CT-AI ISTQB Certified Tester AI Testing Exam Question and Answers

ï‚· A. This statement is incorrect. Current AI is true AI and there is no reason to believe that this fact will change over time.

AI is an evolving field, and the definition of what constitutes AI can change as technology advances.

ï‚· B. This statement is correct. In general, what is considered AI today may change over time.

The term AI is dynamic and has evolved over the years. What is considered AI today might be viewed as standard computing in the future. Historically, as technologies become mainstream, they often cease to be considered "AI".

ï‚· C. This statement is incorrect. What is considered AI today will continue to be AI even as technology evolves and changes.

This perspective does not account for the historical evolution of the definition of AI. As new technologies emerge, the boundaries of AI shift.

ï‚· D. This statement is correct. In general, today the term AI is utilized incorrectly.

While some may argue this, it is not a universal truth. The term AI encompasses a broad range of technologies and applications, and its usage is generally consistent with current technological capabilities.

From the confusion matrix:

True Positives (TP) = 15

False Positives (FP) = 5

False Negatives (FN) = 15

True Negatives (TN) = 65

Accuracy= (TP + TN) / Total

= (15 + 65) / 100

=80%

Precision= TP / (TP + FP)

= 15 / (15 + 5)

= 15 / 20

=75%

Section3.2 – Functional Performance Criteriain the syllabus explains accuracy and precision exactly these ways when evaluating ML classification performance.

Option B is therefore the only correct pair of values.

Flexibility in AI systems is crucial for various reasons, particularly because it allows for easier modification and adaptation of the system as a whole.

AI systems are inherently flexible (A): This statement is not correct. While some AI systems may be designed to be flexible, they are not inherently flexible by nature. Flexibility depends on the system's design and implementation.

AI systems require changing operational environments; therefore, flexibility is required (B): While it's true that AI systems may need to operate in changing environments, this statement does not directly address the importance of flexibility for the modification of the system.

Flexible AI systems allow for easier modification of the system as a whole (C): This statement correctly describes the importance of flexibility. Being able to modify AI systems easily is critical for their maintenance, adaptation to new requirements, and improvement.

Self-learning systems are expected to deal with new situations without explicitly having to program for it (D): This statement relates to the adaptability of self-learning systems rather than their overall flexibility for modification.

Hence, the correct answer isC. Flexible AI systems allow for easier modification of the system as a whole.

The question asks which characteristic is least likely to cause safety-related issues for an AI system. Let's evaluate each option:

Non-determinism (A): Non-deterministic systems can produce different outcomes even with the same inputs, which can lead to unpredictable behavior and potential safety issues.

Robustness (B): Robustness refers to the ability of the system to handle errors, anomalies, and unexpected inputs gracefully. A robust system is less likely to cause safety issues because it can maintain functionality under varied conditions.

High complexity (C): High complexity in AI systems can lead to difficulties in understanding, predicting, and managing the system's behavior, which can cause safety-related issues.

Self-learning (D): Self-learning systems adapt based on new data, which can lead to unexpected changes in behavior. If not properly monitored and controlled, this can result in safety issues.

ï‚· A. Using a pixel comparison of the GUI before and after the change to check the differences.

Pixel comparison is a traditional method and does not involve AI. It compares images at the pixel level, which can be effective but is not an intelligent approach. It is not considered an AI usage and is the least likely usage of AI for detecting GUI changes.

ï‚· B. Using computer vision to compare the GUI before and after the test object changes.

Computer vision involves using AI techniques to interpret and process images. It is a likely usage of AI for detecting changes in the GUI.

ï‚· C. Using vision-based detection of the GUI layout changes before and after test object changes.

Vision-based detection is another AI technique where the layout and structure of the GUI are analyzed to detect changes. This is a typical application of AI.

ï‚· D. Using a ML-based classifier to flag if changes in GUI are to be flagged for humans.

An ML-based classifier can intelligently determine significant changes and decide if they need human review, which is a sophisticated AI application.

The ISTQB CT-AI syllabus defines these two concepts clearly inSection 2.1 – Flexibility and Adaptability. Flexibility is described as the ability of a system to operate in situationsnot explicitly covered in its original requirements, while adaptability refers to how easily the system can bemodifiedto handle new environments or conditions. The syllabus stresses that both flexibility and adaptability are crucial, particularly inself-learning AI systemsthat may need to respond to changes in their environment and adjust their behavior accordingly. It states that systems must be capable of determiningwhenandhowto adjust behavior in evolving situations, especially when the operational environment is not fully known at deployment time . This directly aligns with OptionA.

