is the diversity and quantity of labeled utterances for the problematic intent.

Full explanation →

382

Multi-Selectmedium

A company uses Azure Custom Vision to build a classifier for defect detection on a manufacturing line. They have labeled images of products with and without defects. Which TWO actions should they take to improve model performance?

Select 2 answers

A.Train for more iterations without validation.

B.Use images with balanced numbers of defect and non-defect samples.

C.Set the learning rate manually using the Custom Vision API.

D.Increase the number of images per tag, including variations in lighting and angle.

E.Reduce the number of images per tag to avoid overfitting.

AnswersB, D

Balanced datasets prevent bias toward majority class.

Why this answer

Balanced datasets prevent the model from becoming biased toward the majority class (e.g., non-defect images), which is critical for defect detection where defects are rare. Azure Custom Vision uses a weighted loss function during training, and class imbalance can cause the model to predict the majority class for most inputs, reducing recall for defects. Balanced samples ensure the model learns discriminative features for both classes equally.

Exam trap

The trap here is that candidates may think reducing images prevents overfitting (Option E) or that manual learning rate tuning (Option C) is possible in Custom Vision, but the service abstracts hyperparameter tuning and requires sufficient, varied data for robust defect detection.

Full explanation →

383

MCQhard

Your knowledge mining solution ingests documents from multiple tenants. Each tenant's data must be isolated and searchable only by that tenant. You have a single Azure AI Search service. How should you implement multi-tenancy?

A.Use separate skillsets for each tenant

B.Create a separate search service for each tenant

C.Use a single index with a tenant ID field and filter queries by that field

D.Use separate data sources within the same index

AnswerC

A tenant ID field with query filters enforces logical isolation within one index, satisfying the single-service constraint. Filters apply at query time, so each tenant retrieves only its own documents. This is the documented approach for multi-tenancy when index-per-tenant isolation is unnecessary, though filter discipline must be enforced consistently.

Why this answer

Azure AI Search supports multi-tenancy within a single service by using a shared index with a tenant ID field. Each document is tagged with a tenant identifier, and queries are scoped using OData `$filter` expressions (e.g., `$filter=tenantId eq 'tenant123'`). This ensures data isolation while keeping costs low and management simple, as only one search service and one index are needed.

Exam trap

The trap here is that candidates confuse data sources (which are just ingestion pipelines) with data partitioning, leading them to think separate data sources or skillsets provide isolation, when in fact only query-time filtering or separate indexes enforce tenant boundaries.

How to eliminate wrong answers

Option A is wrong because skillsets define enrichment pipelines (e.g., OCR, entity extraction) and are not used for data isolation; they apply to all documents in an index regardless of tenant. Option B is wrong because creating a separate search service for each tenant is unnecessarily expensive and complex, violating the requirement to use a single Azure AI Search service. Option D is wrong because separate data sources within the same index still store all documents together; data sources only define where data is pulled from, not how it is partitioned or secured at query time.

Full explanation →

384

MCQmedium

You have defined the custom WebApiSkill shown in the exhibit. The skill calls an Azure Function that can process up to 10 documents per second. However, you notice that the skill is failing with 429 errors. What is the most likely cause?

A.The timeout of 30 seconds is too short for the function to respond

B.The batch size of 5 is too large, causing the function to receive too many documents at once

C.The context '/document' is incorrect, causing all documents to be processed as one

D.The degreeOfParallelism of 3 causes too many concurrent requests, exceeding the function's capacity

AnswerD

degreeOfParallelism of 3 lets the indexer dispatch three concurrent calls per batch, so bursts exceed the function's 10 documents per second limit. The function throttles with 429 responses; lowering parallelism to 1 aligns throughput with capacity.

Why this answer

The `degreeOfParallelism` of 3 causes the AI Search enrichment pipeline to invoke the Azure Function with up to 3 concurrent batches, each of size 5, resulting in up to 15 documents per second. Since the function can only handle 10 documents per second, this exceeds its capacity and triggers HTTP 429 (Too Many Requests) errors.

Exam trap

The trap here is that candidates often focus on the batch size as the sole cause of rate limiting, overlooking that `degreeOfParallelism` multiplies the effective request rate, which is the actual trigger for 429 errors.

How to eliminate wrong answers

Option A is wrong because a 30-second timeout is typically sufficient for an Azure Function processing documents; 429 errors indicate rate limiting, not timeout. Option B is wrong because a batch size of 5 means the function receives 5 documents per invocation, which is within the 10-document-per-second capacity if only one batch is processed at a time. Option C is wrong because the context '/document' is the standard path for per-document processing in AI Search skills; using it does not cause all documents to be processed as one, but rather each document is processed individually.

Full explanation →

385

Multi-Selectmedium

Which TWO Azure AI services can be used together to build a solution that transcribes customer service calls and detects sentiment?

Select 2 answers

A.Azure AI Language (sentiment analysis)

B.Azure AI Speech (speech-to-text)

C.Azure AI Translator

D.Azure AI Speech (text-to-speech)

E.Azure AI Language (conversational language understanding)

AnswersA, B

Azure AI Language's sentiment analysis returns per-sentence and document-level scores, which suits transcribed call audio where emotion shifts mid-conversation. Combined with speech-to-text transcription, it satisfies the stem's requirement to detect sentiment across customer service calls, operating on the text output rather than the audio itself.

Why this answer

Azure AI Speech (option B) is correct because its speech-to-text capability transcribes the audio of customer service calls into text, which is the required first step for this solution. Azure AI Language (option A) is correct because its sentiment analysis feature evaluates the transcribed text and returns sentiment scores/labels (positive, negative, neutral, mixed), satisfying the detection requirement. Together, B feeds transcription output into A for sentiment evaluation.

Option C (Azure AI Translator) is not needed since the scenario does not require language translation. Option D (text-to-speech) is the reverse of what is needed—it synthesizes speech from text rather than transcribing calls. Option E (conversational language understanding) extracts intents and entities for conversational apps, not sentiment, so it does not fulfill the requirement.

Exam trap

The trap here is that candidates may confuse Azure AI Speech's text-to-speech with speech-to-text, or mistakenly think Azure AI Translator or conversational language understanding can perform sentiment analysis, when in fact only the specific sentiment analysis feature of Azure AI Language is designed for that task.

Full explanation →

386

MCQhard

You are building an agent using Azure AI Agent Service that must execute code in a sandbox environment. The code should be able to install Python packages. Which action type should you use?

A.function

B.code_interpreter

C.openApi

D.httpRequest

AnswerB

The code_interpreter action type runs Python in a Microsoft-managed sandbox, satisfying the requirement to execute code safely. It supports installing packages at runtime via pip within that sandbox, so dependencies can be added dynamically. This directly meets the stem's constraint that the agent install Python packages during execution.

Why this answer

The code_interpreter action type in Azure AI Agent Service provides a sandboxed environment where code can be executed, and it supports installing Python packages via pip. This allows the agent to run code that requires additional libraries not pre-installed.

Exam trap

AI-102 often tests the differences between action types; candidates may confuse code_interpreter with function or httpRequest, but only code_interpreter provides a sandbox for executing code and installing packages.

How to eliminate wrong answers

Option A is wrong because the function action type is for calling custom functions defined in the agent, not for executing arbitrary code in a sandbox. Option C is wrong because openApi is for calling external APIs defined by OpenAPI specifications, not for code execution. Option D is wrong because httpRequest is for making HTTP requests, not for executing code or installing packages.

Full explanation →

387

MCQmedium

You are building a web application that allows users to upload images of restaurant receipts and extract the total amount and merchant name. The receipts may be crumpled, rotated, or have handwritten notes. Which Azure AI service should you use to reliably extract this information?

A.Azure AI Vision Spatial Analysis

B.Azure AI Document Intelligence prebuilt receipt model

C.Azure AI Face API

D.Azure AI Vision Read API

AnswerB

The prebuilt receipt model in Azure AI Document Intelligence is specifically trained to extract key fields such as total, merchant name, transaction date, and line items from receipts. It handles variations in receipt formats, including rotation and crumpling, and returns structured JSON with confidence scores, making it ideal for this scenario without custom parsing.

Why this answer

The prebuilt receipt model in Azure AI Document Intelligence is purpose-built to extract structured fields like total and merchant name from receipts, even when they are crumpled or rotated. It uses a combination of OCR and machine learning to understand receipt layouts and output key-value pairs, eliminating the need for custom parsing or additional services.

Exam trap

The trap here is assuming that the Read API's text extraction is sufficient, but it lacks the semantic understanding to identify specific fields like total amount without extra processing.

Full explanation →

388

MCQhard

You are deploying a generative AI assistant with Azure OpenAI Service. Company policy requires that the assistant must never output profanity, and that any attempt to elicit such content must be blocked at the service level rather than filtered in application code. You need the least administrative effort to enforce this. What should you do?

A.Add a system message instructing the model to never use profanity.

B.Enable Azure API Management policies to inspect and rewrite responses containing profanity.

C.Set the model temperature to 0 and top_p to 1 to reduce the chance of profanity.

D.Create a custom content filter with a blocklist for profanity terms and associate it with the deployment.

AnswerD

Azure OpenAI content filters support custom blocklists of terms that are blocked at the service level when associated with a deployment. Creating a blocklist for profanity and attaching it to the deployment enforces the policy without application-side filtering. This is the least-effort service-level control that directly targets the prohibited terms and blocks elicitation attempts.

Why this answer

Custom content filters in Azure OpenAI let you define blocklists of specific terms and apply them to a deployment. Terms in the blocklist are blocked at the service level, which aligns with the policy that prohibited content must be stopped by the service rather than by application code. Associating the filter with the deployment enforces the rule for all calls.

Exam trap

The trap here is treating a system message or sampling parameter as a content-safety control, when only service-level filtering such as a custom blocklist reliably blocks prohibited terms.

Full explanation →

389

MCQeasy

Your company wants to moderate user-uploaded images for adult content. Which Azure AI service should you use?

A.Azure AI Content Safety

B.Azure AI Face

C.Azure AI Document Intelligence

D.Azure AI Vision Image Analysis

AnswerA

Azure AI Content Safety provides dedicated image moderation with severity-scored adult, racy, and violent classifications, directly satisfying the requirement to filter user-uploaded images. Unlike Computer Vision, which offers only basic adult/gory flags, Content Safety returns graded severity levels, enabling precise threshold-based blocking aligned with your moderation policy.

Why this answer

Azure AI Content Safety is the correct service because it is specifically designed to detect and moderate inappropriate content, including adult content, in images and text. It provides severity-based classifications (safe, low, medium, high) for categories such as hate, self-harm, sexual, and violence, making it ideal for user-uploaded image moderation.

Exam trap

The trap here is that candidates often confuse Azure AI Vision Image Analysis (which can detect adult content via the 'adult' flag in its Analyze Image API) with the dedicated Azure AI Content Safety service, but the exam expects the service purpose-built for content moderation with granular severity levels and broader category support.

How to eliminate wrong answers

Option B is wrong because Azure AI Face is focused on detecting, analyzing, and recognizing human faces, not on moderating adult content. Option C is wrong because Azure AI Document Intelligence is designed to extract text, key-value pairs, and tables from documents, not to analyze images for adult content. Option D is wrong because Azure AI Vision Image Analysis provides general image descriptions, object detection, and optical character recognition, but lacks the specific content moderation categories and severity scoring needed for adult content detection.

Full explanation →

390

MCQeasy

Your company has a large set of PDF documents stored in Azure Blob Storage. You need to index these documents in Azure Cognitive Search so that users can search the text content. What is the first step you should take?

A.Create an index with a field for each metadata property.

B.Create a skillset to extract text from PDFs.

C.Create a data source that connects to Azure Blob Storage.

D.Create an indexer that runs daily.

AnswerC

Creating a data source establishes the connection between the search indexer and Azure Blob Storage, which is the prerequisite for pulling PDF content. The indexer cannot traverse or extract text from the container until this data source object exists, so it must precede defining the index, skillset, or indexer itself.

Why this answer

In Azure Cognitive Search, the pipeline always begins with a data source that defines the connection to the underlying content store. Before you can create an indexer, skillset, or index, you must register the Azure Blob Storage container as a data source so the service knows where to pull the PDFs from. Only after the data source exists can an indexer crawl it and push content through the enrichment pipeline.

Exam trap

AI-102 often tests the ordering of the Cognitive Search pipeline, and candidates mistakenly jump to the indexer or skillset because those feel like the 'real work' — but the data source is always the mandatory first object.

How to eliminate wrong answers

Option A is wrong because an index defines the searchable schema, but it cannot be created meaningfully until the source content and its fields are known — and it is not the first step in the ingestion pipeline. Option B is wrong because a skillset is an optional enrichment layer (OCR, entity extraction, etc.) that runs after an indexer has already pulled documents from a data source. Option D is wrong because an indexer requires a data source to exist first; scheduling it daily is a later configuration step, not the initial one.

