D831 Introduction to AI and Security
Access The Exact Questions for D831 Introduction to AI and Security
💯 100% Pass Rate guaranteed
🗓️ Unlock for 1 Month
Rated 4.8/5 from over 1000+ reviews
- Unlimited Exact Practice Test Questions
- Trusted By 200 Million Students and Professors
What’s Included:
- Unlock Actual Exam Questions and Answers for D831 Introduction to AI and Security on monthly basis
- Well-structured questions covering all topics, accompanied by organized images.
- Learn from mistakes with detailed answer explanations.
- Easy To understand explanations for all students.
Access and unlock Multiple Practice Question for Introduction to AI and Security to help you Pass at ease.
Free D831 Introduction to AI and Security Questions
Which data format is commonly used for storing and exchanging structured data between systems?
-
JavaScript Object Notation (JSON)
-
Graphics Interchange Format (GIF)
-
MPEG Audio Layer 3 (MP3)
-
HyperText Markup Language (HTML)
Explanation
Explanation
Correct Answer: A. JavaScript Object Notation (JSON)
JavaScript Object Notation (JSON) is a lightweight, text-based format widely used for storing and exchanging structured data between systems. It is easy for both humans to read and write and for machines to parse and generate, making it the standard format for APIs, web applications, and AI systems that exchange data.
The other options are incorrect because GIF is an image file format, MP3 is an audio file format, and HTML is a markup language used to structure and display web pages rather than to exchange structured data between systems.
How can prompt injection attacks be mitigated through training data curation?
-
By filtering inputs to prevent harmful prompts
-
By aligning model behavior with human values
-
By encrypting the training data
-
By including adversarial examples
Explanation
Explanation
Correct Answer: D. By including adversarial examples
Training data curation can help mitigate prompt injection attacks by incorporating adversarial examples into the training process. These are intentionally crafted inputs that simulate malicious or misleading prompts. By exposing the model to such examples during training, it learns to recognize and resist prompt injection attempts, improving its robustness and resilience.
The other options are incorrect because filtering inputs is a runtime defense rather than a training data curation method, aligning model behavior with human values is a broader AI alignment goal rather than a specific data curation technique for prompt injection, and encrypting training data protects data confidentiality but does not directly improve resistance to prompt-based attacks.
A European company is integrating AI technology into its operations to automate decision-making processes. Which law should they consider to ensure transparency and accountability in AI systems?
-
Health Insurance Portability and Accountability Act (HIPAA)
-
California Consumer Privacy Act (CCPA)
-
General Data Protection Regulation (GDPR)
-
European Union Artificial Intelligence (EU AI) law
Explanation
Explanation
Correct Answer: D. European Union Artificial Intelligence (EU AI) law
The European Union AI Act (EU AI law) is specifically designed to regulate AI systems by establishing requirements for transparency, accountability, risk management, human oversight, and safety. It adopts a risk-based approach, placing stricter obligations on high-risk AI systems to ensure they are developed and used responsibly within the European Union.
The other options are incorrect because HIPAA governs the privacy and security of health information in the United States, CCPA is a California law focused on consumer privacy rights, and GDPR regulates the processing and protection of personal data in the European Union. While GDPR applies to some automated decision-making involving personal data, the EU AI law specifically addresses the governance, transparency, and accountability of AI systems.
What is an example of a hardware agent in AI?
-
Autonomous car
-
Chatbot
-
Image recognition software
-
Voice assistant
Explanation
Explanation
Correct Answer: A. Autonomous car
A hardware agent is an AI system that interacts with the physical world using sensors to perceive its environment and actuators to perform actions. An autonomous car is a hardware agent because it uses cameras, radar, LiDAR, and other sensors to detect its surroundings and controls steering, acceleration, and braking to navigate safely.
The other options are incorrect because a chatbot, image recognition software, and a voice assistant are primarily software agents. They process information and perform tasks in digital environments without directly interacting with the physical world through hardware components such as sensors and actuators.
How do "if-then" rules contribute to decision-making in AI systems?
-
They automate the process of data analysis.
-
They assist in problem-solving tasks.
-
They are primarily used for language understanding tasks.
-
They provide a structured framework for making decisions.
Explanation
Explanation
Correct Answer: D. They provide a structured framework for making decisions.
"If-then" rules are a fundamental form of rule-based logic in AI. They define specific conditions ("if") and the corresponding actions or outcomes ("then"), allowing an AI system to make consistent and predictable decisions. These rules provide a clear, structured framework for decision-making, especially in expert systems and rule-based AI applications.
