AI-900 Practice Exam 66 — 20 Questions

Question 1

A developer is using Azure OpenAI Service to generate Python code snippets. They notice that the generated code often contains repetitive function definitions and loops. Which parameter should be increased to reduce this repetition?

Accepted Answer

Frequency penalty. The frequency penalty parameter reduces repetition by penalizing tokens that have already appeared in the generated text, making the model less likely to reuse the same functions or loops. Increasing this value directly discourages the model from generating repetitive patterns, which is exactly the issue described.

Answer

Temperature

Answer

Max tokens

Answer

Top P

Question 2

A company uses a generative AI model to answer customer questions about their products. They observe that the model sometimes produces factually incorrect or fabricated information. To reduce these inaccuracies, they want to provide the model with relevant, up-to-date product documentation as context before generating a response. Which technique is being applied?

Accepted Answer

Grounding. B is correct because grounding is the technique of providing a generative AI model with specific, authoritative source data (such as product documentation) as context before generating a response. This anchors the model's output to verified facts, directly reducing hallucinations and fabricated information by constraining the generation to the provided context.

Answer

Prompt Engineering

Answer

Fine-tuning

Answer

Reinforcement Learning from Human Feedback (RLHF)

Question 3

A bank deploys an AI system to approve personal loan applications. After six months, an audit reveals that applicants from certain postal codes receive significantly lower approval rates than applicants from other postal codes, even when their income and credit scores are comparable. Which Microsoft responsible AI principle is most directly violated by this outcome?

Accepted Answer

Fairness. The AI system's approval decisions produce systematically different outcomes for applicants from different postal codes despite comparable income and credit scores, which directly violates the Fairness principle. Fairness requires that AI systems treat all individuals and groups equitably, avoiding discrimination based on sensitive attributes like location. The audit evidence shows the model has learned spurious correlations between postal code and loan risk, leading to biased approval rates.

Answer

Transparency

Answer

Inclusiveness

Answer

Reliability and safety

Question 4

A company uses an AI system to help screen job applications. The system ranks candidates based on their resumes. The company wants to ensure that if a candidate asks why they were not selected, the company can provide a clear explanation of the factors that influenced the AI's decision. Which Microsoft responsible AI principle is most directly relevant?

Accepted Answer

Transparency. Transparency is the responsible AI principle that requires AI systems to be understandable and interpretable. In this scenario, the company needs to provide a clear explanation of why a candidate was not selected, which directly aligns with transparency's goal of making AI decisions explainable to users.

Answer

Accountability

Answer

Privacy and security

Answer

Reliability and safety

Question 5

A city council deploys an AI system to analyze surveillance footage and automatically issue traffic violation fines. They want to ensure the system does not disproportionately target one type of vehicle (e.g., bicycles over cars) when issuing fines. Which Microsoft responsible AI principle is most directly relevant?

Accepted Answer

Fairness. The scenario describes a risk of algorithmic bias where the AI system might disproportionately issue fines to bicycles over cars. The Microsoft responsible AI principle of Fairness directly addresses this by requiring that AI systems treat all groups equitably and avoid discrimination based on protected attributes. Ensuring fairness involves auditing the model's predictions across different vehicle types and mitigating any statistical disparities.

Answer

Inclusiveness

Answer

Reliability and safety

Answer

Transparency

Question 6

A company uses Azure OpenAI Service to generate long technical reports. To manage costs, the development team needs to accurately estimate the number of tokens that a given prompt will consume before making any API call. Which Azure OpenAI Service feature should they use to obtain this estimate?

Accepted Answer

The Token Counter tool in Azure OpenAI Studio. The Token Counter tool in Azure OpenAI Studio is specifically designed to estimate the number of tokens a prompt will consume before making an API call. This allows developers to predict costs accurately by calculating token usage for both input and expected output, without incurring actual API charges.

Answer

The Chat Completions API

Answer

The Embeddings API

Answer

The Content Filter configuration

Question 7

A company deploys an AI chatbot on its website to answer customer questions. The company wants to be transparent about the nature of the interaction. Which Microsoft responsible AI principle is most directly relevant to ensuring users know they are communicating with an AI and not a human?

Accepted Answer

Transparency. Transparency is the Microsoft responsible AI principle that requires AI systems to be designed so that users are aware they are interacting with an AI, not a human. In the context of a chatbot, this means clearly disclosing the AI nature of the system, such as through a label or introductory message, to avoid deception and build trust.

Answer

Inclusiveness

Answer

Privacy and security

Answer

Accountability

Question 8

A company deploys an AI system to screen job resumes. The system consistently rejects candidates from a certain university, but the company cannot determine which features led to the decision or how the model arrived at that outcome. Which Microsoft responsible AI principle is most directly violated?

