2026 - ISTQB AI Testing (CT-AI) Certification - Crash Course

About this course

A complete practical course to make you understand AI Testing with ISTQB exam readiness.Traditional software testing assumes predictable logic and fixed expected outputs. AI systems don’t work that way. They learn from data, evolve over time, behave probabilistically, and often operate as black boxes. This shift breaks many traditional testing assumptions. In this course, you will learn how to test AI-based systems the right way, using globally accepted ISTQB AI Testing (CT-AI) principles, explained clearly and practically for testers.This course starts by building strong foundations. You will first understand what AI really is, how AI-based systems differ from conventional software, and why new testing strategies are required. You’ll then learn machine learning fundamentals — supervised, unsupervised, and reinforcement learning — not as a data scientist, but from a tester’s mindset.As the course progresses, you’ll explore the complete ML lifecycle, focusing on what testers must validate at each stage: data preparation, training, validation, testing, deployment, and ongoing monitoring. You’ll learn how poor data quality, bias, imbalance, and mislabeling directly impact model behavior and test outcomes.You will deeply understand ML performance metrics, their limitations, and how to detect overfitting and underfitting. The course then moves into testing AI-specific quality characteristics such as fairness, ethics, safety, transparency, interpretability, and explainability (XAI).Finally, you’ll apply AI-specific testing techniques like adversarial testing, metamorphic testing, data poisoning, A/B testing, and exploratory testing, and also learn how AI itself can be used to enhance testing through test generation and defect prediction.This course strictly follows the

Course objectives

• prepare testers for the ISTQB AI Testing (CT-AI) certification
• build a foundational understanding of machine learning concepts
• enable effective testing of AI systems through learned techniques