The other options are incorrect because automating data analysis is not the primary purpose of "if-then" rules, assisting in problem-solving is a broader function that does not specifically describe their role in decision-making, and language understanding relies on natural language processing techniques rather than simple rule-based "if-then" logic.
How does machine learning contribute to outlier detection in data cleaning?
-
By ignoring outliers to focus on the majority of the data
-
By removing all data points that deviate from the mean
-
By introducing more outliers for analysis
-
By training models to detect data patterns and anomalies
Explanation
Explanation
Correct Answer: D. By training models to detect data patterns and anomalies
Machine learning contributes to outlier detection by learning the normal patterns within a dataset and identifying data points that significantly deviate from those patterns. These anomalies can indicate errors, fraud, unusual events, or other important observations that require further investigation. This approach improves the accuracy and quality of data cleaning by detecting outliers more effectively than relying solely on simple statistical methods.
The other options are incorrect because ignoring outliers may overlook important anomalies, removing all data points that deviate from the mean could eliminate valid observations, and introducing more outliers does not improve data cleaning or anomaly detection.
What is a common evasion technique used in evasion attacks?
-
Conducting regular security audits
-
Using strong passwords
-
Installing antivirus software
-
Hiding malicious code
Explanation
Explanation
Correct Answer: D. Hiding malicious code
Evasion attacks are designed to avoid detection by security systems such as antivirus software, intrusion detection systems, or AI-based classifiers. A common evasion technique is hiding malicious code through methods like obfuscation, encryption, packing, or polymorphism, making it harder for security tools to recognize and flag the threat.
The other options are incorrect because conducting regular security audits is a defensive practice, using strong passwords helps with access security but is not an evasion technique, and installing antivirus software is also a defensive measure rather than a method used to evade detection.
How does semi-structured data differ from structured data?
-
It is easily searchable.
-
It follows a strict data model.
-
It is stored in a relational database.
-
It lacks a predefined schema.
Explanation
Explanation
Correct Answer: D. It lacks a predefined schema.
Semi-structured data does not follow a rigid, predefined schema like structured data. Instead, it uses flexible organizational elements such as tags, keys, or metadata to organize information. Common examples include JSON, XML, and email messages. This flexibility allows semi-structured data to accommodate varying data formats while still being easier to process than completely unstructured data.
The other options are incorrect because being easily searchable is not a defining characteristic of semi-structured data, following a strict data model describes structured data, and being stored in a relational database is a characteristic of structured data rather than semi-structured data.
Why is it essential for developers to select appropriate machine learning (ML) model architectures based on the specific requirements of the AI application?
-
To ensure compatibility with all programming languages
-
To increase the computational complexity of the model
-
To simplify the training process
-
To optimize performance and accuracy of the AI model
Explanation
Explanation
Correct Answer: D. To optimize performance and accuracy of the AI model
Selecting the right ML model architecture is critical because different AI tasks (e.g., classification, regression, image recognition, NLP) require different structural designs to perform effectively. The appropriate architecture ensures the model can learn patterns efficiently, generalize well to new data, and deliver high accuracy and performance aligned with the application’s goals.
The other options are incorrect because compatibility with programming languages is not the primary reason for selecting model architectures, increasing computational complexity is generally undesirable rather than a goal, and while architecture choice can influence training efficiency, the main objective is not simply to simplify training but to maximize model effectiveness and accuracy.
How can statistical approaches aid in identifying outliers in a dataset?
-
By analyzing data distribution using quartiles
-
By creating separate datasets for potential outliers
-
By randomly assigning values to potential outliers
-
By increasing the dataset's complexity
Explanation
Explanation
Correct Answer: A. By analyzing data distribution using quartiles
Statistical approaches identify outliers by examining the distribution of data. One common method uses quartiles and the interquartile range (IQR) to determine whether data points fall unusually far above or below the rest of the dataset. Values outside the typical range are flagged as potential outliers for further investigation.
The other options are incorrect because creating separate datasets does not identify outliers, randomly assigning values would distort the data rather than detect unusual observations, and increasing the dataset's complexity does not help identify outliers and may make analysis more difficult.
How to Order
Select Your Exam
Click on your desired exam to open its dedicated page with resources like practice questions, flashcards, and study guides.Choose what to focus on, Your selected exam is saved for quick access Once you log in.
Subscribe
Hit the Subscribe button on the platform. With your subscription, you will enjoy unlimited access to all practice questions and resources for a full 1-month period. After the month has elapsed, you can choose to resubscribe to continue benefiting from our comprehensive exam preparation tools and resources.
Pay and unlock the practice Questions
Once your payment is processed, you’ll immediately unlock access to all practice questions tailored to your selected exam for 1 month .