Option B reverses definitions—the syllabus states flexibility (not adaptability) relates to unspecified situations. Option C is incorrect: self-learning systems requirebothflexibility and adaptability; they are not categorized as one or the other. Option D incorrectly defines flexibility; the syllabus defines adaptability—not flexibility—as ease of modification.

Thus,Option Acorrectly reflects the syllabus.

To calculate the accuracy, recall, and specificity from the confusion matrix provided, we use the following formulas:

Confusion Matrix:

Actually Rotten: 45 (True Positive), 8 (False Positive)

Actually Fresh: 5 (False Negative), 42 (True Negative)

Accuracy:

Accuracy is the proportion of true results (both true positives and true negatives) in the total population.

Formula: Accuracy=TP+TNTP+TN+FP+FN\text{Accuracy} = \frac{TP + TN}{TP + TN + FP + FN}Accuracy=TP+TN+FP+FNTP+TN​

Calculation: Accuracy=45+4245+42+8+5=87100=0.87\text{Accuracy} = \frac{45 + 42}{45 + 42 + 8 + 5} = \frac{87}{100} = 0.87Accuracy=45+42+8+545+42​=10087​=0.87

Recall (Sensitivity):

Recall is the proportion of true positive results in the total actual positives.

Formula: Recall=TPTP+FN\text{Recall} = \frac{TP}{TP + FN}Recall=TP+FNTP​

Calculation: Recall=4545+5=4550=0.9\text{Recall} = \frac{45}{45 + 5} = \frac{45}{50} = 0.9Recall=45+545​=5045​=0.9

Specificity:

Specificity is the proportion of true negative results in the total actual negatives.

Formula: Specificity=TNTN+FP\text{Specificity} = \frac{TN}{TN + FP}Specificity=TN+FPTN​

Calculation: Specificity=4242+8=4250=0.84\text{Specificity} = \frac{42}{42 + 8} = \frac{42}{50} = 0.84Specificity=42+842​=5042​=0.84

Therefore, the correct combinations of accuracy, recall, and specificity are 0.87, 0.9, and 0.84 respectively.

The ISTQB CT-AI syllabus definesoverfittingin Section3.2 – ML Model Evaluationas a condition where an ML model learns the training data too precisely—including noise and irrelevant detail—resulting in poor performance on unseen data. Overfitting is characterized byhigh accuracy on training data but low accuracy on validation or real-world data. OptionAperfectly matches this definition: a model trained only on one university’s student data generalizes poorly to students from other universities. This is a textbook example of overfitting because the model has essentially memorized patterns unique to a narrow dataset, instead of learning generalizable relationships applicable across environments .

Option B instead describessample biasor inadequate training diversity, not overfitting. Option C involvestransfer learningor model extension, unrelated to overfitting. Option D indicatesinsufficient training data qualityor lack of meaningful features, but not overfitting. Only Option A reflects the syllabus definition directly: overly specialized training leading to reduced predictive performance on new data.

Thus,Ais the correct and syllabus-aligned example of overfitting.

ï‚· Technology Most Typically Used to Implement AI:Genetic algorithms are a well-known technique used in AI. They are inspired by the process of natural selection and are used to find approximate solutions to optimization and search problems. Unlike search engines, procedural programming, or case control structures, genetic algorithms are specifically designed for evolving solutions and are commonly employed in AI implementations.

 Reference:ISTQB_CT-AI_Syllabus_v1.0, Section 1.4 AI Technologies, which identifies different technologies used to implement AI​​.

The syllabus describes that data pre-processing includes cleaning (e.g., fixing or removing invalid data) and transforming data (e.g., changing data types such as numbers stored as strings).

"Transformation: The format of the given data is changed… converting categorical data into numerical data, changing image formats…"

(Reference: ISTQB CT-AI Syllabus v1.0, Section 4.1.1, Page 34 of 99)

Section4.3 – Test Levels for AI Systemsclearly defines ML model testing as the level at which testers evaluate whether an ML model fulfills itsfunctional performance criteria, including accuracy, precision, recall, F1, robustness, stability, and fairness. Therefore, Option C is the correct and syllabus-aligned statement.

