AI in cybersecurity (defense and offense) - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the role of AI in cybersecurity defense?

AI helps detect threats quickly by analyzing patterns and unusual activities, allowing faster response to attacks.

beginner

How can AI be used offensively in cybersecurity?

Attackers can use AI to create smarter malware, automate attacks, and find weaknesses faster than humans.

intermediate

What is anomaly detection in AI-based cybersecurity?

It is when AI learns normal system behavior and spots unusual actions that might indicate a cyberattack.

intermediate

Name one challenge of using AI in cybersecurity.

AI systems can sometimes make mistakes, like false alarms or missing new types of attacks, requiring human oversight.

intermediate

Why is AI considered a double-edged sword in cybersecurity?

Because it can protect systems by detecting threats but also be used by attackers to improve their hacking methods.

What is one way AI helps in cybersecurity defense?

A Slowing down internet speed

B Creating phishing emails

C Deleting system backups

D Detecting unusual network activity

AI analyzes data to find unusual activity that may indicate a cyberattack.

How might attackers use AI offensively?

A Automating attacks to find vulnerabilities

B Improving firewall rules

C Backing up data securely

D Monitoring employee productivity

Attackers use AI to automate and improve the efficiency of their attacks.

What does anomaly detection mean in AI cybersecurity?

A Finding normal behavior patterns

B Spotting unusual or suspicious actions

C Deleting old files

D Increasing network speed

Anomaly detection means identifying actions that differ from normal behavior, which may signal threats.

Which is a limitation of AI in cybersecurity?

A It slows down computers intentionally

B It never makes mistakes

C It can generate false alarms

D It replaces all human jobs

AI can sometimes wrongly flag safe actions as threats, requiring human review.

Why is AI called a double-edged sword in cybersecurity?

A It can both protect and be used to attack systems

B It only helps attackers

C It only helps defenders

D It is always harmful

AI can improve defense but also be exploited by attackers to enhance their methods.

Explain how AI is used in cybersecurity defense and give an example.

Think about how AI spots unusual behavior to protect systems.

Describe one way attackers might use AI offensively and why it is a concern.

Consider how AI can help hackers work faster and smarter.

Practice

1. What is one main way AI helps in cybersecurity defense?

easy

A. By slowing down network traffic to prevent attacks

B. By quickly detecting unusual activities that may be threats

C. By creating viruses to attack systems

D. By replacing all human cybersecurity experts

Solution

1. Step 1: Understand AI's role in defense

   AI analyzes data fast to spot patterns that look like attacks.

2. Step 2: Compare options

   Only By quickly detecting unusual activities that may be threats describes AI detecting threats quickly, which is a key defense use.

3. Final Answer:

   By quickly detecting unusual activities that may be threats -> Option B

4. Quick Check:

   AI detects threats fast = D [OK]

Hint: AI defense means spotting threats fast, not replacing humans [OK]

Common Mistakes:

• Thinking AI replaces all humans
• Confusing defense with attack roles
• Believing AI slows network traffic

2. Which of the following is a correct example of AI use in cybersecurity offense?

easy

A. AI encrypting data to protect it

B. AI scanning for vulnerabilities to fix them

C. AI monitoring network traffic for attacks

D. AI generating phishing emails that look real

Solution

1. Step 1: Identify AI use in offense

   Offense means attacking or tricking systems, like phishing emails.

2. Step 2: Match options to offense

   Only AI generating phishing emails that look real describes AI creating phishing emails, an offensive action.

3. Final Answer:

   AI generating phishing emails that look real -> Option D

4. Quick Check:

   AI offense = phishing emails = A [OK]

Hint: Offense means attack; phishing emails are an attack method [OK]

Common Mistakes:

• Confusing offense with defense tasks
• Choosing options about protecting data
• Thinking AI only defends, not attacks

3. Consider this scenario: An AI system monitors network traffic and flags 95% of real attacks but also wrongly flags 10% of normal traffic as attacks. What is a likely outcome?

medium

A. The system alerts on many real attacks but also causes false alarms

B. The system never makes mistakes and blocks all traffic

C. The system misses most attacks and lets threats through

D. The system only flags normal traffic and ignores attacks

Solution

1. Step 1: Analyze detection rates

   95% detection means most attacks are caught; 10% false alarms means some normal traffic is wrongly flagged.

2. Step 2: Interpret outcomes

   The system works well but causes some false alerts, which is common in AI detection.

3. Final Answer:

   The system alerts on many real attacks but also causes false alarms -> Option A

4. Quick Check:

   High detection + some false alarms = C [OK]

Hint: High detection with some false alarms means alerts plus mistakes [OK]

Common Mistakes:

• Assuming no false alarms happen
• Thinking 95% detection means perfect security
• Confusing false alarms with missed attacks

4. A cybersecurity team uses AI to detect malware but notices it misses new malware types. What is a likely cause and fix?

medium

A. Cause: AI trained only on old malware; Fix: update training data with new samples

B. Cause: AI hardware is slow; Fix: buy faster computers

C. Cause: AI is too sensitive; Fix: reduce detection thresholds

D. Cause: AI is offline; Fix: disconnect it from the network

Solution

1. Step 1: Identify why AI misses new malware

   AI learns from past data; if training lacks new malware, it won't recognize them.

2. Step 2: Determine the fix

   Updating training data with new malware samples helps AI detect new threats.

3. Final Answer:

   Cause: AI trained only on old malware; Fix: update training data with new samples -> Option A

4. Quick Check:

   Old data causes misses; update data fixes it = A [OK]

Hint: AI needs fresh data to detect new threats [OK]

Common Mistakes:

• Blaming hardware speed for detection issues
• Reducing sensitivity can miss more threats
• Disconnecting AI stops detection completely

5. A company wants to use AI to improve cybersecurity. They plan to combine AI with human experts. Which approach best balances defense and offense risks?

hard

A. Let AI act alone to block all suspicious activity immediately

B. Only use humans and avoid AI to prevent mistakes

C. Use AI to detect threats and humans to verify alerts before action

D. Use AI to create offensive attacks to test defenses without human input

Solution

1. Step 1: Understand combined AI and human roles

   AI can quickly find threats; humans add judgment to reduce errors.

2. Step 2: Evaluate options for balance

   Use AI to detect threats and humans to verify alerts before action uses AI detection plus human verification, balancing speed and accuracy.

3. Final Answer:

   Use AI to detect threats and humans to verify alerts before action -> Option C

4. Quick Check:

   AI + human teamwork = best balance = B [OK]

Hint: Combine AI speed with human judgment for best security [OK]

Common Mistakes:

• Relying only on AI without human checks
• Avoiding AI completely reduces efficiency
• Using AI offensively without control risks damage