Accepted Answer

Transparency. The scenario describes a system that makes decisions without providing any insight into how or why those decisions were made. Transparency, as a Microsoft responsible AI principle, requires that AI systems be understandable and that their decisions can be explained. Since the company cannot determine which features led to the rejection or how the model arrived at the outcome, the lack of explainability directly violates the Transparency principle.

Answer

Fairness

Answer

Accountability

Answer

Privacy and security

Question 9

A company uses Azure OpenAI to build a customer service chatbot. They want to prevent malicious users from injecting prompts that cause the chatbot to behave unexpectedly, such as revealing its system instructions. Which responsible AI consideration is most directly relevant?

Accepted Answer

Reliability and Safety. Prompt injection attacks target the system by embedding malicious instructions in user input, causing the model to override its original directives or reveal sensitive information. This directly undermines the reliability and safety of the AI system, as the chatbot's behavior becomes unpredictable and potentially harmful. Azure OpenAI's safety systems (e.g., content filtering, abuse detection) are designed to mitigate such risks, making Reliability and Safety the most relevant responsible AI consideration.

Answer

Fairness

Answer

Privacy and Security

Answer

Inclusiveness

Question 10

A customer support team wants to use Azure AI Language to automatically analyze incoming support emails. They need to extract the product name mentioned in each email and determine whether the customer's sentiment is positive, negative, or neutral. They have no labeled data for custom training. Which two prebuilt Azure AI Language features should they use together?

Accepted Answer

Named entity recognition (NER) and sentiment analysis. Named entity recognition (NER) extracts specific entities like product names from text, while sentiment analysis determines the emotional tone (positive, negative, neutral). Both are prebuilt, no-code features in Azure AI Language that require no labeled data for custom training, making them the correct pair for this use case.

Answer

Key phrase extraction and language detection

Answer

Conversational language understanding (CLU) and translation

Answer

Text summarization and personal identifying information (PII) detection

Question 11

A financial institution uses an AI model to approve loan applications. A customer is denied a loan and requests an explanation for the decision. The development team cannot explain how the model reached its conclusion because the model is a deep neural network with complex layers. Which Microsoft responsible AI principle is being violated in this scenario?

Accepted Answer

Accountability. The scenario describes a model whose internal reasoning cannot be explained by the development team, which directly violates the Microsoft responsible AI principle of Accountability. Accountability requires that organizations be able to explain AI decisions and take ownership of their outcomes. A deep neural network that acts as a 'black box' prevents the team from providing the required explanation to the customer, thus breaking this principle.

Answer

Fairness

Answer

Reliability and Safety

Answer

Privacy and Security

Question 12

A data scientist is training a logistic regression model to predict customer churn using a small dataset with 500 records and 200 features. The model achieves 97% accuracy on the training set but only 65% on a held-out test set, indicating severe overfitting. The data scientist wants to reduce overfitting by automatically eliminating irrelevant features. Which technique should the data scientist apply?

Accepted Answer

Apply L1 regularization (Lasso) to the model. L1 regularization (Lasso) adds a penalty equal to the absolute value of the magnitude of coefficients, which can shrink some coefficients exactly to zero. This performs automatic feature selection by eliminating irrelevant features, directly addressing the overfitting caused by having 200 features on only 500 records. The high training accuracy (97%) versus low test accuracy (65%) is a classic sign of overfitting that L1 regularization mitigates by reducing model complexity.

Answer

Apply L2 regularization (Ridge) to the model

Answer

Use k-fold cross-validation to select the best model

Answer

Increase the number of training samples by data augmentation

Question 13

A healthcare company deploys an AI system to assist doctors in diagnosing skin conditions from images. The system is a deep neural network that does not provide explanations for its predictions. The company implements a process where every AI recommendation is logged, and a medical team reviews any adverse outcomes to determine if the system or a human made an error. The company also clearly assigns responsibility for the system's outputs to a specific clinical oversight committee. Which Microsoft responsible AI principle is most directly being implemented by these actions?

Accepted Answer

Accountability. The actions described—logging AI recommendations, reviewing adverse outcomes to determine error source, and assigning a clinical oversight committee—directly implement the Accountability principle. Accountability requires that organizations clearly assign responsibility for AI system outputs and have processes to review and address impacts, which is exactly what the company does by creating a human review loop and a designated committee.

Answer

Fairness

Answer

Reliability and safety

Answer

Transparency

Question 14

A hospital uses an AI system to prioritize patient appointments based on urgency. The system is trained on historical data. The team wants to ensure that the system does not discriminate against patients based on age or disability. Which Microsoft responsible AI principle should most directly guide the design of this system?