Option A is incorrect because input data testing focuses onvalidity and correctness of data entering the model, not interactions with all system components. Option B is incorrect: acceptance testing in the syllabus focuses primarily onbusiness and stakeholder requirements, not specifically explainability. Explainability testing may occur at multiple levels depending on context. Option D is also incorrect because API testing belongs tointegration testing, not system testing, even when AI is consumed as a service.

Thus,Option Cis the only statement that precisely matches syllabus definitions.

A. Regression

Regression is used to predict a continuous value and is not suitable for grouping people based on similar tastes.

B. Association

Association is used to find relationships between variables in large datasets, often in the form of rules (e.g., market basket analysis). It does not directly group individuals but identifies patterns of co-occurrence.

C. Clustering

Clustering is an unsupervised learning method used to group similar data points based on their features. It is ideal for identifying groups of people with similar tastes without prior knowledge of the group labels. This technique will help the company segment its customer base effectively.

D. Classification

Classification is a supervised learning method used to categorize data points into predefined classes. It requires labeled data for training, which is not the case here as we want to identify groups without predefined labels.

Therefore, the correct answer isCbecause clustering is the most suitable method for grouping people with similar tastes for targeted product marketing​​.

The syllabus defines bias as:

"Bias is the systematic difference in treatment of certain objects, people or groups in comparison to others."

It also discusses:

"Sample bias can occur if the data used for training the model does not represent the operational environment, or if some relevant faulty conditions are excluded deliberately."

(Reference: ISTQB CT-AI Syllabus v1.0, Section 7.6 and 8.3)

The ISTQB CT-AI syllabus explains inSection 4.5 – Testing AI-Specific Risksthat adversarial testing is a structured test activity in which testers applyadversarial attacks—crafted or perturbed inputs—to intentionally expose weaknesses in the ML model. The purpose is to identify vulnerabilities that could be exploited throughdata poisoning,evasion attacks, orinput manipulation. OptionCcorrectly reflects this syllabus definition: adversarial testing is aboutusing attacks to locate weaknesses so they can be removed or mitigated.

Option A is incorrect because regression testing does not verify data sourcing policies; it verifies unchanged functionality after modifications. Option B is incorrect because adversarial examples are oftenaddedto training datasets to improverobustness(a practice called adversarial training), not excluded. Option D is incorrect because AIB testing is not described as superior to exploratory data analysis in outlier detection; both have different purposes, and EDA remains essential for data quality assessment.

Thus,Option Cis consistent with syllabus-defined adversarial testing.

The syllabus highlights that testing for unnecessary human intervention is a key focus for autonomous AI-based systems:

"For autonomous AI-based systems, testers must ensure that the system does not prompt for unnecessary human intervention, as this contradicts the autonomy concept."

(Reference: ISTQB CT-AI Syllabus v1.0, Section 8.2, page 59 of 99)

Defect prediction models aim to identify parts of the software that are likely to contain defects by analyzing historical data and code quality metrics. The primary goal is to use this predictive information to allocate testing and maintenance resources effectively. Let's break down why option D is the correct choice:

Understanding Classification Models:

Classification models are a type of supervised learning algorithm used to categorize or classify data into predefined classes or labels. In the context of defect prediction, the classification model would classify parts of the code as either "defective" or "non-defective" based on the input features.

Input Data - Code Quality Metrics:

The input data for these classification models typically includes various code quality metrics such as cyclomatic complexity, lines of code, number of methods, depth of inheritance, coupling between objects, etc. These metrics help the model learn patterns associated with defects.

Historical Data:

Historical versions of the code along with their defect records provide the labeled data needed for training the classification model. By analyzing this historical data, the model can learn which metrics are indicative of defects.

Why Option D is Correct:

Option D specifies using a classification model to predict the presence of defects by using code quality metrics as input data. This accurately describes the process of defect prediction using historical bug data and quality metrics.

Eliminating Other Options:

A. Identifying the relationship between developers and the modules developed by them: This does not directly involve predicting defects based on code quality metrics and historical data.

B. Search of similar code based on natural language processing: While useful for other purposes, this method does not describe defect prediction using classification models and code metrics.

