Free Practice Questions for ISTQB Certified Tester AI Testing (CT-AI) V2.0 Certification

This subdomain covers the fundamental concepts of artificial intelligence, including its history, types, and applications. It introduces the basic terminology and principles that underpin AI systems, such as machine learning, deep learning, and generative AI. The content also addresses the current state of AI and its impact on software testing.

This subdomain focuses on the specific quality characteristics relevant to AI-based systems, as defined by standards like ISO/IEC 25059. It discusses attributes such as functional suitability, performance efficiency, compatibility, usability, reliability, security, maintainability, and portability in the context of AI. The subdomain also covers how these characteristics differ from traditional software and how they influence testing strategies.

This subdomain addresses the definition and evaluation of acceptance criteria for AI-based systems. It includes methods for establishing measurable criteria that account for the probabilistic and non-deterministic nature of AI. The content covers techniques for validating that an AI system meets business and user requirements, considering factors like model accuracy, fairness, and explainability.

This subdomain provides an overview of machine learning, including supervised, unsupervised, and reinforcement learning. It explains key concepts such as training, validation, and testing datasets, overfitting, underfitting, and model selection. The content also introduces the machine learning lifecycle and its implications for testing.

This subdomain covers the role of data in machine learning, including data collection, preprocessing, labeling, and augmentation. It discusses data quality issues such as bias, completeness, and representativeness, and their impact on model performance. The content also addresses data privacy and ethical considerations relevant to testing.

This subdomain focuses on metrics used to evaluate the performance of classification models, such as accuracy, precision, recall, F1-score, and ROC-AUC. It explains how to calculate and interpret these metrics, and their relevance to testing. The content also covers the selection of appropriate metrics based on business objectives and the trade-offs between different metrics.

This subdomain introduces neural networks, including their architecture, activation functions, and training processes. It covers simple neural networks and their testing challenges, such as understanding model behavior and detecting errors. The content also discusses techniques for testing neural networks, including adversarial testing and coverage metrics.

This subdomain provides an overview of the unique challenges and approaches in testing AI-based systems. It covers the differences between testing traditional software and AI systems, including handling non-determinism, probabilistic outputs, and the oracle problem. The content also introduces test strategies and techniques specific to AI testing.

This subdomain focuses on testing generative AI systems, such as large language models (LLMs). It covers evaluation methods for generated content, including quality, relevance, and safety. The content addresses challenges like hallucination, bias, and ethical concerns, and discusses testing techniques such as prompt engineering and output validation.

This subdomain describes the test levels specific to machine learning systems, including offline and online testing. It explains the scope and objectives of each level, such as model evaluation, integration testing, and monitoring in production. The content also covers the role of continuous testing and feedback loops in ML systems.

This subdomain covers testing of input data for machine learning systems, including data validation, schema checking, and statistical analysis. It addresses techniques for detecting data drift, anomalies, and biases. The content also discusses the importance of data quality for model performance and methods for automating data testing.

This subdomain focuses on testing the machine learning model itself, including functional and non-functional testing. It covers techniques such as metamorphic testing, adversarial testing, and back-to-back testing. The content also addresses model evaluation using performance metrics, and testing for robustness, fairness, and explainability.

This subdomain covers testing activities integrated into the machine learning development process. It includes testing of data pipelines, feature engineering, and model training code. The content addresses continuous integration and delivery practices for ML, and the role of version control and reproducibility in testing.