Full explanation →

391

MCQeasy

You are developing a chat application that uses Azure OpenAI Service to answer customer queries. The solution must ensure that the model does not generate harmful or offensive content. Which Azure AI service should you configure?

A.Azure AI Bot Service

B.Azure AI Search

C.Azure AI Content Safety

D.Azure AI Language

AnswerC

Azure AI Content Safety provides dedicated harm classification across categories such as hate, violence and self-harm, filtering both prompts and completions. Configuring it on the Azure OpenAI endpoint enforces the requirement that the model never generates harmful or offensive content.

Why this answer

Azure AI Content Safety is specifically designed to detect and filter harmful or offensive content in text and images, making it the appropriate service to integrate with an Azure OpenAI chat application to enforce content safety policies. It provides APIs for content moderation, including severity-based filtering for hate, self-harm, sexual, and violence categories, which directly addresses the requirement to prevent the model from generating harmful output.

Exam trap

The trap here is that candidates often confuse Azure AI Language's text analytics capabilities (like sentiment analysis) with content safety, assuming that language understanding inherently includes harm detection, but Azure AI Content Safety is a separate, specialized service for content moderation.

How to eliminate wrong answers

Option A is wrong because Azure AI Bot Service is a platform for building, testing, and deploying conversational agents (bots), not a content moderation or safety service; it does not natively filter harmful content from model outputs. Option B is wrong because Azure AI Search is a search-as-a-service solution for indexing and querying data, with no built-in content safety or moderation capabilities for generated text. Option D is wrong because Azure AI Language provides natural language processing features like sentiment analysis, key phrase extraction, and question answering, but it does not include content safety filters for detecting harmful or offensive content.

Full explanation →

392

Multi-Selectmedium

You are building an Azure AI Search solution that uses a custom skill to enrich documents. The custom skill is implemented as an Azure Function. You need to ensure that the custom skill can access the documents' content and output enriched data. Which two actions should you perform? (Choose two.)

Select 2 answers

A.Add the Azure Function's output to the index's field mappings.

B.Ensure the Azure Function accepts a JSON payload with the expected input and returns a JSON response.

C.Set the custom skill's batch size to 1 to ensure sequential processing.

D.Configure the indexer to use a managed identity to access the Azure Function.

E.Define the custom skill in the skillset with the correct input source and output mappings.

AnswersB, E

Azure AI Search sends a JSON payload containing the input values to the custom skill and expects a JSON response with the output values. The Azure Function must be implemented to parse the request and return the appropriate structure, otherwise the skill will fail or produce no output.

Why this answer

The custom skill must be defined in the skillset with proper input and output mappings to integrate with the enrichment pipeline. Additionally, the Azure Function must handle the JSON payload correctly. These two actions ensure the skill can access document content and return enriched data for indexing.

Exam trap

The trap here is focusing on security or performance settings like managed identity or batch size, which are not required for basic functionality of a custom skill.

Full explanation →

393

MCQhard

You are a data scientist for Contoso Pharmaceuticals. The company has thousands of research documents in PDF format stored in Azure Blob Storage. You need to build an Azure Cognitive Search solution that enables researchers to search for documents based on chemical compound names, disease mentions, and experimental results. The solution must extract these entities using a custom AI model built in Azure AI Language. Additionally, the solution must support semantic search for natural language queries. The search index must be updated daily with new documents. You have an existing Azure AI Language custom entity extraction model that recognizes chemical compounds and diseases. The model is deployed as an endpoint. You need to configure the enrichment pipeline. What should you do?

A.Create a custom skill in the skillset that calls the custom entity extraction endpoint via HTTP.

B.Deploy the custom model to Azure AI Document Intelligence and use a Document Intelligence skill.

C.Add the custom entity extraction as a field mapping in the indexer.

D.Use the built-in Entity Recognition skill and configure it to use your custom model endpoint.

AnswerA

A custom skill in the skillset invokes your deployed Azure AI Language endpoint over HTTP, letting the enrichment pipeline attach the extracted chemical compounds and diseases as indexable fields. This satisfies the requirement to use your existing custom entity extraction model rather than a built-in skill.

Why this answer

To integrate a custom AI model from Azure AI Language into an Azure Cognitive Search enrichment pipeline, you need to create a custom skill in the skillset that calls the custom entity extraction endpoint via HTTP. The built-in Entity Recognition skill only supports prebuilt models and cannot be configured to use a custom model endpoint. Deploying the model to Azure AI Document Intelligence and using a Document Intelligence skill is not appropriate because the model is already deployed in Azure AI Language as a custom entity extraction model.

Field mappings in the indexer are for direct field-to-field mappings from the data source to the index, not for calling external AI services. Therefore, Option A is the correct approach.

Full explanation →

394

MCQeasy

You are building an agent in Microsoft Copilot Studio that needs to send a confirmation email after a user completes a survey. The email should be sent using the user's email address collected during the conversation. Which feature should you use to send the email?

A.Create a Power Automate flow that sends an email and call it from the topic.

B.Use the Send an email action directly in Copilot Studio.

C.Use an adaptive card with an email button.

D.Use the email channel to send a response.

AnswerA

Power Automate provides the Office 365 Outlook connector with a Send an email action, letting the topic pass the collected address as a dynamic input. Copilot Studio topics alone cannot dispatch SMTP mail, so the flow satisfies the send-after-survey requirement.

Why this answer

Microsoft Copilot Studio does not have a native 'Send an email' action; it relies on Power Automate flows to perform external actions like sending emails. By creating a flow that uses the user's email address collected during the conversation and calling it from the topic, you can send a confirmation email after the survey is completed.

Exam trap

The trap here is that candidates assume Copilot Studio has built-in email actions similar to Power Automate, but Microsoft deliberately separates conversation logic from external integrations to enforce a modular architecture.

How to eliminate wrong answers

Option B is wrong because Copilot Studio does not include a built-in 'Send an email' action; it lacks native email capabilities and must delegate such tasks to Power Automate. Option C is wrong because an adaptive card with an email button only provides a clickable interface to open the user's default mail client; it does not programmatically send an email from the bot. Option D is wrong because the email channel in Copilot Studio is used to receive and respond to messages via email, not to send outbound emails like a confirmation.

Full explanation →

395

MCQmedium

You are developing an app that analyzes images of restaurant receipts. The app must extract the merchant name, transaction date, and total amount from each receipt. You want to minimize development effort and use a prebuilt Azure AI service. Which service should you use?

A.Azure AI Document Intelligence with the prebuilt-receipt model

B.Azure AI Language with custom named entity recognition

C.Azure AI Vision with the Read API

D.Azure AI Custom Vision with a classification model

AnswerA

The prebuilt-receipt model in Azure AI Document Intelligence is specifically trained to extract key fields from receipts, including merchant name, transaction date, and total. It returns structured JSON with these fields, so minimal development effort is needed. It handles common receipt variations and is the recommended prebuilt solution for receipt processing.

Why this answer

Azure AI Document Intelligence provides a prebuilt receipt model that is purpose-built to extract merchant name, transaction date, and total from receipt images. It returns structured data without requiring custom model training or text parsing, which aligns with the goal of minimizing development effort. Other services either return unstructured text or require custom training.

Exam trap

The trap here is assuming that the Read API's OCR output is sufficient for structured field extraction, when it only provides raw text.

Full explanation →

396

MCQmedium

Your knowledge mining solution uses Azure AI Search. Users complain that search results are not relevant. You have enabled semantic search but results still lack context. What should you do to improve relevance?

A.Ensure the index includes a semantic configuration with title and content fields

B.Increase the number of partitions to handle more data

C.Configure a scoring profile with boosting based on metadata

D.Increase the number of replicas to improve query performance

AnswerA

Semantic search ranks results using a semantic configuration that designates title and content fields for caption and answer generation. Without that configuration, semantic ranking cannot use the fields, so relevance and contextual captions remain poor.

Why this answer

Semantic search in Azure AI Search requires a semantic configuration that explicitly maps the title and content fields to be used for semantic ranking. Without this configuration, the search engine cannot apply the deep neural network models that understand context and intent, so results remain based on keyword matching even when the semantic search feature is enabled.

Exam trap

The trap here is that candidates assume enabling the semantic search feature alone is sufficient, but Azure AI Search requires an explicit semantic configuration to map the fields that the semantic model will use for reranking.

How to eliminate wrong answers

Option B is wrong because increasing the number of partitions scales the index for larger data volumes but does not improve relevance or semantic understanding of search results. Option C is wrong because scoring profiles with boosting based on metadata can adjust ranking weights but do not provide the contextual, language-understanding capabilities that semantic search offers. Option D is wrong because increasing replicas improves query throughput and availability, not the relevance or contextual quality of search results.

Full explanation →

397

MCQhard

You are building a generative AI solution that uses Azure OpenAI function calling to let the model invoke backend APIs. During testing, the model sometimes invents parameter values that cause API errors. You need to make function invocation more reliable. What should you do?

A.Define clear function schemas with typed parameters, required fields, and descriptions, and validate arguments before executing the API call.

B.Set tool_choice to auto and increase the temperature to encourage more creative argument generation.

C.Disable parallel tool calls and force the model to call only one function per turn.

D.Include the full OpenAPI specification of every backend API in the system message.

AnswerA

Function calling relies on the model producing arguments that match the declared JSON schema. Precise types, required fields, and descriptions guide the model, while server-side validation rejects invalid arguments before they reach the API. Together they reduce invented values and prevent downstream errors.

Why this answer

Reliable function calling depends on well-defined schemas and validation. Typed parameters, required fields, and descriptive names help the model map user intent to correct arguments, and validating before execution prevents invalid values from reaching the API. Higher temperature, oversized specifications, and parallel-call limits do not enforce argument correctness.

Exam trap

The trap here is believing that adding more API documentation or creative sampling improves argument accuracy.

Full explanation →

398

MCQhard

You have the above skillset in Azure AI Search. The indexer processes a document with 12,000 characters of content. How many entity recognition skill executions occur?

A.4

B.2

C.3

D.1

AnswerC

Entity recognition splits input into 5,000-character chunks, so 12,000 characters yields three executions: two full chunks plus a 2,000-character remainder. This satisfies the stem's chunking constraint, where each skill invocation processes at most 5,000 characters, making three the accurate count.

Why this answer

The Azure AI Search entity recognition skill has a maximum text length per execution of 5,000 characters. A document with 12,000 characters is split into chunks of up to 5,000 characters, resulting in three chunks (5,000 + 5,000 + 2,000). Each chunk triggers one skill execution, so three executions occur.

Exam trap

The Azure AI Search entity recognition skill has a maximum text length per execution of 5,000 characters. Candidates might incorrectly divide 12,000 by 5,000 and round down to 2, or assume a single execution can handle the entire document, ignoring the chunking behavior.

How to eliminate wrong answers

Option A is wrong because 4 executions would require more than 15,000 characters (4 × 5,000), but the document has only 12,000 characters. Option B is wrong because 2 executions would cover only 10,000 characters (2 × 5,000), leaving 2,000 characters unprocessed. Option D is wrong because 1 execution can handle only up to 5,000 characters, but the document has 12,000 characters, so it must be split into multiple chunks.

Full explanation →

399

MCQeasy

A development team is building an Azure OpenAI chat application and wants the model to return structured JSON that matches a defined schema containing an 'intent' field and a 'confidence' field. The application will parse the response directly. Which deployment parameter should the team use to constrain the output format?

A.Set the 'temperature' parameter to 0.

B.Set the 'top_p' parameter to 1.

C.Configure a structured output format, such as a JSON schema response format, on the request.

D.Increase the 'max_tokens' value to allow the full JSON document to be generated.

AnswerC

Structured outputs, exposed through the response format parameter with a JSON schema, constrain generation so the completion conforms to the supplied schema, including the required intent and confidence fields. This is the purpose-built mechanism for machine-parseable responses and removes the need for fragile prompt-only formatting instructions or post-processing repairs in the application.

Why this answer

Constraining a model to emit machine-parseable JSON with specific fields is accomplished with structured outputs, where a JSON schema is supplied in the request's response format. This guarantees schema conformance far more reliably than sampling parameters or length limits, which only influence variability or truncation.

Exam trap

The trap here is confusing determinism with structure: lowering temperature or raising max_tokens does not make a model emit schema-valid JSON.

Full explanation →

400

MCQmedium

A news organization uses Azure Video Indexer to generate transcripts of live broadcasts. They notice that the speaker names are not appearing in the transcript. What is the most likely cause?

A.The video resolution is too low for OCR.

B.The speaker identification model has not been trained with voice samples.