C. Clustering of similar code modules to predict based on similarity: Clustering is an unsupervised learning technique and does not directly align with the supervised learning approach typically used in defect prediction models.

The ISTQB CT-AI syllabus explains supervised learning under Section1.6 – Machine Learning Approaches. It definesclassificationas predictingcategorical labels, whereasregressionpredictscontinuous numerical values. OptionB—deciding whether an object is a bicycle or a motorcycle—fits the definition of classification precisely because the model chooses between discrete categories. The syllabus also uses similar examples to illustrate classification tasks, reinforcing that this is the correct interpretation .

Option A is incorrect because image recognition of a dog is aclassificationtask, not regression. Option C is incorrect because predicting a 10% price rise involves forecasting anumerical value, which is aregressionproblem. Option D is incorrect because classification can involveany number of classes, not only two. Multiclass classification is explicitly mentioned in the syllabus.

Therefore, OptionBis the only answer aligned with the syllabus’ definitions.

TheISTQB Certified Tester AI Testing Syllabus v1.0explicitly categorizes different AI implementation technologies in Section1.4 – AI Technologies. Within this section, AI methods are grouped into categories, one of which is“Reasoning techniques.”These reasoning techniques includerule engines, deductive classifiers, case-based reasoning, and procedural reasoning. Because deductive classifiers are directly listed under this set of reasoning approaches, they are recognized as a reasoning-based AI technology.

Reasoning techniques differ from machine learning approaches because they rely onstructured, predefined rules or logicto reach conclusions. Deductive classifiers use logical inference and symbolic reasoning to classify inputs by applying encoded knowledge. This makes them fundamentally different from statistical or data-driven ML algorithms.

The other options—Linear regression,Random Forest, andGenetic algorithms—are listed by the syllabus asmachine learning techniques, not reasoning methods. Linear regression performs numerical prediction, Random Forest is an ensemble decision-tree ML model, and genetic algorithms are optimization-based ML approaches inspired by evolutionary processes. None of these involve symbolic logical deduction.

Thus, based on the authoritative definitions in the syllabus,Deductive classifiers (Option A)is the only technology classified as a reasoning technique.

Pairwise testing is used to handle the large number of combinations of parameters that can arise in complex systems like self-driving cars. The question asks which of the given options isleast likelyto be a reason for the explosion in the number of parameters.

Different Road Types (A): Self-driving cars must operate on various road types, such as highways, city streets, rural roads, etc. Each road type can have different characteristics, requiring the car's system to adapt and handle different scenarios. Thus, this is a significant factor contributing to the growth of parameters.

Different Weather Conditions (B): Weather conditions such as rain, snow, fog, and bright sunlight significantly affect the performance of self-driving cars. The car's sensors and algorithms must adapt to these varying conditions, which adds to the number of parameters that need to be considered.

ML Model Metrics to Evaluate Functional Performance (C): While evaluating machine learning (ML) model performance is crucial, it does not directly contribute to the explosion of parameter combinations in the same way that road types, weather conditions, and car features do. Metrics are used to measure and assess performance but are not themselves variable conditions that the system must handle.

Different Features like ADAS, Lane Change Assistance, etc. (D): Advanced Driver Assistance Systems (ADAS) and other features add complexity to self-driving cars. Each feature can have multiple settings and operational modes, contributing to the overall number of parameters.

Hence, theleast likelyreason for the incredible growth in the number of parameters isC. ML model metrics to evaluate the functional performance.

The question asks which test is least likely to be performed during the ML model testing phase. Let's consider each option:

Testing the accuracy of the classification model (A): Accuracy testing is a fundamental part of the ML model testing phase. It ensures that the model correctly classifies the data as intended and meets the required performance metrics.

Testing the API of the service powered by the ML model (B): Testing the API is crucial, especially if the ML model is deployed as part of a service. This ensures that the service integrates well with other systems and that the API performs as expected.

Testing the speed of the training of the model (C): This is least likely to be part of the ML model testing phase. The speed of training is more relevant during the development phase when optimizing and tuning the model. During testing, the focus is more on the model's performance and behavior rather than how quickly it was trained.

Testing the speed of the prediction by the model (D): Testing the speed of prediction is important to ensure that the model meets performance requirements in a production environment, especially for real-time applications.

