AI-103 Practice Test 18 — 15 Questions

Question 1

You are processing large PDF documents. You only need the text in the table of contents. What is the most cost-effective approach?

Accepted Answer

Use the 'pages' query parameter. The 'pages' query parameter in Azure AI Document Intelligence allows you to specify which pages to process, such as a single page or a range. If you only need the table of contents, you can identify the page number(s) containing it and pass the 'pages' parameter to limit processing to those pages. This reduces the number of pages processed, thereby lowering cost, since Document Intelligence pricing is based on the number of pages analyzed.

Answer

Use a custom template model

Answer

Split the PDF into individual pages before uploading

Answer

Process the entire PDF

Question 2

You want to monitor the cost and latency of your agentic application in production. Which service should you integrate with your Azure AI Foundry project?

Accepted Answer

Azure Application Insights.. Azure Application Insights is the Application Performance Monitoring (APM) service that Azure AI Foundry integrates with to capture traces, latency, token usage, and cost telemetry for agentic applications. It provides the dashboards, query language (KQL), and alerting needed to monitor production behavior.

Answer

Azure OpenAI Studio.

Answer

Azure AI Search.

Answer

Azure AI Content Safety.

Question 3

You are managing an enterprise Azure AI Search deployment. During an audit, you notice that index data is encrypted at rest using Microsoft-managed keys. Management mandates that you switch to customer-managed keys (CMK) stored in Azure Key Vault immediately without losing index data or rebuilding the service. How should you implement this?

Accepted Answer

Configure encryption with customer-managed keys in the Azure portal on the existing search service instance and assign necessary Key Vault access permissions.. Azure AI Search supports enabling customer-managed key (CMK) encryption on an existing service without data loss or service recreation. You configure CMK in the Azure portal (or via ARM/Bicep) under the Encryption blade, then grant the search service's managed identity the necessary Key Vault permissions (get, wrapKey, unwrapKey). The service re-encrypts the index encryption key with your CMK, and no reindexing or rebuild is required.

Answer

Delete the search service and provision a brand new one with CMK enabled from inception.

Answer

Modify the DNS records to point to a new CMK-encrypted endpoint.

Answer

Run an Azure Automation runbook that exports indices to blob storage, deletes the service, and restores it.

Question 4

You are writing a Python script using the Azure AI Language SDK (`azure-ai-textanalytics`). Which client class should you instantiate to connect to your language resource?

Accepted Answer

TextAnalyticsClient. The `azure-ai-textanalytics` SDK (now part of the `azure-ai-language` family) exposes `TextAnalyticsClient` as the primary client class for interacting with Azure AI Language services such as sentiment analysis, key phrase extraction, and language detection. You instantiate it with your endpoint and an `AzureKeyCredential` or `DefaultAzureCredential`. The class name directly matches the package name, making it the correct entry point for this SDK version.

Answer

LanguageServiceClient

Answer

TextAnalysisClient

Answer

CognitiveServicesClient

Question 5

You are testing an image analysis application in Python. You want to retrieve object bounding boxes using Azure AI Vision 4.0. Which feature string should you pass to the client?

Accepted Answer

VisualFeatures.OBJECTS. In the Azure AI Vision 4.0 SDK, object detection is requested by passing VisualFeatures.OBJECTS to the analyze call, which returns bounding boxes and labels for detected objects. The enum value maps to the 'objects' feature string on the wire. This is distinct from TAGS (whole-image labels), CAPTION (single-sentence description), and READ (OCR text).

Answer

VisualFeatures.TAGS

Answer

VisualFeatures.CAPTION

Answer

VisualFeatures.READ

Question 6

Your enterprise organization provisions multiple Azure AI Services accounts across different subscriptions. Security mandates that all API keys must be automatically rotated every 30 days without human intervention. Which Azure service configuration should you implement?

Accepted Answer

Azure Key Vault automated secret rotation using an Azure Function. Azure Key Vault automated secret rotation using an Azure Function is correct because Key Vault supports rotation policies that can trigger an Azure Function to generate new keys and update the AI Services account automatically. This enables fully automated, scheduled rotation without human intervention. The Azure Function can call the Azure AI Services management API to regenerate keys and store the new keys back in Key Vault.