C.The video format is not supported.

D.The language is not set correctly.

AnswerB

Speaker identification in Azure Video Indexer requires a trained voice model built from submitted speaker audio samples; without enrolment, the service cannot map voices to named individuals. The stem's missing speaker names therefore point to an untrained model, since transcription itself still succeeds but attribution stays anonymous.

Why this answer

Speaker names are missing because Azure Video Indexer's speaker identification feature requires pre-trained voice samples to match speakers to their identities. Without a custom voice model trained on known speakers' audio, the service can only label speakers as 'Speaker #1', 'Speaker #2', etc., but cannot assign actual names. This is a supervised learning process where the model must be trained with labeled voice samples before it can recognize and name speakers.

Exam trap

The trap here is that candidates may confuse speaker identification with automatic diarization or assume that speaker names are automatically extracted from the video metadata, when in fact Azure Video Indexer requires explicit training of a custom Person Model with voice samples to assign names.

How to eliminate wrong answers

Option A is wrong because OCR (optical character recognition) is used for extracting text from video frames, not for identifying speakers or generating transcripts; low resolution would affect text extraction but not speaker name assignment. Option C is wrong because Azure Video Indexer supports a wide range of common video formats (e.g., MP4, MOV, AVI), and an unsupported format would cause a failure to index or generate any transcript, not just missing speaker names. Option D is wrong because setting the language incorrectly would affect the accuracy of the speech-to-text transcription (e.g., wrong words or gibberish), but it would not prevent speaker names from appearing; speaker identification is a separate model that requires training regardless of language.

Full explanation →

401

MCQmedium

Your company uses Microsoft Copilot for Microsoft 365. You need to ensure that Copilot only accesses data from approved SharePoint sites and does not use any other organizational data. What should you configure?

A.Configure Conditional Access policies in Microsoft Entra ID.

B.Use Microsoft Purview Data Map to catalog the approved sites.

C.Apply sensitivity labels to the approved SharePoint sites.

D.Set up data retention policies in Microsoft Purview.

AnswerC

Sensitivity labels can be used to restrict Copilot's data sources.

Why this answer

Sensitivity labels can be configured to restrict Microsoft Copilot for Microsoft 365 to only access content from approved SharePoint sites. By applying a sensitivity label that includes the 'mark content' or 'encrypt content' setting with a specific scope, you can use the 'Microsoft Copilot for Microsoft 365' condition under 'Access control' to block Copilot from processing data from unlabeled or non-approved sites. This ensures Copilot respects the label's policy and excludes all other organizational data.

Exam trap

The trap here is that candidates often confuse data access control with authentication or data lifecycle management, leading them to select Conditional Access or retention policies instead of recognizing that sensitivity labels provide the specific Copilot data source restriction capability.

How to eliminate wrong answers

Option A is wrong because Conditional Access policies in Microsoft Entra ID control user authentication and access to applications, not the data sources that Copilot can index or retrieve content from. Option B is wrong because Microsoft Purview Data Map is a metadata catalog for data governance and discovery, not a mechanism to enforce access restrictions on Copilot's data sources. Option D is wrong because data retention policies in Microsoft Purview manage how long data is kept and when it is deleted, not which data Copilot is allowed to access.

Full explanation →

402

MCQhard

You have configured a system message for an Azure OpenAI chat completion deployment as shown in the exhibit. Users are reporting that the assistant sometimes refuses to answer questions that are clearly within the scope of the provided data. What is the most likely issue?

A.The system message encourages the model to err on the side of caution, leading to false-negative refusals.

B.The system message explicitly prohibits making up information, which is correct behavior.

C.The system message does not include instructions to use the provided data.

D.The temperature parameter is set too high, causing the model to hallucinate.

AnswerA

Overly defensive system message wording instructs the model to decline when uncertain, so legitimate in-scope questions fall below its confidence threshold and trigger refusals. Softening that instruction, while keeping grounding constraints, restores answers without permitting out-of-scope responses.

Why this answer

The system message likely contains overly cautious language (e.g., 'only answer if you are certain' or 'do not speculate'), which causes the model to refuse answering even when the data clearly supports the response. This is a known behavior in Azure OpenAI chat completions where the system message's tone and constraints directly influence refusal rates, leading to false-negative refusals.

Exam trap

Microsoft often tests the misconception that refusal issues are caused by missing data instructions or high temperature, when in fact the root cause is the system message's overly cautious phrasing that induces false-negative refusals.

How to eliminate wrong answers

Option B is wrong because explicitly prohibiting making up information is a standard best practice to reduce hallucination, not a cause of false-negative refusals; it does not inherently make the model overly cautious. Option C is wrong because the system message in the exhibit (as described) does include instructions to use the provided data, so the issue is not a missing directive but the cautious phrasing. Option D is wrong because a high temperature parameter increases randomness and creativity, leading to hallucinations or off-topic responses, not systematic refusal to answer within scope.

Full explanation →

403

Multi-Selectmedium

Which TWO actions can you take to reduce the cost of using Azure OpenAI Service for a chat application?

Select 2 answers

A.Increase the frequency penalty.

B.Enable content filtering.

C.Set the max_tokens parameter to a lower value.

D.Increase the max_tokens parameter to allow longer responses.

E.Use a smaller model like GPT-3.5 instead of GPT-4.

AnswersC, E

Reduces token count per response.

Why this answer

Reducing the max_tokens parameter directly limits the number of tokens generated per API call, which lowers the cost since Azure OpenAI Service bills per token (both input and output). By capping the response length, you avoid paying for unnecessarily long completions.

Exam trap

The trap here is that candidates often confuse cost-saving techniques with performance-tuning parameters, mistakenly thinking that adjusting penalty settings or content filtering reduces token consumption, when in fact only limiting token output or using a cheaper model directly lowers the bill.

Full explanation →

404

MCQhard

You are building a generative AI feature that drafts marketing copy in English and must also produce accurate French and German versions. The team wants a single deployment that handles all three languages without maintaining separate prompt templates per language, and the translations must preserve the marketing tone. Which Azure OpenAI capability should the solution rely on?

A.Use a single model deployment and instruct it in the system message to generate in the requested language while preserving tone.

B.Translate the English output using Azure AI Translator and then post-process with a grammar model.

C.Deploy a separate model instance per language and route requests based on the target locale.

D.Fine-tune the model on parallel marketing copy in all three languages before deploying it.

AnswerA

Azure OpenAI chat models are multilingual and can follow instructions that specify the target language and desired style. A single deployment with a system message describing the marketing tone and the requested output language avoids separate templates per language. This directly satisfies the goal of one deployment handling English, French, and German with consistent tone.

Why this answer

Azure OpenAI chat models are trained on multilingual data and respond well to explicit language and style instructions. Specifying the target language and the marketing tone in the system message lets one deployment serve English, French, and German without maintaining separate templates. Separate deployments, an external translation pipeline, or fine-tuning all add complexity that the scenario does not require and do not better guarantee tone preservation.

Exam trap

The trap here is assuming multilingual output requires separate deployments or a dedicated translation service, when a single deployment with clear language and tone instructions is sufficient.

Full explanation →

405

MCQmedium

You are deploying an Azure AI Services resource by using an ARM template as part of an automated pipeline. The template must create the resource, a key vault, and a role assignment that allows a specific managed identity to read the resource key from the key vault. The deployment fails with an authorization error when creating the role assignment. You need to resolve the failure. What should you do?

A.Assign the User Access Administrator role to the pipeline service principal at the target scope before the deployment.

B.Add the Contributor role to the pipeline service principal at the subscription scope.

C.Enable the key vault for template deployment and add the pipeline identity to the key vault access policy.

D.Change the template to use a system-assigned managed identity for the Azure AI Services resource instead of the key vault.

AnswerA

Creating role assignments requires the Microsoft.Authorization/roleAssignments/write permission, which is included in User Access Administrator or Owner. Granting that role to the pipeline identity at the deployment scope lets the ARM template create the role assignment successfully, resolving the authorization failure.

Why this answer

Role assignment creation is a privileged operation that requires permissions such as Microsoft.Authorization/roleAssignments/write, provided by Owner or User Access Administrator. Contributor and key vault access policies do not include that permission, so granting User Access Administrator to the pipeline identity at the target scope is what allows the template to complete.

Exam trap

The trap here is assuming Contributor is sufficient for every deployment action, when Contributor deliberately excludes the ability to grant access to others.

Full explanation →

406

Multi-Selectmedium

You are building a generative AI feature where an Azure OpenAI model must call internal functions, such as looking up an order status and issuing a refund, based on the user's natural language request. You need the model to decide which function to invoke and to incorporate the function result into its answer. Which two actions should you take? (Choose two.)

Select 2 answers

A.Execute the returned function call in your application and send the result back in a subsequent message with the tool role.

B.Attach the internal API documentation as an 'On Your Data' source so the model can retrieve and call the endpoints.

C.Fine-tune the model on historical order and refund conversations so it memorizes the internal APIs.

D.Raise the temperature so the model explores multiple possible functions before answering.

E.Define each function with a name, description, and JSON parameter schema and pass them in the tools parameter of the chat request.

AnswersA, E

The model only proposes a function call; it does not execute internal code. The application must run the requested lookup or refund and then return the output to the model in a tool-role message so the model can compose a grounded final answer. This round trip is what connects the model's decision to real business data and completes the interaction.

Why this answer

Function calling requires two coordinated steps: declaring the callable functions with their parameter schemas in the request, and then executing the model's requested call and returning the result in a tool-role message. Together these let the model decide on the correct internal operation and produce a final answer grounded in real system data.

Exam trap

The trap here is believing fine-tuning or document retrieval can invoke internal APIs, when only declared tools plus application-side execution return usable results.

Full explanation →

407

MCQhard

A multinational corporation uses Azure AI Language to analyze customer feedback in multiple languages. The solution must detect the language of incoming text and then perform sentiment analysis. Which approach minimizes latency and cost?

A.Use the sentiment analysis API with multilingual support

B.Use the Translator service to translate text to English, then call sentiment analysis

C.Store text in Azure AI Search and use cognitive skills for sentiment

D.Call the language detection API followed by the sentiment analysis API

AnswerD

This is the correct approach because you first call the language detection API to identify the language, then call the sentiment analysis API with that language code. This combination directly meets the requirement with minimal latency and cost, avoiding translation or indexing overhead.

Why this answer

The requirement is to detect the language of incoming text and then perform sentiment analysis. The sentiment analysis API in Azure AI Language does not detect language; it requires the language to be specified. Therefore, you must first call the language detection API to identify the language, then pass that language code to the sentiment analysis API.

While this involves two API calls, it is still more efficient than translating text (Option B) which adds significant latency and cost, or using cognitive skills with Azure AI Search (Option C) which adds indexing overhead. Option A is wrong because the sentiment analysis API cannot detect language on its own; it requires the language as an input parameter. Thus, Option D minimizes latency and cost by only performing the necessary steps without unnecessary transformations.

Exam trap

The trap is that candidates may assume the sentiment analysis API automatically detects language, but in reality it requires the language to be specified. Alternatively, they might think translation is needed, but language detection plus native sentiment analysis is more efficient for multilingual support.

How to eliminate wrong answers

Option B is wrong because translating text to English before sentiment analysis adds the latency and cost of an extra Translator API call, and translation may alter nuances or sentiment, reducing accuracy. Option C is wrong because storing text in Azure AI Search and using cognitive skills introduces unnecessary infrastructure overhead, indexing delays, and additional costs for search and skillset execution, which is not optimal for a simple real-time sentiment analysis pipeline. Option D is wrong because calling the language detection API first, then the sentiment analysis API, requires two separate API calls, doubling the latency and cost compared to using the single multilingual sentiment API that handles detection internally.

Full explanation →

408

MCQeasy

You are planning to deploy an Azure AI Services multi-service resource. You need to ensure that the resource can be used by applications running in an Azure Kubernetes Service (AKS) cluster without embedding keys in the application code. What should you do?

A.Store the Azure AI Services key in a Kubernetes secret and mount it as an environment variable in the application pods.

B.Configure Azure AD Pod Identity or Workload Identity for the AKS cluster and assign the Cognitive Services User role to the identity on the Azure AI Services resource.

C.Enable a system-assigned managed identity on the AKS cluster and assign the Cognitive Services User role to it on the Azure AI Services resource.

D.Use Azure Key Vault to store the Azure AI Services key and retrieve it at runtime by using the AKS cluster's managed identity.

AnswerB

Workload Identity (or the older Pod Identity) allows Kubernetes pods to use a managed identity to authenticate to Azure services. By assigning the Cognitive Services User role to that identity on the Azure AI Services resource, the application can obtain a token from Microsoft Entra ID and call the service without any keys. This meets the keyless requirement.