The syllabus describes metamorphic testing as:

“Testing involves defining metamorphic relations and then applying those relations to check that the transformations result in expected outcomes, even when the expected output of the system is unknown or not well-defined.”

In this scenario, applying the metamorphic relation (speed differences) and checking the transformed outcome (arrival times) fits the definition of metamorphic testing.

(Reference: ISTQB CT-AI Syllabus v1.0, Section 9.5, page 69 of 99)

When written requirements are given in text documents, the best way to generate test cases is by using Natural Language Processing (NLP). Here's why:

Natural Language Processing (NLP): NLP can analyze and understand human language. It can be used to process textual requirements to extract relevant information and generate test cases. This method is efficient in handling large volumes of textual data and identifying key elements necessary for testing.

Why Not Other Options:

Analyzing source code for generating test cases: This is more suitable for white-box testing where the code is available, but it doesn't apply to text-based requirements.

Machine learning on logs of execution: This approach is used for dynamic analysis based on system behavior during execution rather than static textual requirements.

GUI analysis by computer vision: This is used for testing graphical user interfaces and is not applicable to text-based requirements.

Decision trees are a foundational algorithm used in supervised machine learning. The syllabus describes:

"A decision tree is a tree-like ML model whose nodes represent decisions and whose branches represent possible outcomes."

(Reference: ISTQB CT-AI Syllabus v1.0, Section 3.4)

A. Add 10% of the rows randomly and create another model and compare the R-Square scores of both the models.

Adding more data to the training set can affect the R-Square score, but it does not directly verify the correctness of the implementation.

B. Train various models by changing the order of input features and verify that the R-Square score of these models vary significantly.

Changing the order of input features should not significantly affect the R-Square score if the implementation is correct, but this approach is more about testing model robustness rather than correctness of the implementation.

C. Compare the R-Square score of the model obtained using two different implementations that utilize two different programming languages while using the same algorithm and the same training and testing data.

This approach directly compares the performance of two implementations of the same algorithm. If both implementations produce similar R-Square scores on the same training and testing data, it suggests that the new implementation "X" is correct.

D. Drop 10% of the rows randomly and create another model and compare the R-Square scores of both the models.

Dropping data can lead to variations in the R-Square score but does not directly verify the correctness of the implementation.

Therefore, optionCis the most appropriate strategy because it directly compares the performance of the new implementation "X" with another implementation using the same algorithm and datasets, which helps in verifying the correctness of the implementation​​.

The ISTQB CT-AI syllabus explains that reusing pre-trained models is strongly related tosimilarity between the original task and the new task. Section1.8 – Pre-trained Models and Transfer Learningstates that reuse is effective when the new task is similar to the original one, such as adapting a cat-classifier to classify dog breeds. The syllabus warns about risks related toinput differences,data preparation inconsistencies,inherited shortcomings, andexplainability issues. These are legitimate objections (matching options A, B, and C) because large differences in image inputs or patterns can undermine transfer learning; misclassification risk can increase; and explainability often decreases when reusing pre-trained models .

However,output differences are NOT a valid concernhere. Both the leaf-based and flower-based ML models classifythe same plant species, meaning theiroutputs are identical. The syllabus does not identify output mismatch as a transfer-learning risk. Real risks concerninputs,bias inheritance,model transparency, andtraining differences—not output labels. Therefore, OptionDdescribes an objection that can be safelyneglected, because output classes are the same and do not hinder reuse.

The syllabus states:

“When testing probabilistic and non-deterministic systems, the same input may produce different outputs. Tests need to be run several times to produce statistically valid test results, ensuring that an appropriate number of answers are accurate.”

(Reference: ISTQB CT-AI Syllabus v1.0, Section 8.4, page 58 of 99)

The ISTQB CT-AI syllabus describes thatself-learning systems continuously adjust their behaviorduring operation as new data arrives. Section4.1 – Challenges of Testing AI-Based Systemshighlights that such systems are vulnerable todata attacks, particularly through adversarial inputs, poisoning, or malicious drift. The risk arises because unexpected changes in the input distribution may alter the learned model in harmful ways. OptionD – Unexpected changescorresponds directly to this syllabus-defined risk.

Option A refers to system specification issues but does not relate to data attacks. Option B discusses environment complexity, which makes testing difficult but is not tied to adversarial threats. Option C (insufficient testing time) affects quality but does not specifically increase vulnerability to malicious data manipulation.

