How does concept drift differ from data drift?

Concept drift changes the relationship between inputs and outputs, while data drift changes the distribution of input data.

Data drift is caused by model updates, while concept drift is caused by external factors.

Concept drift is permanent, while data drift is temporary.

Data drift changes the relationship between inputs and outputs, while concept drift changes the input distribution.

Which mitigation strategy is specifically aimed at reducing automation bias in AI-based systems?

Reducing the number of test cases to speed up testing.

Increasing the complexity of the AI model to improve accuracy.

Requiring human verification of AI outputs for high-risk decisions.

Using only automated testing tools to evaluate the AI system.

Which performance-related information should be documented for an AI component?

The color scheme of the user interface.

The names of the developers who wrote the code.

The metrics used to evaluate the model, such as accuracy and precision.

The number of lines of code in the AI component.

In a hospital, an AI system alerts nurses to potential patient deterioration. The system has a known false alarm rate, but nurses have started to ignore alerts because they have been wrong in the past. Which risk is most directly related to this situation?

Automation bias, but in the opposite direction: under-trusting the AI can lead to missed critical alerts.

Concept drift, because the AI's performance has changed over time.

Overfitting, because the model has learned the training data too well.

Data drift, because the input data distribution has changed.

A company is developing an AI system for image recognition. The test team needs to document the AI component to support future audits and compliance checks. Which documentation approach is most aligned with this need?

Write a brief summary of the AI's purpose and leave it at that.

Create a comprehensive data sheet that includes the dataset's origin, collection methods, and any known biases.

Store the model's source code in a version control system.

Document only the model's final accuracy on the test set.

Certified Tester AI Testing

Domain 7Objective 3

Specialized Testing Considerations CT-AI Practice Questions (Page 2)

Part of the Domain 7: Testing AI-Based Systems Overview domain, which makes up ~5% of our current practice bank. ISTQB does not publish an official question count, but from its 60-minute exam (~25–40 total, ~1–2 in this domain), expect 1–1 from this objective — we provide 35 practice questions to prepare you well beyond it. (estimate)

35questions here

7free pages

10concepts

Questions 6–10

6foundation · easy
How does concept drift differ from data drift?
Select an answer first
7foundation · easy
Which mitigation strategy is specifically aimed at reducing automation bias in AI-based systems?
Select an answer first
8foundation · easy
Which performance-related information should be documented for an AI component?
Select an answer first
9application · medium
In a hospital, an AI system alerts nurses to potential patient deterioration. The system has a known false alarm rate, but nurses have started to ignore alerts because they have been wrong in the past. Which risk is most directly related to this situation?
Select an answer first
10application · medium
A company is developing an AI system for image recognition. The test team needs to document the AI component to support future audits and compliance checks. Which documentation approach is most aligned with this need?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CT-AI

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by ISTQB. “CT-AI” is a trademark of its owner, used for identification only.