Explain Core AI Principles and Terminology CY0-001 Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   What is a primary relevance of artificial intelligence to cybersecurity?

   • AIt allows security systems to learn from data and adapt to new and evolving threats.
   • BIt enables the automatic generation of encryption keys for all data at rest.
   • CIt guarantees that all cyberattacks will be prevented before they occur.
   • DIt replaces the need for human security analysts entirely.
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   What is the basic principle of machine learning?

   • AA system stores all possible inputs and outputs in a database for instant retrieval.
   • BA system follows a set of manually written rules to make decisions.
   • CA system improves its performance on a task by learning from data without being explicitly programmed.
   • DA system relies on human intervention to update its logic after each new threat is discovered.
   Reveal Solution Discussion Select an answer first
4. 18expert · hard

   A security team is using an NLP model to analyze chat logs for potential data exfiltration. The model sometimes fails to detect exfiltration when employees use slang or code words. What is the most likely reason for this limitation?

   • AThe model is using too many features
   • BThe model was trained on formal language and may not generalize well to informal or evolving language such as slang
   • CThe model is overfitting to the test set
   • DThe model is underfitting the training data
   Reveal Solution Discussion Select an answer first
5. 19expert · hard

   A security team is using an AI model to filter phishing emails. They notice that the model is more likely to flag emails written in certain languages as phishing, even when they are legitimate. This is because the training data contained more phishing examples in those languages. What is the most appropriate way to address this bias?

   • AIncrease the model's complexity to better fit the training data
   • BRemove all non-English emails from the training data
   • CIgnore the issue because the model is only for filtering emails
   • DCollect more diverse training data and re-evaluate the model's performance across different language groups
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   What is a key difference between deep learning and traditional machine learning?

   • ADeep learning can only be used for image recognition, while traditional machine learning is used for all other tasks.
   • BDeep learning uses multiple layers of neural networks to automatically learn hierarchical features from data.
   • CDeep learning is a rule-based system, while traditional machine learning uses statistical models.
   • DDeep learning requires manual feature extraction, while traditional machine learning learns features automatically.
   Reveal Solution Discussion Select an answer first