Why this answer

To enable keyless authentication for applications in AKS, you should use Azure AD Workload Identity (or Pod Identity) to associate a managed identity with the pods. That identity must be granted the Cognitive Services User role on the Azure AI Services resource. The application can then use DefaultAzureCredential to obtain a token and call the service without any keys.

Exam trap

The trap here is assuming that enabling a managed identity on the AKS cluster is enough, when in fact you must configure workload identity to make that identity available to the pods.

Full explanation →

409

Multi-Selecteasy

You are using Azure OpenAI Service to generate code snippets. The output must be safe and free of security vulnerabilities. Which TWO practices should you follow? (Select TWO.)

Select 2 answers

A.Increase the temperature parameter to encourage diversity

B.Rely on the model's built-in safety features

C.Use Azure AI Content Safety to filter outputs

D.Fine-tune the model with a dataset of secure code examples

E.Include a system message instructing the model to follow secure coding practices

AnswersC, E

Content safety can detect and block harmful code.

Why this answer

Azure AI Content Safety is a dedicated service that provides an additional layer of filtering for harmful or inappropriate content, including security vulnerabilities, beyond what the model itself offers. It allows you to define custom severity thresholds and blocklists, ensuring that generated code snippets are safe before they reach the user. This is a recommended practice for production deployments to mitigate risks like injection attacks or exposure of sensitive patterns.

Exam trap

The trap here is that candidates often assume the model's built-in safety features are sufficient (Option B) or that increasing temperature (Option A) improves safety by adding randomness, when in fact Azure AI Content Safety is the explicit, exam-tested tool for output filtering in generative AI solutions.

Full explanation →

410

Multi-Selecthard

Which THREE components are required to build a custom chat application using Azure OpenAI Service that can answer questions based on your own private data?

Select 3 answers

A.The 'Add your data' feature configured in Azure OpenAI Studio.

B.Azure AI Search index.

C.An Azure OpenAI Service deployment.

D.A fine-tuned custom model.

E.An Azure OpenAI embeddings model deployment.

AnswersA, B, C

Enables grounding on private data.

Why this answer

The 'Add your data' feature in Azure OpenAI Studio provides a no-code interface to connect your private data sources (e.g., Azure Blob Storage, local files) to an Azure OpenAI chat model. It automatically chunks the data, creates an Azure AI Search index, and configures the retrieval-augmented generation (RAG) pipeline, enabling the model to answer questions grounded in your proprietary content without fine-tuning.

Exam trap

The trap here is that candidates often confuse fine-tuning (Option D) with retrieval-augmented generation, assuming that custom data requires model retraining, when in fact the 'Add your data' feature uses a RAG approach that does not modify the base model.

Full explanation →

411

MCQmedium

Your company wants to build a custom generative AI model that generates architectural designs. The model should be trained on the company's proprietary dataset of floor plans and designs. Which Azure service should you use?

A.Azure OpenAI Service

B.Azure Machine Learning

C.Azure AI Vision

D.Azure AI Document Intelligence

AnswerB

Azure Machine Learning provides the training pipelines, compute and model registration needed to fine-tune a generative model on proprietary floor-plan data. It satisfies the custom-training requirement that prebuilt Azure OpenAI models cannot meet with private datasets.

Why this answer

Azure Machine Learning is the correct service because it provides a comprehensive platform for training custom generative AI models using your own proprietary datasets. It supports deep learning frameworks like PyTorch and TensorFlow, enabling you to build, train, and deploy a custom generative model for architectural designs, whereas Azure OpenAI Service is limited to pre-trained models and fine-tuning, not custom training from scratch.

Exam trap

The trap here is that candidates often confuse 'fine-tuning' (offered by Azure OpenAI Service) with 'custom training from scratch' (offered by Azure Machine Learning), leading them to incorrectly select Azure OpenAI Service when the question explicitly requires training a model on proprietary data with custom architecture.

How to eliminate wrong answers

Option A is wrong because Azure OpenAI Service only allows fine-tuning of existing pre-trained models (like GPT-4) on your data, not training a custom generative model from scratch on proprietary floor plans; it lacks the flexibility to define custom architectures. Option C is wrong because Azure AI Vision is designed for image analysis tasks (e.g., object detection, OCR) and does not support generative model training or generation of new designs. Option D is wrong because Azure AI Document Intelligence is specialized for extracting structured data from documents (e.g., forms, invoices) and cannot be used to train or generate architectural designs.

Full explanation →

412

MCQhard

You have deployed a generative AI model using Azure Machine Learning. The model is used for generating financial reports. You need to monitor the model's performance and detect data drift in the input data. What should you use?

A.Azure Machine Learning data drift monitoring

B.Azure Monitor

C.Application Insights

D.Azure AI Language

AnswerA

Azure Machine Learning data drift monitoring directly satisfies the requirement to detect drift in the model's input data. It computes statistical divergence between a baseline dataset and recent inference data, covering numerical and categorical features, and raises alerts when distributions shift. Model performance metrics alone would not surface input-distribution changes for the financial reports.

Why this answer

Azure Machine Learning data drift monitoring is the correct choice because it is specifically designed to detect statistical changes in input data over time, which is critical for generative AI models used in financial reporting where data distributions can shift due to market changes or new regulations. It compares the current input data distribution against a baseline dataset using metrics like Wasserstein distance or Population Stability Index, and triggers alerts when drift exceeds a threshold. This ensures the model's outputs remain reliable and compliant, which is a core requirement for generative AI solutions in regulated industries.

Exam trap

The trap here is that candidates confuse general monitoring tools (Azure Monitor, Application Insights) with the specialized data drift detection capability in Azure Machine Learning, assuming any monitoring tool can detect statistical drift in input data.

How to eliminate wrong answers

Option B is wrong because Azure Monitor is a general-purpose observability service for infrastructure and application metrics (e.g., CPU, memory, request rates), not for detecting statistical data drift in model inputs. Option C is wrong because Application Insights focuses on application performance monitoring (APM) and telemetry (e.g., request failures, dependencies), not on comparing data distributions or detecting drift in input features. Option D is wrong because Azure AI Language is a service for natural language processing tasks (e.g., sentiment analysis, key phrase extraction), not for monitoring model performance or detecting data drift in input data.

Full explanation →

413

MCQmedium

You are troubleshooting an agentic solution where the agent is not returning responses within acceptable time limits. You suspect the agent is making too many sequential calls to external tools. Which strategy should you recommend to reduce latency?

A.Increase the max token limit

B.Enable parallel tool execution

C.Add more tools to distribute the load

D.Reduce the thread history length

AnswerB

Parallel tool execution dispatches independent external tool calls concurrently rather than sequentially, so total latency reflects the slowest call instead of the sum of all calls. This directly addresses the stem's constraint of excessive sequential tool invocations exceeding acceptable response times.

Why this answer

Enabling parallel tool execution allows the agent to invoke multiple external tools simultaneously rather than sequentially, directly reducing the total latency caused by serial tool calls. This is a core optimization in agentic frameworks like Semantic Kernel or AutoGen, where tool calls are independent and can be dispatched concurrently.

Exam trap

The trap here is that candidates confuse throughput improvements (like adding tools or increasing token limits) with latency reduction, when the real bottleneck is the sequential dependency of tool calls.

How to eliminate wrong answers

Option A is wrong because increasing the max token limit does not affect the number or sequence of tool calls; it only allows longer responses, which can actually increase latency. Option C is wrong because adding more tools increases the workload and potential sequential calls, worsening latency rather than distributing load in a meaningful way. Option D is wrong because reducing thread history length may free context window space but does not change the sequential execution pattern of tool calls, so it has no direct impact on latency from tool orchestration.

Full explanation →

414

MCQeasy

Your company uses Azure AI Content Safety to moderate user-generated content in a chat application. You need to detect and block sexual content in multiple languages. Which pre-built category should you configure?

A.Sexual

B.Self-harm

C.Hate

D.Violence

AnswerA

The Sexual category directly targets sexual content, satisfying the multilingual detection requirement because Azure AI Content Safety's pre-built categories are language-agnostic, applying trained classifiers across supported languages without per-language configuration. Configuring it blocks such material at the severity threshold you set, meeting the stem's need to detect and block sexual content across multiple languages.

Why this answer

Azure AI Content Safety provides pre-built severity-based categories for content moderation. The 'Sexual' category is specifically designed to detect and block explicit sexual content, including text and images, across multiple languages. This makes it the correct choice for your requirement to moderate sexual content in a chat application.

Exam trap

The trap here is that candidates may confuse 'Sexual' with broader categories like 'Hate' or 'Violence', not realizing that Azure AI Content Safety has a dedicated pre-built category for sexual content with specific detection capabilities.

How to eliminate wrong answers

Option B (Self-harm) is wrong because it focuses on content related to self-injury or suicide, not sexual material. Option C (Hate) is wrong because it targets hate speech based on protected attributes like race or religion, not sexual content. Option D (Violence) is wrong because it detects violent acts or threats, which are distinct from sexual content.

Full explanation →

415

MCQeasy

You need to generate realistic synthetic data using Azure OpenAI Service to train a machine learning model. The data must be diverse and cover edge cases. Which approach should you use?

A.Use prompt engineering with detailed instructions to generate varied examples.

B.Fine-tune the model on a small dataset of real examples.

C.Use Azure OpenAI embeddings to generate similar data points.

D.Set a high temperature parameter only.

AnswerA

Prompt engineering lets you specify tone, format, demographic spread and explicit edge-case scenarios within the instruction, so the model generates varied synthetic examples covering the required diversity. Fine-tuning needs existing data, and templates alone cannot produce the breadth the training set demands.

Why this answer

Prompt engineering with detailed instructions allows you to explicitly control the diversity, structure, and edge-case coverage of generated synthetic data without requiring a pre-existing dataset. By crafting system messages and user prompts that specify variations in attributes, formats, and boundary conditions, you can produce a wide range of realistic examples that mimic real-world distributions, which is essential for training robust machine learning models.

Exam trap

The trap here is that candidates often assume fine-tuning or embeddings are the only ways to generate realistic data, overlooking that prompt engineering with detailed instructions is the most direct and flexible method for producing diverse synthetic data without requiring a pre-existing labeled dataset.

How to eliminate wrong answers

Option B is wrong because fine-tuning on a small dataset of real examples would bias the model toward the limited patterns in that dataset, reducing diversity and failing to generate edge cases, which contradicts the requirement for varied synthetic data. Option C is wrong because Azure OpenAI embeddings are used to measure semantic similarity between text inputs, not to generate new data points; they can retrieve or cluster existing data but cannot produce novel synthetic examples. Option D is wrong because setting a high temperature parameter alone increases randomness in token selection but does not provide the structured control needed to ensure coverage of specific edge cases or diverse scenarios; it may produce incoherent or irrelevant outputs without detailed prompt guidance.

Full explanation →

416

Multi-Selecthard

You are deploying a solution that uses Azure OpenAI Service to generate financial reports. You need to ensure the outputs are accurate and consistent. Which TWO parameters should you adjust? (Choose two.)

Select 2 answers

A.Set presence_penalty to 0.5.

B.Set temperature to 0.

C.Set max_tokens to 2000.

D.Set frequency_penalty to 0.7.

E.Set top_p to 0.1.

AnswersB, E

Low temperature makes output more deterministic.

Why this answer

Setting temperature to 0 makes the model deterministic, always choosing the most likely next token. This is critical for financial reports where consistency and reproducibility are required, as it eliminates randomness in the output.

Exam trap

The trap here is that candidates often confuse parameters that control creativity (temperature, top_p) with those that control repetition (frequency_penalty, presence_penalty) or output length (max_tokens), leading them to select options that do not directly address accuracy and consistency.

Full explanation →

417

MCQhard

A financial services firm uses Azure AI Language to analyze earnings call transcripts. They need to extract key phrases and identify sentiment for each speaker's turn. Which approach should they use?

A.Call the prebuilt sentiment analysis API on the entire transcript

B.Split the transcript by speaker turns and call key phrase extraction and sentiment analysis on each part

C.Use QnA Maker to extract Q&A pairs per speaker

D.Use Text Analytics for health to extract entities and sentiment

AnswerB

Key phrase extraction and sentiment analysis operate per document, so the transcript must be segmented by speaker turn to attribute results correctly. Calling both services on each segment satisfies the stem's requirement to analyse sentiment and key phrases for every individual speaker.

Why this answer

The requirement is to extract key phrases and identify sentiment per speaker turn. The Azure AI Language key phrase extraction and sentiment analysis APIs operate on individual text inputs. By splitting the transcript by speaker turns, each segment can be analyzed independently, providing per-speaker insights.

Processing the entire transcript as a single document would aggregate sentiment and key phrases, losing per-speaker granularity.

