Chapter6(part2) AI in Cybersecurity - AMIS1003

Flashcards covering AI-powered cyber attacks, defenses, benefits, and risks based on the Chapter 6-part 2 lecture notes.

How is Artificial Intelligence (AI) defined in the context of computer systems?

AI refers to computer systems that can perform tasks that normally require human intelligence, such as learning, reasoning, problem-solving, and decision-making.

How has AI changed the nature of cyber attacks compared to traditional methods?

Traditionally, attacks required significant technical knowledge and manual effort. Today, AI enables attackers to automate aspects of cybercrime, making attacks faster, more scalable, and more convincing.

What are the specific characteristics of AI-powered phishing attacks?

Personalized messages based on social media information, correct grammar and professional language, realistic company branding and formatting, automatic translation into multiple languages, and large-scale automated email generation.

What are the four key features of AI-based malware?

1. Self-Learning Capabilities (modifying behavior to avoid detection); 2. Evasion Techniques (changing code structure to bypass security software); 3. Automated Decision-Making (determining the most valuable files to target); 4. Rapid Propagation (efficient identifying and spreading to vulnerable systems).

How do AI systems facilitate automated vulnerability discovery?

They can simultaneously scan thousands of websites, analyze software code for vulnerabilities, identify outdated software versions, detect configuration errors, and recommend potential attack paths.

What are the three steps involved in Voice Cloning technology?

Step 1: Voice Collection (from social media, interviews, etc.); Step 2: AI Training (analyzing tone, pitch, style, and pronunciation); Step 3: Voice Generation (creating synthetic speech that resembles the target).

What are the common applications of Deepfake technology in cybercrime?

Executive impersonation, political misinformation, corporate fraud, and fake news dissemination.

According to the CyberArk data published in May 2024, what percentage of cyber decision makers believe AI-powered malware will do harm in 2024?

$$40\,\%$$

How is Machine Learning (ML) defined in the cybersecurity context?

A subset of AI that enables systems to learn from data and improve over time without explicit programming.

How does an AI-Based Intrusion Detection System (IDS) differ from a traditional IDS?

Traditional systems rely on predefined rules and signatures, whereas AI-powered systems can learn normal network behavior, detect unknown attacks, and adapt to changing threats.

What is 'Predictive Threat Intelligence' in AI cybersecurity?

A shift from reactive to proactive security where AI analyzes historical attack data and identifies emerging trends to strengthen defenses before attacks occur.

What is the difference between a False Positive and a False Negative in AI-powered security?

A False Positive occurs when legitimate activity is incorrectly identified as malicious (e.g., a traveling employee flagged as an attacker). A False Negative occurs when a genuine attack goes undetected.

What are the primary privacy concerns mentioned regarding AI systems in cybersecurity?

Collection of personal information, data misuse, unauthorized access, and regulatory compliance challenges.

What are some repetitive tasks AI can automate to reduce operational costs?

Log review, threat classification, vulnerability scanning, and security reporting.

What are the capabilities of AI-powered chatbots used in social engineering?

Real-time responses, personalized interactions, emotional manipulation, and multi-language communication.