AI Security, Ethics and Governance →mediumMultiple ChoiceObjective-mapped

Detecting Adversarial Attacks: AI Monitoring Log Analysis

Exhibit

Refer to the exhibit.

```
[2025-04-01 14:23:45] INFO: Model inference call for job_id=123
[2025-04-01 14:23:45] ALERT: Drift detected on feature 'transaction_amount' - PSI: 0.35 (threshold: 0.20)
[2025-04-01 14:23:46] ALERT: Unusual request pattern from IP 10.0.0.55: 100 queries in 5 seconds (limit: 50)
[2025-04-01 14:23:47] WARN: Model 'fraud_detection_v2' confidence score dropped below 0.8 for 15 consecutive predictions
[2025-04-01 14:23:48] ALERT: Response time for inference increased to 200ms (baseline: 50ms)
```

Refer to the exhibit. A security analyst reviews the monitoring log for an AI fraud detection model. Which of the following is the most likely cause of the multiple alerts?

Quick Answer

The correct answer is an adversarial attack attempt because the monitoring log reveals a triad of anomalies—feature drift, an unusually high query rate, and degraded model performance—that together signal a deliberate probing of the AI fraud detection system. Unlike data poisoning, which corrupts training data and would manifest earlier, or model retraining, which causes temporary instability without a spike in queries, an adversarial attack actively exploits model vulnerabilities by sending crafted inputs to induce misclassifications, creating the real-time drift and performance drop seen in the log. On the CompTIA AI+ AI0-001 exam, this scenario tests your ability to distinguish between different AI security threats by correlating multiple log signals rather than focusing on a single symptom. A common trap is to mistake the high query rate for a network issue, but remember that network problems cause latency, not feature drift. Memory tip: think “Drift + Query Spike = Adversarial Probe” to quickly identify this attack pattern.

⚠ Common exam trap

The AI0-001 exam often tests the distinction between data poisoning (training-phase attack) and adversarial attacks (inference-phase attack), and the trap here is that candidates confuse the sudden onset of alerts with a training data issue, overlooking that adversarial attacks specifically target the model's decision boundary during live operation.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

An adversarial attack attempt

An adversarial attack attempt is the most likely cause because the monitoring log shows multiple alerts triggered by subtle, crafted perturbations in input data designed to cause the AI fraud detection model to misclassify legitimate transactions as fraudulent or vice versa. Unlike data poisoning, which corrupts the training dataset over time, adversarial attacks target the model's inference phase, exploiting its sensitivity to small input variations to produce incorrect outputs without altering the underlying training data.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Data poisoning of the training dataset
Why it's wrong here
Poisoning would not cause real-time drift and query spikes.
✗
A network hardware failure
Why it's wrong here
Hardware failure does not explain drift or query pattern.
✓
An adversarial attack attempt
Why this is correct
Multiple concurrent alerts indicate active probing or evasion.
✗
A scheduled model retraining process
Why it's wrong here
Retraining would not cause high query rate or confidence drop.

About these practice questions

One of 754 original AI0-001 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

Same concept, more angles

1 more way this is tested on AI0-001

These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.

Variation 1. A security team discovers that an AI-based anomaly detection system frequently misclassifies benign network traffic as malicious when the source IP is from a specific geographic region. Which type of AI vulnerability is most likely being exploited?

medium

A.Data poisoning
B.Model inversion
✓ C.Adversarial evasion
D.Membership inference

Why C: The scenario describes an AI-based anomaly detection system that misclassifies benign traffic from a specific geographic region as malicious. This is a classic example of an adversarial evasion attack, where an attacker crafts inputs (in this case, network traffic) that appear benign to human analysts but cause the AI model to misclassify them. The geographic bias suggests the attacker is exploiting the model's learned decision boundary, likely by manipulating features such as source IP or packet timing to evade detection.

Written by Johnson Ajibi, MSc IT Security

Senior Network & Security Engineer · founder of Courseiva

This AI0-001 practice question is part of Courseiva's free CompTIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI0-001 exam.