Exam trap

The trap here is that candidates may assume the prebuilt sentiment analysis API can handle multi-speaker transcripts by default, but it processes the entire input as one document, so splitting by speaker turns is necessary for per-speaker granularity.

How to eliminate wrong answers

Option A is wrong because calling the prebuilt sentiment analysis API on the entire transcript would return a single overall sentiment score and key phrases for the whole document, not per-speaker turn, failing to meet the requirement for speaker-level analysis. Option C is wrong because QnA Maker (now part of Azure AI Language as custom question answering) is designed to extract question-answer pairs from FAQ-like content, not to perform key phrase extraction or sentiment analysis on conversational transcripts. Option D is wrong because Text Analytics for health is a specialized domain model for extracting medical entities (e.g., diagnoses, medications) and sentiment from clinical notes, not suitable for financial earnings call transcripts.

Full explanation →

418

MCQhard

Refer to the exhibit. You are reviewing a content safety policy for an Azure AI Foundry deployment. The policy rate limits to 20 requests per minute. A user submits 50 requests in one minute. How many requests are allowed?

A.20

B.50

C.100

D.None, all are blocked.

AnswerA

The rate limit caps throughput at 20 requests per minute, so only the first 20 submissions are processed; the remaining 30 are rejected with HTTP 429 responses. This satisfies the policy's stated constraint directly, regardless of how many requests the user submits within that window.

Why this answer

The content safety policy enforces a rate limit of 20 requests per minute. When a user submits 50 requests in one minute, the rate limiter allows only the first 20 requests and blocks the remaining 30. This is a standard token-bucket or sliding-window rate-limiting behavior in Azure AI Foundry, where exceeding the limit results in HTTP 429 (Too Many Requests) for excess requests.

Exam trap

Microsoft often tests the misconception that exceeding a rate limit blocks all requests, when in fact the limit is a threshold that allows the first N requests and denies the rest.

How to eliminate wrong answers

Option B is wrong because it assumes all 50 requests are allowed, ignoring the explicit rate limit of 20 per minute. Option C is wrong because 100 is not related to any limit in the policy; it may confuse the rate limit with a burst or quota value. Option D is wrong because the policy does not block all requests; it allows up to the limit (20) and then blocks the excess.

Full explanation →

419

MCQeasy

You are using Azure AI Search to index customer support tickets. You want to automatically extract the customer's sentiment and key phrases from each ticket. Which Azure AI service should you integrate as a skillset?

A.Azure AI Document Intelligence

B.Azure AI Computer Vision

C.Azure AI Translator

D.Azure AI Language

AnswerD

Azure AI Language provides the sentiment analysis and key phrase extraction capabilities the skillset requires, satisfying the stem's demand for both enrichments from one service. Its built-in Text Analytics skills integrate directly into Azure AI Search indexers, so each ticket is scored and mined for phrases during indexing without custom model code.

Why this answer

Azure AI Language provides pre-built capabilities for sentiment analysis and key phrase extraction, which are exactly the skills needed to process customer support ticket text. When integrated as a skillset in Azure AI Search, it enriches the indexing pipeline by automatically extracting these insights from each document. The other services focus on different modalities (vision, translation, document structure) and do not offer native sentiment or key phrase extraction.

Exam trap

The AI-102 exam often tests the distinction between Azure AI Language (for text analytics) and Azure AI Document Intelligence (for document structure extraction), leading candidates to mistakenly choose Document Intelligence when the task involves analyzing text content rather than extracting form fields.

How to eliminate wrong answers

Option A is wrong because Azure AI Document Intelligence is designed for extracting structured data (like tables, forms, and key-value pairs) from scanned documents, not for analyzing sentiment or extracting key phrases from text. Option B is wrong because Azure AI Computer Vision analyzes images and video, not text content, so it cannot perform sentiment analysis or key phrase extraction on ticket text. Option C is wrong because Azure AI Translator focuses on language translation, not on extracting sentiment or key phrases from the original language text.

Full explanation →

420

MCQeasy

A company is building a chatbot that must handle user queries in multiple languages. The chatbot uses Azure AI Language Service. Which feature should be used to detect the language of incoming messages before routing them to the appropriate language model?

A.Sentiment Analysis

B.Key Phrase Extraction

C.Language Detection

D.Entity Recognition

AnswerC

Language Detection returns the detected language name and ISO code for incoming text, letting the chatbot route each message to the correct language model before processing. It is the prerequisite step; translation and sentiment analysis operate only after the language is known.

Why this answer

Language Detection is the correct feature because it is specifically designed to identify the language of text input, returning a language name and a confidence score. In a multi-language chatbot, this detection step is essential to route the query to the appropriate language-specific model or handler. Azure AI Language Service provides a dedicated pre-built capability for language detection, which can be called via the REST API or SDK.

Exam trap

The trap here is that candidates confuse Language Detection with other text analytics features like Sentiment Analysis or Entity Recognition, assuming any 'analysis' feature can identify language, but only Language Detection is purpose-built for this task.

How to eliminate wrong answers

Option A is wrong because Sentiment Analysis determines the emotional tone (positive, negative, neutral) of text, not the language. Option B is wrong because Key Phrase Extraction identifies important words or phrases in the text, but does not detect the language. Option D is wrong because Entity Recognition identifies named entities like people, places, or organizations, not the language of the text.

Full explanation →

421

MCQhard

You are building a solution that uses Azure AI Language to summarize long customer support emails. The emails are in English, and each email can be up to 10,000 characters. You need to generate a concise summary that preserves the most important sentences and returns the summary along with the original sentences it selected. Which feature and approach should you use?

A.Use key phrase extraction and concatenate the returned phrases into a summary.

B.Use abstractive summarization and send the entire 10,000-character email in a single request.

C.Use extractive summarization and split each email into chunks that respect the per-document character limit before sending requests.

D.Use named entity recognition to extract the main subjects and build a summary from the entities.

AnswerC

Extractive summarization returns the most important sentences from the input, which matches the need to preserve original sentences and know which ones were selected. Because the per-document limit is lower than 10,000 characters, splitting each email into compliant chunks before calling the API avoids request errors while still producing sentence-level summaries.

Why this answer

Extractive summarization returns the most important original sentences, which satisfies the requirement to preserve sentences and identify which ones were selected. Since the per-document character limit is below 10,000 characters, each email must be split into compliant chunks before calling the API to avoid request failures.

Exam trap

The trap here is assuming summarization accepts the same document length as other features, when extractive summarization has a lower per-document character limit that requires chunking.

Full explanation →

422

MCQeasy

A company uses Azure Face API to verify employee identities for building access. They need to ensure that only live faces are used, not photos or videos. Which feature should they enable?

A.Set a high confidence threshold for face matching.

B.Face identification with a large person group.

C.Enable liveness detection using session-based verification.

D.Face detection with attributes such as age and emotion.

AnswerC

Session-based liveness detection challenges the subject to perform a randomised action, then analyses the response to confirm a live person rather than a static photo or replayed video. This satisfies the requirement to reject spoofing attempts during identity verification.

Why this answer

Azure Face API's liveness detection with session-based verification is specifically designed to prevent spoofing attacks using photos, videos, or masks. It analyzes subtle cues such as micro-movements, texture, and depth to confirm the presence of a live person, ensuring that only live faces are accepted for identity verification.

Exam trap

The trap here is that candidates may confuse confidence thresholds or face attributes with liveness detection, not realizing that only session-based verification actively checks for spoofing through motion and depth analysis.

How to eliminate wrong answers

Option A is wrong because setting a high confidence threshold only increases the strictness of face matching scores, but does not differentiate between a live face and a spoofed image or video. Option B is wrong because face identification with a large person group is used to match a detected face against a database of enrolled persons, but it does not verify liveness or detect presentation attacks. Option D is wrong because face detection with attributes like age and emotion extracts demographic and emotional information from a face, but it cannot determine whether the face is live or a reproduction.

Full explanation →

423

MCQhard

You are a generative AI engineer at a financial services company. The company uses Azure OpenAI Service to generate investment summaries. You have deployed a GPT-4 model with a content filter set to 'Low' for hate speech. The model frequently generates summaries that include biased language against certain demographics. You need to reduce biased outputs while maintaining the ability to generate detailed financial analysis. You cannot afford to retrain the model. You have the following options: A) Change the content filter severity to 'High' for all categories, B) Add a system message instructing the model to avoid bias and provide examples of unbiased summaries in the prompt, C) Use the Azure AI Language service to detect bias in the output and regenerate if bias is found, D) Deploy a different model like GPT-3.5 which has less bias. Which course of action should you take?

A.Use the Azure AI Language service to detect bias in the output and regenerate if bias is found.

B.Add a system message instructing the model to avoid bias and provide examples of unbiased summaries in the prompt.

C.Change the content filter severity to 'High' for all categories.

D.Deploy a different model like GPT-3.5 which has less bias.

AnswerB

Adding a system message with explicit anti-bias instructions and unbiased examples steers the model's outputs through prompt engineering, requiring no retraining. Raising the filter to High may block legitimate financial content, while detection and regeneration adds latency and cost.

Why this answer

Adding a system message that explicitly instructs the model to avoid bias and providing few-shot examples of unbiased summaries is the most direct and cost-effective way to steer the model's behavior without retraining. This approach leverages prompt engineering to influence the model's output distribution while preserving the ability to generate detailed financial analysis.

Exam trap

The trap is assuming that content filters can address bias; candidates often pick 'High' severity thinking it will filter biased language, but content filters target harmful content categories, not bias, and may over-block legitimate content.

How to eliminate wrong answers

Option A is wrong because using Azure AI Language to detect bias and regenerate adds latency, cost, and complexity, and it does not prevent biased outputs from being generated in the first place. Option C is wrong because setting content filters to 'High' for all categories may block legitimate financial content and does not specifically target bias; content filters are designed for harmful content categories, not nuanced bias. Option D is wrong because deploying a different model like GPT-3.5 does not guarantee less bias and may degrade the quality of detailed financial analysis.

Full explanation →

424

MCQeasy

You are developing a custom chatbot using Azure AI Bot Service and Language Understanding (CLU). The chatbot needs to escalate to a human agent when the user's sentiment is negative. Which component should you use to detect sentiment?

A.Azure AI Language sentiment analysis

B.Azure Cognitive Search

C.Orchestration workflow

D.QnA Maker

AnswerA

Azure AI Language sentiment analysis returns per-utterance sentiment scores and confidence values, which the bot can evaluate to trigger escalation. CLU handles intent and entity extraction only, not sentiment, so it cannot satisfy the negative-sentiment escalation condition. This component directly meets the requirement to detect sentiment within the conversation flow.

Why this answer

Azure AI Language sentiment analysis is the correct component because it provides pre-built sentiment detection capabilities that analyze text and return sentiment labels (positive, negative, neutral) and confidence scores. This directly meets the requirement to detect negative user sentiment in chatbot conversations, enabling escalation to a human agent when needed.

Exam trap

The trap here is that candidates may confuse Azure Cognitive Search (a search service) with AI Language services, or assume that QnA Maker includes sentiment analysis, when in fact only Azure AI Language provides dedicated sentiment detection.

How to eliminate wrong answers

Option B (Azure Cognitive Search) is wrong because it is designed for indexing and searching documents, not for analyzing sentiment in real-time chat messages. Option C (Orchestration workflow) is wrong because it manages routing between multiple language models or skills, but does not perform sentiment analysis itself. Option D (QnA Maker) is wrong because it is a service for creating question-and-answer knowledge bases from FAQ-like content, and lacks native sentiment detection capabilities.

Full explanation →

425

MCQhard

You are implementing a knowledge mining solution for a legal firm. The solution must ingest large volumes of legal documents (PDFs and Word files) stored in Azure Blob Storage. You need to extract text, recognize named entities (e.g., parties, judges, case numbers), and index the content for full-text search. The solution should also support redaction of sensitive information before indexing. Which combination of Azure AI services should you use?

A.Azure AI Document Intelligence, Azure AI Translator, and Azure AI Search

B.Azure AI Document Intelligence, Azure AI Video Indexer, and Azure AI Search

C.Azure AI Document Intelligence, Azure AI Language, custom skill for redaction, and Azure AI Search

D.Azure AI Document Intelligence, Azure AI Content Safety, and Azure AI Search

AnswerC

Azure AI Document Intelligence extracts text and layout from PDFs and Word files, while Azure AI Language performs named entity recognition for parties, judges and case numbers. A custom skill redacts sensitive content before indexing, satisfying the pre-indexing redaction constraint, and Azure AI Search delivers the required full-text search index.

Why this answer

It combines Azure AI Document Intelligence for OCR and text extraction from PDFs and Word files, Azure AI Language for named entity recognition (e.g., parties, judges, case numbers), a custom skill for redaction (to remove sensitive information before indexing), and Azure AI Search to index the cleaned content for full-text search. This stack directly addresses all requirements: ingestion, entity extraction, redaction, and search indexing.