Answer

Microsoft Defender for Cloud JIT access policies

Answer

Azure AI Services built-in automatic key rotation policy

Answer

Azure Monitor alert rules tied to Microsoft Entra ID

Question 7

Which THREE features are provided by the Azure AI Foundry evaluation service?

Accepted Answer

Visualization of evaluation results. Option A is correct because the Azure AI Foundry evaluation service provides a results dashboard/portal experience where you can visualize evaluation runs, compare metrics across models or datasets, and drill into per-row scores. Option B is correct because the service ships built-in evaluators for RAG and generative AI scenarios, including quality metrics such as coherence, fluency, groundedness, and relevance, which can be applied without writing custom scoring logic. Option D is correct because the evaluation service supports custom evaluators, allowing you to bring your own evaluation code (for example, Python-based custom metrics) and register it alongside built-in evaluators. Option C is not part of the evaluation service: fine-tuning is a separate model customization capability, and evaluation does not automatically retrain or fine-tune models based on failures. Option E is not provided by the evaluation service: UI generation is unrelated to evaluation, which focuses on measuring and reporting model output quality rather than generating application interfaces.

Answer

Automated model fine-tuning based on failures

Answer

Automatic user interface generation

Question 8

You are reviewing the Azure Cost Management dashboard to analyze spending on Azure AI services. Which grouping option allows you to see costs broken down by specific resource names?

Accepted Answer

Resource. In Azure Cost Management, the 'Resource' grouping dimension breaks down costs by individual resource name (e.g., a specific Azure OpenAI account, a specific AI Search service). This is the only grouping that surfaces per-resource granularity, which is what the question asks for. Resource type groups by category (e.g., 'Microsoft.CognitiveServices/accounts'), not by name.

Answer

Resource type

Answer

Meter category

Answer

Location

Question 9

Which TWO strategies are recommended to prevent 'Model Over-Reliance' or 'Automation Bias' in agents?

Accepted Answer

Include a clear disclaimer that the agent is an AI and may provide incorrect information.. Option A is correct because a clear disclaimer that the agent is an AI and may provide incorrect information sets appropriate user expectations and prompts critical evaluation of the agent's output, directly countering automation bias (the tendency to over-trust automated systems). Option C is correct because providing citations and links back to original source documents lets users verify claims against authoritative sources, which reduces blind reliance on the model and supports grounded, evidence-based decision-making. Options B, D, and E do not belong: disabling 'I don't know' responses encourages the agent to fabricate answers instead of expressing uncertainty, removing human oversight eliminates the human-in-the-loop check that mitigates over-reliance, and using the highest available model temperature increases randomness and hallucination risk rather than improving trustworthiness.

Answer

Disable the ability for the agent to say 'I don't know'.

Answer

Remove all human oversight from the process.

Answer

Use the highest available model temperature.

Question 10

Your company uses Azure Policy to enforce governance across all subscriptions. A developer attempts to deploy an Azure OpenAI resource in a region that is prohibited by corporate policy. What happens during the deployment?

Accepted Answer

The deployment is blocked and returns an error explaining the policy violation.. Azure Policy operates as a guardrail at the control plane: when a deployment violates a Deny-effect policy, the Resource Manager blocks the request and returns an error describing the policy violation. The resource is never created, and the developer must remediate the request to comply.

Answer

Azure automatically moves the resource to an allowed region.

Answer

The resource is deployed successfully, but an email alert is sent to the administrator.

Answer

The deployment succeeds and automatically applies a remediation task.

Question 11

You are implementing opinion mining in an application that analyzes hotel reviews. Which THREE pieces of information does opinion mining return for an evaluated sentence? (Choose three.)

Accepted Answer

Sentiment polarity (positive, negative, neutral) for the aspect. Opinion mining (aspect-based sentiment analysis) decomposes an evaluated sentence into its key components: option B is correct because it returns the sentiment polarity (positive, negative, or neutral) associated with a specific aspect, which is the core output of the analysis. Option C is correct because it identifies the target aspect, such as 'room' or 'staff', i.e., the entity or feature being evaluated in the sentence. Option D is correct because it extracts the assessment evaluation, such as 'clean' or 'rude', which is the descriptive term expressing the opinion about the aspect. Options A and E are incorrect because compiler error logs and network bandwidth utilization metrics are unrelated to natural language opinion mining outputs; they belong to software development and network monitoring domains, respectively.