Accepted Answer

Inclusiveness. The Inclusiveness principle directly guides the design of this system because it requires AI systems to empower everyone and avoid discrimination, including on the basis of age or disability. By proactively identifying and mitigating bias in the training data and model predictions, the hospital can ensure that appointment prioritization does not unfairly disadvantage any patient group. This principle is specifically about fairness and accessibility, making it the most relevant for preventing discriminatory outcomes.

Answer

Privacy and security

Answer

Reliability and safety

Answer

Transparency

Question 15

What is the difference between a chatbot and a conversational AI agent?

Accepted Answer

Chatbots use fixed rules/decision trees; conversational AI uses NLP/ML for flexible, context-aware responses. Option B is correct because chatbots traditionally rely on predefined rules or decision trees to handle user inputs, limiting them to scripted interactions. In contrast, conversational AI agents leverage natural language processing (NLP) and machine learning (ML) to understand intent, manage context, and generate dynamic, human-like responses. This allows conversational AI to handle ambiguous phrasing, maintain multi-turn dialogue state, and adapt to user behavior without explicit programming for every scenario.

Answer

Chatbots are always voice-based; conversational AI is text-only

Answer

Chatbots are more expensive to build than conversational AI

Answer

They are the same technology with different marketing terms

Question 16

A healthcare clinic uses an AI system to triage patients by urgency. The system consistently assigns lower priority to patients presenting with rare symptoms compared to those with common symptoms, even when the rare symptoms indicate a serious condition. The clinic wants to ensure the system treats all patients equitably. According to Microsoft's Responsible AI principles, which principle is most directly relevant to addressing this disparity?

Accepted Answer

Fairness. The AI system's consistent assignment of lower priority to patients with rare symptoms, despite those symptoms indicating serious conditions, is a clear case of algorithmic bias that leads to unfair treatment outcomes. Microsoft's Fairness principle directly addresses this by requiring AI systems to allocate resources and make decisions without discrimination or favoritism, ensuring equitable treatment across all patient groups regardless of symptom prevalence.

Answer

Inclusiveness

Answer

Transparency

Answer

Accountability

Question 17

A hospital deploys an AI system to assist with diagnosing diseases from medical images. A doctor disagrees with the system's diagnosis and overrules it. The hospital wants to document this interaction for legal and audit purposes. Which Microsoft responsible AI principle is most directly relevant?

Accepted Answer

Accountability. The scenario involves documenting a human override of an AI system's diagnosis for legal and audit purposes, which directly relates to accountability. Accountability in responsible AI ensures that organizations can answer for their AI systems' decisions by maintaining clear records of interactions, including when humans overrule AI outputs. This principle requires traceability and governance mechanisms, such as audit trails, to assign responsibility for outcomes.

Answer

Fairness

Answer

Reliability and safety

Answer

Transparency

Question 18

What is 'speculative decoding' and how does it improve LLM inference speed?

Accepted Answer

Using a small draft model to generate candidate tokens that a large model verifies in parallel — improving throughput. Speculative decoding improves LLM inference speed by using a small, fast draft model to generate multiple candidate tokens in sequence, which are then verified in parallel by the large target model. This parallel verification allows the large model to accept or reject entire blocks of tokens at once, significantly reducing the number of sequential autoregressive steps required. The technique leverages the observation that draft models can produce acceptable continuations most of the time, and the large model only needs to correct mistakes, leading to higher throughput without sacrificing output quality.

Answer

Predicting user input before they finish typing to pre-compute responses

Answer

Generating speculative forecasts about future events using language model knowledge

Answer

Running model inference on the CPU while the GPU processes the next request in parallel

Question 19

What is the 'presence penalty' parameter in Azure OpenAI API calls?

Accepted Answer

A flat penalty discouraging repetition of any token already present in the response. The 'presence penalty' parameter in Azure OpenAI API calls applies a flat penalty to any token that has already appeared in the response so far, reducing the model's likelihood of repeating that token. This helps generate more diverse and less repetitive text by discouraging the reuse of tokens already present in the output sequence.

Answer

A parameter requiring AI systems to acknowledge their presence as AI to users

Answer

A parameter indicating whether the AI is present online or offline

Answer

The minimum number of characters that must be present in a response

Question 20

What is 'speech synthesis markup language' (SSML) in Azure AI Speech?

Accepted Answer

An XML language for controlling TTS speech rate, pitch, pauses, and emphasis. SSML is an XML-based markup language that allows you to fine-tune text-to-speech (TTS) output by controlling prosodic elements such as speech rate, pitch, volume, pauses, and emphasis. In Azure AI Speech, SSML tags like , , and are embedded in the input text to produce more natural and expressive synthesized speech.

Answer

A markup language for styling how text is displayed on screen in Azure apps

Answer

A security standard for marking sensitive spoken content for redaction

Answer

A standard for describing the structure of multi-turn dialogue in speech interfaces