Exam trap

The trap here is that candidates often confuse Azure AI Content Safety (for moderation) with redaction capabilities, or assume Azure AI Translator can handle entity recognition, when in fact redaction requires a custom skill and entity recognition requires Azure AI Language.

How to eliminate wrong answers

Option A is wrong because Azure AI Translator is a translation service, not designed for named entity recognition or redaction; it would not extract legal entities or support redaction. Option B is wrong because Azure AI Video Indexer is for analyzing video and audio content, not for processing legal documents (PDFs/Word files); it cannot extract text or entities from documents. Option D is wrong because Azure AI Content Safety is for detecting harmful or offensive content (e.g., hate speech, violence), not for recognizing named entities or performing redaction of sensitive information like case numbers or party names.

Full explanation →

426

MCQeasy

You are a data engineer at a university. The university wants to digitize its historical student records (paper forms) to make them searchable. The records are scanned as images (JPEG) and stored in Azure Blob Storage. Each form contains handwritten fields: student name, ID number, date of birth, and degree. You need to extract these fields and index them in Azure AI Search. The solution must use Azure AI Services and minimize manual labeling effort. Which approach should you take?

A.Use Azure AI Custom Vision to train a model to detect handwriting regions, then use Azure AI Vision OCR to read text.

B.Use Azure AI Search with a blob indexer and a skillset that includes OCR skill and Entity Recognition skill.

C.Use Azure AI Document Intelligence to train a custom extraction model with a few labeled samples, then deploy as a custom skill in Azure AI Search.

D.Use Azure AI Vision OCR to extract text from images, then use Azure AI Language to extract entities like name, date, and degree.

AnswerC

Document Intelligence custom extraction models learn from as few as five labelled samples, satisfying the minimise-labelling constraint. Deploying it as a custom skill lets Azure AI Search invoke the model during indexing, so handwritten name, ID, date of birth and degree fields are extracted and mapped into searchable index fields.

Why this answer

Azure AI Document Intelligence (formerly Form Recognizer) is specifically designed to extract structured fields from forms with handwritten text. By training a custom extraction model with a few labeled samples, you minimize manual labeling effort while achieving high accuracy for fields like student name, ID, date of birth, and degree. The model can then be deployed as a custom skill in Azure AI Search to index the extracted data.

Exam trap

The trap here is that candidates often confuse general OCR (Azure AI Vision) with form-specific extraction (Azure AI Document Intelligence), overlooking that Document Intelligence is purpose-built for structured field extraction from forms with minimal labeling.

How to eliminate wrong answers

Option A is wrong because Azure AI Custom Vision is for image classification and object detection, not for handwriting recognition or OCR; it cannot extract text from handwritten fields. Option B is wrong because Azure AI Search's built-in OCR skill extracts raw text but lacks the ability to identify specific fields like student name or degree without additional custom logic, and Entity Recognition skill is designed for named entities in text, not for form field extraction. Option D is wrong because Azure AI Vision OCR extracts all text from an image but does not parse it into structured fields; Azure AI Language's entity recognition would require additional post-processing and manual mapping to identify specific form fields, increasing effort.

Full explanation →

427

Multi-Selecteasy

You are developing an Azure AI solution that uses pre-built models from Azure AI Vision to analyze images. The solution must be able to detect objects and read printed text. Which TWO capabilities should you use?

Select 2 answers

A.OCR (legacy)

B.Facial detection

C.Object detection

D.Image tagging

E.Read (OCR)

AnswersC, E

Object detection returns bounding boxes and labels for multiple objects within an image, directly satisfying the requirement to detect objects. Azure AI Vision's pre-built model provides this without training, meeting the scenario's use of standard capabilities.

Why this answer

Object detection (C) is correct because it is the Azure AI Vision pre-built capability that locates and classifies multiple objects within an image, returning bounding boxes and labels, which directly satisfies the requirement to detect objects. Read (OCR) (E) is correct because it is the modern Azure AI Vision OCR engine that extracts printed and handwritten text from images and documents, satisfying the requirement to read printed text. The legacy OCR (A) option is an older, deprecated recognition model with weaker accuracy and limited language support, so it is not the recommended choice for new solutions.

Facial detection (B) only returns face locations and attributes and does not detect general objects or read text, and image tagging (D) produces descriptive labels for the overall image rather than object locations or extracted text, so neither meets the stated requirements.

Exam trap

The trap here is that candidates often confuse Image Tagging (which only provides labels) with Object Detection (which provides both labels and spatial localization), and may mistakenly choose the legacy OCR API instead of the modern Read API for text extraction.

Full explanation →

428

MCQeasy

You are implementing a generative AI solution using Azure OpenAI. You need to ensure that the model's outputs do not contain certain inappropriate words or phrases. Which feature should you configure?

A.System message instructions

B.Grounding with your data

C.Content filters

D.Max tokens limit

AnswerC

Content filters in Azure OpenAI evaluate both prompts and completions against configurable severity thresholds across categories such as hate, violence, and sexual content, blocking inappropriate words or phrases. This directly enforces the required output restriction without retraining or prompt engineering.

Why this answer

Content filters in Azure OpenAI are specifically designed to detect and block inappropriate words or phrases in both prompts and completions. They operate at the service level, applying configurable severity thresholds for categories like hate, violence, sexual content, and self-harm, ensuring model outputs adhere to policy without requiring prompt engineering or data modifications.

Exam trap

Microsoft often tests the misconception that system messages (Option A) are sufficient for content safety, when in fact they are only behavioral guidelines and lack the enforcement mechanism of dedicated content filters.

How to eliminate wrong answers

Option A is wrong because system message instructions guide model behavior and tone but cannot reliably enforce content restrictions; they are advisory and can be overridden by the model, especially in edge cases. Option B is wrong because grounding with your data (using Azure Cognitive Search) augments prompts with your own data for relevance and accuracy, but it does not filter or block inappropriate content from the model's generated responses. Option D is wrong because the max tokens limit controls the length of the output, not its content; it cannot prevent the model from generating inappropriate words or phrases within the allowed token count.

Full explanation →

429

Multi-Selecthard

You are designing an agentic solution using Azure AI Agent Service. The agent needs to perform actions on behalf of users, such as sending emails and updating databases. The solution must use managed identities for authentication to Azure resources. Which TWO configurations are required?

Select 2 answers

A.Store connection strings in Azure Key Vault and reference them in the agent's configuration

B.Create a service principal in Microsoft Entra ID and assign RBAC roles to the agent's resource

C.Use DefaultAzureCredential in the agent's code to authenticate to Azure services

D.Configure the agent to use an API key for each external service

E.Assign a system-assigned managed identity to the Azure resource hosting the agent

AnswersC, E

DefaultAzureCredential chains managed identity credentials automatically, so the agent authenticates to Azure resources without stored secrets — satisfying the stem's managed identity requirement. In Azure AI Agent Service, this credential resolves the assigned identity at runtime, enabling email and database actions on behalf of users without embedding connection strings or service principal keys.

Why this answer

Option E is correct because a system-assigned managed identity must first be enabled on the Azure resource hosting the agent (for example, the Azure AI Foundry/Azure AI Services resource or the compute running the agent code) so that Microsoft Entra ID can issue tokens for that resource without storing secrets. Option C is correct because DefaultAzureCredential is the recommended credential chain in the Azure Identity SDK; it automatically picks up the managed identity (via ManagedIdentityCredential/EnvironmentCredential) when running in Azure, allowing the agent code to authenticate to Azure services such as Azure SQL, Storage, or Microsoft Graph without embedding credentials. Option A is not required and contradicts the managed-identity requirement, since connection strings/secrets in Key Vault are a secret-based pattern rather than identity-based authentication.

Option B is not required because managed identities are service principals managed by the platform; you do not manually create a service principal in Microsoft Entra ID, and RBAC role assignments are made to the managed identity's principal, not to a separately created app registration. Option D is incorrect because API keys are static secrets and the scenario explicitly mandates managed identities for authentication to Azure resources.

Exam trap

The trap here is that candidates often confuse managed identities with service principals or API keys, thinking they need to create a separate service principal or store connection strings, when in fact managed identities are automatically managed service principals that require only RBAC role assignments and the use of DefaultAzureCredential (or ManagedIdentityCredential) in code.

Full explanation →

430

MCQmedium

A company wants to build a solution that automatically generates alt text for images on their website to improve accessibility. The alt text must be a concise, human-readable description of the image content. Which Azure AI Vision feature should they use?

A.Custom Vision with a classification model trained on descriptive text

B.Image Analysis with the Detect Objects feature

C.Image Analysis with the Caption feature

D.Image Analysis with the Tags feature

AnswerC

The Caption feature generates a concise, human-readable description of an image, which is exactly what is needed for alt text. It is a prebuilt model that requires no training and returns a single sentence describing the image. This directly meets the accessibility requirement.

Why this answer

The Caption feature in Image Analysis is designed to generate a one-sentence description of an image, which is ideal for alt text. It is prebuilt and requires no training. Other features like Tags or Detect Objects provide keywords or bounding boxes but not a coherent description, making them less suitable for accessibility purposes.

Exam trap

The trap here is confusing Tags with Caption; Tags give keywords, while Caption gives a full sentence suitable for alt text.

Full explanation →

431

MCQmedium

A developer uses the Azure OpenAI API to generate code. They want to ensure that the generated code is in Python. Which parameter should they set?

A.temperature

B.top_p

C.system message

D.max_tokens

AnswerC

The system message sets the model's behavioural context before any user turn, so instructing it there to respond only in Python reliably constrains the generated code's language. A user message or temperature setting cannot enforce this persistent constraint.

Why this answer

The system message is used to set the behavior and context of the AI model, including specifying the desired output format or language. By setting the system message to 'You are a helpful assistant that always writes code in Python', the developer can instruct the model to generate Python code consistently. This parameter is part of the chat completions API and directly influences the model's persona and constraints.

Exam trap

Microsoft often tests the distinction between parameters that control output randomness (temperature, top_p) and those that control output structure or behavior (system message), leading candidates to mistakenly choose temperature or top_p for language specification.

How to eliminate wrong answers

Option A is wrong because temperature controls the randomness of the output, not the language or format of the generated code. Option B is wrong because top_p (nucleus sampling) controls the cumulative probability threshold for token selection, which affects diversity but does not specify the output language. Option D is wrong because max_tokens limits the length of the generated response, not the programming language or content type.

Full explanation →

432

MCQmedium

You are building a generative AI solution with Azure OpenAI Service that must produce concise, factual answers using only a supplied knowledge base. During testing, the model sometimes invents details not present in the knowledge base. You need to reduce hallucinations without retraining the model. What should you do?

A.Increase the max_tokens setting so the model has more room to explain its reasoning.

B.Switch to a larger model deployment with a bigger context window.

C.Lower the temperature and include a system message that instructs the model to answer only from the provided context and to say it does not know when the answer is absent.

D.Fine-tune the base model on the knowledge base so it memorizes the facts.

AnswerC

Lowering temperature reduces sampling randomness, and a system message that constrains the model to the supplied context encourages grounded answers and an explicit 'I don't know' fallback. Together these prompt-engineering techniques reduce hallucination without retraining, and they work within the existing deployment. This directly addresses invented details while keeping the model unchanged.

Why this answer

Hallucinations in grounded generation are commonly reduced with prompt engineering: lower temperature to make output more deterministic, and a system message that restricts answers to the provided context and instructs the model to admit when the answer is not present. These changes need no retraining and can be applied immediately to the deployment.

Exam trap

The trap here is reaching for fine-tuning or a larger model to fix factual errors, when the scenario explicitly rules out retraining and the issue is prompt grounding.

Full explanation →

433

MCQmedium

A company wants to generate personalized product descriptions for its e-commerce site using Azure OpenAI. They need to ensure the model's output adheres to brand guidelines and does not generate prohibited content. Which approach should they use?

A.Use a system message with brand guidelines and apply content filtering.

B.Use prompt engineering with negative prompts and ignore content filtering.

C.Provide few-shot examples in the user message and rely on the model's training.

D.Fine-tune the model with brand guidelines and disable content filtering for performance.

AnswerA

A system message sets persistent behavioural instructions, so brand tone and prohibited-topic rules apply to every completion, while Azure OpenAI's content filter blocks harmful categories independently. Together they satisfy both the brand-guideline and prohibited-content constraints.

Why this answer

Using a system message allows you to embed brand guidelines directly into the conversation context, instructing the model on tone, style, and prohibited content. Azure OpenAI's content filtering provides an additional safety layer by automatically detecting and blocking harmful or policy-violating outputs, ensuring compliance with both brand and regulatory requirements.