Unexpected changes—including data drift, poisoned samples, or maliciously constructed training data—pose the greatest risk. When a self-learning system adapts to altered data patterns, it may unknowingly learn incorrect associations, causing model degradation or manipulation. Therefore,Option Dcorrectly identifies the challenge that increases exposure to data attacks.

A. Feature selection

Feature selection is the process of selecting the most relevant features from the data. While important, it is not directly about handling excess data.

B. Data sampling

Data sampling involves selecting a representative subset of the data for training. When there is more data than needed, sampling can be used to create a manageable dataset that maintains the statistical properties of the full dataset.

C. Data labeling

Data labeling involves annotating data for supervised learning. It is necessary for training models but does not address the issue of having excess data.

D. Data augmentation

Data augmentation is used to increase the size of the training dataset by creating modified versions of existing data. It is useful when there is insufficient data, not when there is excess data.

Therefore, the correct answer isBbecause data sampling is the most relevant activity when dealing with an excess amount of data for training​​.

The syllabus defines precision as:

"Precision = TP / (TP + FP) * 100%. Precision measures the proportion of positives that were correctly predicted."

Using the confusion matrix:

TP = 90

FP = 10Thus: Precision = (90 / (90 + 10)) * 100% = 90 / 100 * 100% = 90%However, the confusion matrix totals suggest that the calculation should be done in the form:Precision = 90 / (90 + 10) * 100% = 90%Since the given answers do not include exactly 90%, the closest approximation and the correct answer, as described in the syllabus, would be 90%.(Reference: ISTQB CT-AI Syllabus v1.0, Section 5.1, page 40 of 99)

The syllabus explains:

"Where the internal structure of an AI-based system is too complex for humans to understand, the system can only be tested as a black box. Even when the internal structure is visible, this provides no additional useful information to help with testing."

This confirms that black-box testing is chosen because the tester does not need to understand the system’s internal structure.

(Reference: ISTQB CT-AI Syllabus v1.0, Section 8.5, page 61 of 99)

The ISTQB CT-AI syllabus (Section4.3 – Test Environments for AI Systems) describes that, unlike conventional software testing, testing AI systems may requirespecialized toolsfor analyzing and explaining the decisions of ML models. This includes visualization tools, explainability frameworks, and diagnostic utilities to understand why the AI made a certain prediction. Since AI decisions may be non-transparent, the test environment must supportexplainability, making OptionBcorrect.

Option A is incorrect: the syllabus does not state that an autonomous AI system requires an autonomous test environment. Option C is incorrect because test environments mustnot learn; they must remain stable to avoid unpredictable testing conditions. Option D is incorrect because multi-agent systems often involve stochastic interactions, and determinism is neither required nor realistic.

Thus,Option Bis the syllabus-accurate choice.

ï‚· Testing Autonomy:Testing for human handover when it should not be relinquishing control is the least appropriate because it contradicts the very definition of autonomous systems. The other tests are relevant to ensuring smooth operation and transitions between human and AI control.

 Reference:ISTQB_CT-AI_Syllabus_v1.0, Sections on Testing Autonomous AI-Based Systems and Testing for Human-AI Interaction​​​​. ​

The ISTQB CT-AI syllabus explains that AI-based decision-support systems benefit strongly fromback-to-back testingandmetamorphic testingwhen oracle problems exist or when limited regression tests are available. In this scenario, CleverPropose replaces an older advisory system.Back-to-back testing(Option A) is ideal because the outputs of the existing conventional system can serve as areference, enabling comparison against the new AI system. This is exactly what the syllabus recommends when AI is replacing a traditional deterministic system.

Metamorphic testing(Option C) is also appropriate, as stated in Section4.6 – Metamorphic Relations. With limited regression tests and complex decision logic, testers can define metamorphic relations such as “if customer income increases, risk rating should not worsen.” These relations allow validation even when exact expected outputs are unavailable.

Exploratory data analysis (Option D) is not a system testing technique. Pairwise testing (Option E) is not well suited for complex AI-based financial advice systems. Adversarial testing (Option B) is more relevant for security-critical or robustness evaluation, not primary system testing for advisory tools.

Thus,A and Care the correct and syllabus-supported choices.