Answer

Compiler error logs for client application code

Answer

Network bandwidth utilization metrics

Question 12

Which TWO actions should you perform to troubleshoot a model that consistently misinterprets tables?

Accepted Answer

Verify labels for alignment and completeness. Option D is correct because when a model consistently misinterprets tables, the first troubleshooting step is to audit the ground-truth labels — verifying that bounding boxes, row/column tags, and table region annotations are aligned with the actual content and that no cells or headers are missing ensures the model is learning from accurate supervision. Option E is correct because table misinterpretation often stems from insufficient variety in the training data; retraining with more diverse samples (different table layouts, border styles, fonts, merged cells, and scan qualities) improves the model's ability to generalize to unseen table structures. Option A is not appropriate because the Read API is a general OCR endpoint for printed and handwritten text and does not provide the table-structure extraction or custom-model tuning needed to fix a misinterpreting custom model. Option B is wrong because disabling the layout model removes the component that detects structural elements like tables, which would worsen rather than troubleshoot the problem. Option C is incorrect because changing the resource region only affects data residency and endpoint location, not the model's table-parsing accuracy.

Answer

Use the Read API instead

Answer

Disable the layout model

Answer

Change the resource region

Question 13

Which TWO file formats are natively supported as inputs for analysis in Azure AI Document Intelligence? (Choose TWO)

Accepted Answer

application/pdf. Azure AI Document Intelligence natively accepts application/pdf (option B) because PDF is one of its primary supported input formats, allowing both digital and scanned documents to be submitted directly to the Analyze Document operations. It also natively supports image/jpeg (option D), since JPEG is among the supported image formats (along with PNG, BMP, TIFF, and HEIF) that the service can process for OCR and document analysis. The other options are not natively supported inputs: text/html (A) is not an accepted content type, application/zip (C) is not a document format the service reads directly, and application/vnd.ms-excel (E) is an Excel spreadsheet format that Document Intelligence does not accept as a native input.

Answer

text/html

Answer

application/zip

Answer

application/vnd.ms-excel

Question 14

You are implementing indirect prompt injection defenses in a RAG application where documents are ingested from public web pages. Which THREE mitigation strategies should you deploy? (Choose THREE)

Accepted Answer

Use strict structural delimiters (like XML tags) to isolate retrieved context from system instructions. Option A is correct because strict structural delimiters such as XML tags (e.g., ...) create a clear boundary between untrusted retrieved document text and trusted system instructions, making it harder for injected instructions inside web content to be interpreted as authoritative commands. Option D is correct because output validation filters inspect model completions for unauthorized actions, data exfiltration patterns, or policy violations, providing a defense-in-depth control that catches injection effects even if the input-side isolation fails. Option E is correct because Azure AI Content Safety's Prompt Shields feature is specifically designed to detect both direct (user) and indirect (document) prompt injection attacks, including spotlighting techniques used in RAG ingestion pipelines. Option B is incorrect because trusting all incoming web page text unconditionally is the exact vulnerability that enables indirect prompt injection, since public pages can contain adversarial instructions. Option C is incorrect because disabling content moderation filters removes a key safety layer and increases the risk that malicious or harmful injected content is processed and acted upon.

Answer

Trust all incoming web page text unconditionally without inspection

Answer

Disable all content moderation filters to maximize processing speed

Question 15

Your application calls Azure AI Vision API and receives an HTTP 429 Too Many Requests response. What does this error signify, and how should your application handle it?

Accepted Answer

You have exceeded your pricing tier rate limits. Implement exponential backoff and retry the request.. HTTP 429 Too Many Requests indicates that the application has exceeded the rate limits for its Azure AI Vision pricing tier (e.g., transactions per second or per minute). The correct handling is to implement exponential backoff with jitter and retry the request after a delay, respecting the 'Retry-After' header if provided. This prevents overwhelming the service and allows the request to succeed once the rate limit resets.

Answer

The service is undergoing maintenance. Fail over to a secondary region.

Answer

The image file format is unsupported. Convert to PNG.

Answer

Your API key is invalid. Refresh your subscription key immediately.