Exam trap

Microsoft often tests the misconception that fine-tuning or prompt engineering alone is sufficient for safety and compliance, when in reality Azure OpenAI requires explicit content filtering and system messages to enforce brand guidelines reliably.

How to eliminate wrong answers

Option B is wrong because ignoring content filtering removes the safety guardrails that prevent prohibited content, and negative prompts alone are unreliable for enforcing brand guidelines. Option C is wrong because few-shot examples in the user message do not guarantee consistent adherence to brand guidelines across all outputs, and relying solely on the model's training ignores the need for explicit content filtering. Option D is wrong because disabling content filtering for performance sacrifices safety and compliance, and fine-tuning alone cannot dynamically enforce brand guidelines as effectively as a system message combined with content filtering.

Full explanation →

434

MCQeasy

You want to use the Azure AI Language service to summarize long customer support conversations into a short summary. Which feature should you use?

A.Sentiment Analysis

B.Conversational Summarization

C.Entity Extraction

D.Key Phrase Extraction

AnswerB

Conversational Summarization is purpose-built for multi-turn dialogue, extracting issues and resolutions across speaker turns rather than treating text as one block. It satisfies the stem's requirement to condense long customer support conversations, unlike document or text summarization, which assume unstructured prose without speaker roles.

Why this answer

Conversational Summarization is the correct feature because it is specifically designed to condense multi-turn dialogues, such as customer support conversations, into concise summaries. Unlike generic text summarization, it understands the conversational flow, speaker turns, and context to produce a coherent summary of the interaction.

Exam trap

The trap here is that candidates often confuse Key Phrase Extraction or Entity Extraction with summarization, but those features only extract discrete items rather than generating a flowing summary of the entire conversation.

How to eliminate wrong answers

Option A is wrong because Sentiment Analysis only detects positive, negative, neutral, or mixed sentiment in text, not the overall summary of a conversation. Option C is wrong because Entity Extraction identifies named entities like people, places, or dates, but does not generate a condensed summary of the dialogue. Option D is wrong because Key Phrase Extraction returns a list of important words or phrases, not a coherent narrative summary of the conversation.

Full explanation →

435

MCQmedium

Refer to the exhibit. { "content_filters": [ { "type": "hate", "action": "block", "severity": "high" }, { "type": "sexual", "action": "block", "severity": "medium" }, { "type": "self_harm", "action": "block", "severity": "low" } ] } You deploy an Azure OpenAI model with the above content filter configuration. A user submits a prompt that the system rates as "hate" at severity level "medium". What happens?

A.The prompt is allowed because the severity is below the threshold.

B.The prompt is blocked because hate content is detected.

C.The prompt is blocked because the severity is medium.

D.The prompt is allowed because the hate filter is not configured for medium.

AnswerA

The hate filter blocks at severity high, so medium-severity hate content falls below that threshold and passes. The action applies only when the rated severity meets or exceeds the configured level, so the prompt is not blocked.

Why this answer

The content filter configuration blocks 'hate' content only at severity level 'high'. Since the user's prompt was rated as 'hate' at severity 'medium', it falls below the configured threshold and is allowed. Azure OpenAI content filters evaluate severity levels (low, medium, high) and apply the configured action only when the detected severity meets or exceeds the specified threshold.

Exam trap

A common mistake in Azure exams is assuming that any detection of a content type (e.g., hate) automatically triggers the configured action, ignoring that only severity levels meeting or exceeding the threshold are blocked.

How to eliminate wrong answers

Option B is wrong because the filter does not block all hate content indiscriminately; it only blocks hate content at severity 'high' or above. Option C is wrong because the severity 'medium' is below the configured threshold of 'high' for the hate filter, so it is not blocked. Option D is wrong because the hate filter is indeed configured (with action 'block' and severity 'high'), but it simply does not apply to severity 'medium'.

Full explanation →

436

MCQmedium

You are configuring an agent in Azure AI Foundry Agent Service to generate responses based on a large set of internal documents. The documents are stored in an Azure Storage account. You need to ensure the agent can retrieve relevant information from these documents to answer user queries. What should you do?

A.Create an Azure AI Search index that contains the documents and connect it to the agent as a knowledge source.

B.Upload the documents to the agent's file storage and enable the file search tool.

C.Use the Azure AI Foundry SDK to embed the documents into the agent's prompt by concatenating their contents.

D.Configure the agent to use a custom tool that calls the Azure Storage REST API to fetch documents on demand.

AnswerA

Azure AI Search is designed to index and query large volumes of documents. By creating an index and connecting it to the agent, you enable the agent to retrieve relevant information based on user queries. This approach scales well and provides advanced search capabilities like semantic ranking.

Why this answer

For large document sets, Azure AI Search provides scalable indexing and retrieval. By creating an index and connecting it as a knowledge source, the agent can query the index and retrieve relevant passages. This is the recommended approach for grounding agent responses in extensive internal documents.

Exam trap

The trap here is assuming that file upload or direct concatenation can handle large document sets, but they are limited by size and token constraints.

Full explanation →

437

MCQhard

Your company uses Azure OpenAI Service to generate product descriptions. You need to ensure that the generated content does not include offensive language and adheres to responsible AI principles. What should you implement?

A.Enable customer-managed key encryption

B.Configure content filters in Azure OpenAI

C.Fine-tune the model with a curated dataset

D.Set usage limits and throttling

AnswerB

Configuring content filters in Azure OpenAI applies severity-based screening across hate, violence, sexual and self-harm categories on both prompts and completions, blocking offensive output before it reaches users. This directly satisfies the stem's requirement to prevent offensive language and uphold responsible AI principles, unlike prompt engineering or moderation applied after generation.

Why this answer

Content filters in Azure OpenAI allow you to define categories (e.g., hate, violence, self-harm) and severity levels (low, medium, high) to automatically block or flag offensive language in generated outputs. This directly enforces responsible AI principles by preventing harmful content from being surfaced to users, without requiring model retraining or encryption changes.

Exam trap

The trap here is that candidates often confuse data security controls (like encryption or throttling) with content safety controls, assuming any 'security' feature can filter offensive language, when in fact only purpose-built content filters can analyze and block harmful text in real time.

How to eliminate wrong answers

Option A is wrong because customer-managed key encryption (CMK) protects data at rest but does not inspect or filter the semantic content of model outputs for offensive language. Option C is wrong because fine-tuning with a curated dataset can reduce but not guarantee the absence of offensive outputs; it cannot dynamically block real-time content violations and requires ongoing dataset maintenance. Option D is wrong because usage limits and throttling control API request rates and quotas, not the quality or safety of the generated text.

Full explanation →

438

MCQhard

You are deploying a generative AI solution that must produce structured JSON order confirmations from free-text customer emails. Downstream systems reject any response that is not valid JSON matching a fixed schema. Which approach best guarantees schema-conformant output from an Azure OpenAI chat completion?

A.Set the temperature to zero and include an example of the exact JSON output in the prompt.

B.Use a structured outputs configuration with a strict JSON schema that defines required fields and types.

C.Set response_format to json_object and describe the desired schema in the system message.

D.Request the output in JSON and run a regex-based validator that repairs any deviations before forwarding.

AnswerB

Structured outputs constrain decoding so the generated tokens conform to a supplied JSON schema, including required properties and allowed types. This provides a strong guarantee that the response validates against the schema, which is exactly what downstream systems demand. Combined with a clear prompt, it removes the need for retry loops or brittle post-processing, making it the most reliable option for strict schema conformance.

Why this answer

Structured outputs constrain token generation to a supplied JSON schema, enforcing required fields and correct types so responses validate against the contract. JSON mode only guarantees syntactic validity, and few-shot examples or post-processing repairs do not provide a hard structural guarantee. For downstream systems that reject any schema deviation, constrained decoding is the most reliable approach.

Exam trap

The trap here is equating JSON mode with schema enforcement, when JSON mode only guarantees valid JSON syntax.

Full explanation →

439

MCQmedium

Your application needs to extract key phrases from customer reviews to identify common topics. Which Azure AI Language feature should you use?

A.Sentiment Analysis

B.Language Detection

C.Named Entity Recognition

D.Key Phrase Extraction

AnswerD

Key Phrase Extraction identifies the main talking points in unstructured text, returning salient terms and phrases. Applied to customer reviews, it surfaces recurring topics directly, matching the stated requirement without needing custom model training or entity categorisation.

Why this answer

Key Phrase Extraction is the correct Azure AI Language feature because it is specifically designed to identify and return a list of key phrases from unstructured text, such as customer reviews, that capture the main topics and themes. This allows you to aggregate common topics across multiple reviews without manual analysis.

Exam trap

The trap here is that candidates may confuse Named Entity Recognition (NER) with Key Phrase Extraction because both deal with extracting information from text, but NER focuses on predefined entity types (e.g., persons, locations) while Key Phrase Extraction identifies any significant topic or phrase relevant to the document's content.

How to eliminate wrong answers

Option A is wrong because Sentiment Analysis determines the overall emotional tone (positive, negative, neutral, or mixed) of text, not the extraction of topics or key phrases. Option B is wrong because Language Detection identifies the language in which the text is written (e.g., English, Spanish), which is unrelated to extracting topic-specific phrases. Option C is wrong because Named Entity Recognition (NER) identifies and categorizes named entities like people, organizations, locations, and dates, but does not extract general key phrases or topics from the text.

Full explanation →

440

MCQeasy

You are designing an Azure AI solution that uses Azure AI Language to analyze customer support transcripts. The solution must identify key phrases, detect sentiment, and extract custom entities specific to your product catalog. Which two Azure AI Language features should you enable?

A.PII Detection

B.Key Phrase Extraction

C.Summarization

D.Custom Entity Extraction

AnswerB, D

Key Phrase Extraction satisfies the requirement to identify key phrases within customer support transcripts. It returns salient noun phrases from unstructured text via the Microsoft Entra ID–authenticated Language resource, complementing sentiment analysis and custom entity extraction. It does not, however, cover sentiment or custom entities, so it is only one of the two features required.

Why this answer

Key Phrase Extraction (Option B) is correct because it identifies the main points and important terms in customer support transcripts, such as 'refund request' or 'account issue,' which directly supports analyzing the content. Custom Entity Extraction (Option D) is correct because it allows you to define and extract domain-specific entities from your product catalog, such as product names or model numbers, using a trained custom entity extraction model. Together, these two features enable both general insight extraction and tailored, product-specific data extraction from the transcripts.

Exam trap

Microsoft often tests the distinction between pre-built features (like Key Phrase Extraction) and custom features (like Custom Entity Extraction), and the trap here is that candidates may incorrectly choose Summarization or PII Detection because they sound relevant to 'analyzing transcripts,' but they do not fulfill the specific requirements of key phrase identification and custom entity extraction.

How to eliminate wrong answers

Option A is wrong because PII Detection is designed to identify and redact personally identifiable information (e.g., names, phone numbers, credit card numbers) for privacy compliance, not to analyze key phrases or extract custom product entities. Option C is wrong because Summarization generates a concise summary of the transcript's main points, which is useful for overview but does not perform key phrase identification or custom entity extraction as required by the question.

Full explanation →

441

MCQeasy

You run the above Azure CLI command. What is the expected output?

A.The primary and secondary keys along with the endpoint

B.The primary and secondary keys

C.A list of endpoints for the service

D.An error because the command is incorrect

AnswerB

The Azure CLI command regenerates or lists the resource's access keys, so the output contains the primary and secondary keys. These credentials authenticate requests to the Azure AI service, matching the expected key pair returned by the operation.

Why this answer

The Azure CLI command `az cognitiveservices account keys list` retrieves only the primary and secondary API keys for the Cognitive Services account. The endpoint is not included in the output; it must be obtained separately using `az cognitiveservices account show`. Therefore, the expected output contains the primary key and secondary key only.

Exam trap

The trap is that candidates may assume `az cognitiveservices account keys list` returns the endpoint along with the keys, but it only returns the primary and secondary keys. The endpoint is obtained via `az cognitiveservices account show`.

How to eliminate wrong answers

Option A is wrong because the command does not return the endpoint; it only returns the keys. Option C is wrong because the command returns keys, not a list of endpoints. Option D is wrong because the command is syntactically correct and will execute successfully.

Full explanation →

442

MCQmedium

Your company uses Azure AI Document Intelligence to process invoices. You need to extract the invoice date and total amount. Which model should you use?

A.Read model

B.Prebuilt invoice model

C.Layout model

D.General document model

AnswerB

The prebuilt invoice model returns structured fields including InvoiceDate and InvoiceTotal, directly satisfying the requirement to extract those two values without custom training. Its pretrained schema covers common invoice layouts, so no labelled dataset or template definition is needed, unlike the general document or custom extraction models.

Why this answer

The Prebuilt invoice model is specifically trained on thousands of invoice documents to extract key fields like invoice date, total amount, vendor details, and line items. It uses deep learning models optimized for invoice layouts, providing higher accuracy and structured output for these fields compared to general models.

Exam trap

The trap here is that candidates often confuse the Layout model's ability to extract text and tables with the specialized field extraction of prebuilt models, leading them to choose the Layout model for invoice data extraction when a purpose-built model exists.

How to eliminate wrong answers

Option A is wrong because the Read model is designed for extracting printed and handwritten text from documents, not for identifying structured fields like invoice date or total amount; it returns raw text without semantic key-value extraction. Option C is wrong because the Layout model focuses on extracting text, tables, and selection marks with spatial relationships, but it does not predefine or extract specific invoice fields like total amount. Option D is wrong because the General document model extracts key-value pairs and entities from unstructured documents, but it is not specialized for invoices and may miss or mislabel critical fields like invoice date and total amount without custom training.

Full explanation →

443

MCQhard

You are building a generative AI solution using Azure OpenAI Service that must generate code snippets. The solution must ensure that the generated code does not include any known vulnerable patterns, such as SQL injection. You need to implement a safeguard. What should you do?

A.Implement a post-processing step that analyzes the generated code with a static analysis tool and rejects or sanitizes unsafe snippets.

B.Add a system message instructing the model to never generate vulnerable code.

C.Fine-tune the model on a dataset of secure code to reduce the likelihood of generating vulnerable patterns.

D.Use Azure AI Content Safety to scan the generated code for vulnerabilities and block unsafe outputs.

AnswerA

A static analysis tool can detect known vulnerable patterns like SQL injection in generated code. By post-processing the model's output, you can enforce security policies before the code is used. This is a robust safeguard that complements the generative model and ensures only safe code is accepted. It is a standard practice in secure code generation.

Why this answer

The most reliable safeguard is to post-process the generated code with a static analysis tool that detects vulnerable patterns like SQL injection. This provides deterministic enforcement, rejecting or sanitizing unsafe outputs before use. While prompt instructions and fine-tuning can help, they do not guarantee security.

Static analysis is a standard, effective method for ensuring code safety.

Exam trap

The trap here is assuming that content safety filters or prompt instructions can reliably prevent code vulnerabilities, when dedicated static analysis is needed.

Full explanation →

444

MCQeasy

You are a content moderator for a social media platform that uses Azure Content Moderator. The platform has a custom blocklist of URLs (e.g., 'example.com/spam') and a custom term list for hate speech. Recently, users have been posting comments that contain a new form of hate speech not yet in the term list. The comments are being allowed through moderation. You need to update the solution to catch these new phrases as quickly as possible. What should you do?

A.Create a new custom model using Custom Vision to detect the new phrases in text

B.Retrain the image classification model using the new phrases as training data

C.Add the new phrases to the existing custom term list using the List Management API

D.Delete the existing term list and recreate it with the new phrases included

AnswerC

The custom term list is the exact-match mechanism Content Moderator applies to text, so adding the new phrases through the List Management API makes them detectable on the next submission. This is the fastest path because term lists update immediately, with no model retraining or redeployment required to catch the emerging hate speech.

Why this answer

Azure Content Moderator's custom term lists allow you to dynamically add new offensive terms or phrases via the List Management API, which immediately updates the moderation screening without retraining or redeploying any model. This provides the fastest way to catch new hate speech patterns as they emerge, as the term list is checked in real-time during content review.

Exam trap

The trap here is that candidates may assume retraining or creating a new model is required for new patterns, but Azure Content Moderator's term lists are designed for rapid, rule-based updates without the overhead of model training, and the List Management API enables immediate addition of terms to an existing list.

How to eliminate wrong answers

Option A is wrong because Custom Vision is designed for image classification, not text phrase detection, and cannot be used to identify hate speech in text comments. Option B is wrong because image classification models are irrelevant to text-based hate speech; retraining such a model would not affect text moderation. Option D is wrong because deleting and recreating the term list is unnecessary and slower; the List Management API supports adding new terms to an existing list without disruption, preserving any existing terms and avoiding downtime.

Full explanation →

445

MCQhard

You are implementing a knowledge mining solution using Azure AI Search. The data source is a large Azure Cosmos DB collection containing customer support tickets. Each ticket has fields: ticket_id, description, category, and resolution. You need to ensure that the search index can support fuzzy search and autocomplete suggestions. What should you configure in the index definition?

A.Set the 'searchable' attribute on the description field and define a suggester

B.Set the 'filterable' attribute on the description field

C.Set the 'sortable' attribute on the ticket_id field

D.Set the 'facetable' attribute on the category field

AnswerA

Marking the description field searchable enables full-text tokenisation, which underpins fuzzy matching, while a suggester builds the dedicated autocomplete and suggestions structures. Both are required because fuzzy search and autocomplete read from different index constructs, and the stem demands support for both.

Why this answer

Fuzzy search requires the 'searchable' attribute on fields to enable full-text search, and autocomplete suggestions require a 'suggester' configured on the index. The suggester defines which fields are used to generate suggestion candidates, and the 'searchable' attribute allows the description field to be tokenized and matched against partial or misspelled queries.

Exam trap

The trap here is that candidates often confuse 'searchable' with 'filterable' or 'facetable', thinking any attribute that enables querying will also support fuzzy search and autocomplete, but only 'searchable' fields are analyzed and tokenized for these features, and a suggester is a separate required configuration.

How to eliminate wrong answers

Option B is wrong because the 'filterable' attribute is used for exact match filtering (e.g., category equals 'billing'), not for fuzzy search or autocomplete; it does not enable partial or approximate matching. Option C is wrong because the 'sortable' attribute on ticket_id only allows ordering results by that field, which has no relevance to fuzzy search or autocomplete suggestions. Option D is wrong because the 'facetable' attribute on category enables faceted navigation (e.g., drill-down counts), but does not support fuzzy matching or suggestion generation.

Full explanation →

446

Multi-Selecthard

An agent uses Azure AI Language to perform sentiment analysis on customer feedback. The team notices that the sentiment scores are sometimes inaccurate for negative feedback. Which TWO improvements should the team consider?

Select 2 answers

A.Pre-process the text to handle negations and sarcasm.

B.Use a custom sentiment analysis model trained on domain-specific data.

C.Switch to a different language model without fine-tuning.

D.Increase the number of decimal places in the sentiment score.

E.Increase the confidence threshold for positive sentiment.

AnswersA, B

Negation and sarcasm invert surface polarity, so the model scores positive words as positive despite negative intent. Pre-processing rewrites or flags these constructions before scoring, directly addressing the inaccurate negative-feedback scores described in the stem.

Why this answer

Option A is correct because Azure AI Language's prebuilt sentiment analysis can misclassify negations (e.g., "not good") and sarcasm, so pre-processing the text to normalize or flag these constructs improves accuracy. Option B is correct because training a custom sentiment analysis model on domain-specific labeled data lets the service learn industry-specific vocabulary and phrasing, which typically boosts accuracy for specialized feedback. Option C is not appropriate because simply swapping language models without fine-tuning does not address domain-specific or negation/sarcasm issues.

Option D is wrong because the number of decimal places in the score is a formatting detail and does not affect model accuracy. Option E is wrong because raising a positive-sentiment confidence threshold only changes classification cutoffs and does not improve the underlying sentiment detection for negative feedback.

Exam trap

The trap here is that candidates often assume that increasing precision or switching models generically will fix inaccuracies, rather than recognizing that domain-specific fine-tuning and text pre-processing are the standard Azure AI Language approaches to handle linguistic edge cases like negations and sarcasm.

Full explanation →

447

MCQmedium

You are building a web application that lets users upload photos of restaurant menus and receive the extracted text. The menus are often photographed at an angle, with uneven lighting and background clutter. You need an Azure AI Vision capability that returns text lines and words with bounding boxes, and you want to minimize development effort. Which Azure AI Vision feature should you use?

A.Azure AI Custom Vision object detection model

B.Azure AI Vision Image Analysis with the Tags feature

C.Azure AI Face API with the OCR attribute enabled

D.Azure AI Vision OCR (Read) API

AnswerD

The Read API is designed for extracting printed and handwritten text from images, including photos taken at angles or with background noise. It returns text lines, words, and bounding boxes in a structured JSON response, so you can parse and display results with minimal custom code. It directly matches the need for menu text extraction without training a custom model.

Why this answer

The Read API in Azure AI Vision is purpose-built for OCR and returns text lines and words with bounding boxes, making it ideal for extracting menu content from photos. It handles real-world conditions like skew and clutter. The other services either return high-level labels, detect objects, or analyze faces, none of which extract the literal text required by the application.

Exam trap

The trap here is assuming that Image Analysis tags or Custom Vision can read text, when only the dedicated Read OCR capability returns the actual characters and their positions.

Full explanation →

448

MCQeasy

You run the PowerShell script shown to audit your Azure AI Agent Service agents. The script outputs that several agents have no tools configured. What is the impact on those agents?

A.The agents cannot be deployed until tools are added

B.The agents can only respond to queries using the model's built-in knowledge, without ability to perform actions

C.The agents cannot start conversations with users

D.The agents will use default tools provided by Azure

AnswerB

Without tools configured, an agent has no function-calling or action capabilities, so it is limited to generating responses from the underlying model's built-in knowledge. It cannot retrieve external data or perform actions on the user's behalf.

Why this answer

Azure AI Agent Service agents without tools configured rely solely on the model's built-in knowledge (e.g., GPT-4o's training data) to generate responses. They cannot execute external actions like calling APIs, querying databases, or running code, which are enabled only when tools (e.g., code interpreter, function calling, or Azure Functions) are explicitly attached. This is by design: tools extend the agent's capabilities beyond the model's static knowledge.

Exam trap

The trap here is that candidates assume agents must have tools to be functional or deployed, but Azure AI Agent Service allows tool-less agents that operate as pure language models, and the exam tests understanding that tools are optional for basic Q&A but required for action-oriented tasks.

How to eliminate wrong answers

Option A is wrong because agents without tools can still be deployed and will function, but with limited capabilities—they simply lack action execution. Option C is wrong because agents can start conversations with users regardless of tool configuration; conversation initiation is controlled by the agent's trigger (e.g., user message or event), not by tool presence. Option D is wrong because Azure does not assign default tools to agents; tools must be explicitly defined in the agent's configuration or via the `tools` parameter in the Azure AI Agent Service SDK.

Full explanation →

449

MCQeasy

You are deploying a generative AI application using Azure OpenAI Service. The application must generate responses that are creative and varied for a marketing campaign. Which parameter should you adjust to increase the randomness of the model's output?

A.Temperature

B.Presence penalty

C.Top-p (nucleus sampling)

D.Frequency penalty

AnswerA

Temperature controls the randomness of the model's output. Higher values (e.g., 1.0 or above) make the output more varied and creative, while lower values make it more deterministic. For a marketing campaign requiring creative and varied responses, increasing the temperature is the correct adjustment.

Why this answer

Temperature is the parameter that directly controls the randomness of the model's output. Higher temperatures lead to more diverse and creative responses, which is suitable for a marketing campaign that requires varied content. Other parameters like top-p and penalties affect diversity in different ways but are not the primary control for randomness.

Exam trap

The trap here is confusing parameters that increase diversity (like top-p or penalties) with the primary parameter for randomness, which is temperature.

Full explanation →

450

MCQhard

You are troubleshooting a Microsoft Copilot Studio agent that calls a custom connector action to retrieve shipment data. The action works in the test canvas, but when the agent is published to a channel, the action returns an authorization error for some users. The connector uses OAuth. You need to resolve the issue. What should you do?

A.Increase the connector's timeout setting so authorization calls have more time to complete.

B.Republish the agent to the channel so the connector definition refreshes for all users.

C.Verify that each user has consented to the OAuth connection and has permission in the source system, then have them sign in to the connection.

D.Change the connector authentication to API key and share the key with all users.

AnswerC

OAuth connections are user-scoped, so each user must consent and hold the necessary permissions in the source system. Users who fail are likely missing consent or lack rights in the shipment system. Having them authenticate and confirming their permissions resolves the authorization errors while preserving per-user security and auditing.

Why this answer

OAuth-protected connectors require each user to consent and to have appropriate rights in the target system. The test canvas often runs under the maker's credentials, which masks per-user authorization problems. Confirming consent and permissions, and having affected users sign in to the connection, fixes the errors without weakening the security model.

Exam trap

The trap here is assuming the connector itself is broken because it works in the test canvas, when the canvas often uses the maker's credentials rather than each end user's.

Full explanation →

Microsoft Azure AI Engineer Associate AI-102 (AI-102) — Questions 376–450