The full syllabus

Sixty-three lectures, start to finish.

Theory first. You start with what these systems actually are and where they quietly go wrong, then work through how you test for it - data quality, drift, bias, adversarial inputs, oracles. Thirteen modules that follow the official ISTQB CT-AI v1 outline, one lecture a day, ending in a timed mock exam. The certificate is the bonus on top; the understanding is the point.

Start the course Try 12 free
63
Lectures · 5-15 min each
13
Modules · 8 weeks
300+
Practice questions
12
Free · no card
- Free taster

The first 12 lectures are free, end to end, no card and no signup wall. That is the whole Intro module plus all of Module 1. The other 51 open up when you join.

M-00
IntroductionFree Welcome, what the ISTQB CT-AI certification is, and how the course is laid out.
3 lectures
  • 01Welcome to the course!Free
  • 02About ISTQB Tester AI certificationFree
  • 03About the courseFree
M-01
Introduction to AI TestingFree Foundational concepts: what AI is, the kinds of AI systems, the technologies and frameworks behind them, and the standards that now apply.
9 lectures
  • 04Lecture 1: Definition of AI and AI EffectFree
  • 05Lecture 2: Narrow, General and Super AIFree
  • 06Lecture 3: AI-based and Conventional SystemsFree
  • 07Lecture 4: AI TechnologiesFree
  • 08Lecture 5: AI Development FrameworksFree
  • 09Lecture 6: Hardware for AI-Based SystemsFree
  • 10Lecture 7: AI as a Service (AIaaS)Free
  • 11Lecture 8: Pre-Trained ModelsFree
  • 12Lecture 9: Standards, Regulations, and AIFree
M-02
Quality Characteristics for AI-Based Systems The quality attributes that are specific to AI: adaptability, autonomy, bias, ethics, transparency, and safety.
8 lectures
  • 13Lecture 1: Flexibility and AdaptabilityQuiz
  • 14Lecture 2: AutonomyQuiz
  • 15Lesson 3: EvolutionQuiz
  • 16Lecture 4: BiasQuiz
  • 17Lecture 5: EthicsQuiz
  • 18Lecture 6: Side Effects and Reward HackingQuiz
  • 19Lecture 7: Transparency, Interpretability and ExplainabilityQuiz
  • 20Lecture 8: Safety and AIQuiz
M-03
Machine Learning (ML) - Overview How ML actually works under the hood: the forms of learning, the workflow, algorithm choice, and the overfitting trap.
5 lectures
  • 21Lecture 1: Forms of MLQuiz
  • 22Lecture 2: ML WorkflowQuiz
  • 23Lecture 3: Selecting a Form of MLQuiz
  • 24Lecture 4: Factors Involved in ML Algorithm SelectionQuiz
  • 25Lecture 5: Overfitting and UnderfittingQuiz
M-04
ML - Data Where most models really go wrong: data preparation, train/validation/test splits, quality issues, and labelling.
5 lectures
  • 26Lecture 1: Data Preparation as Part of the ML WorkflowQuiz
  • 27Lecture 2: Training, Validation and Test Datasets in the ML WorkflowQuiz
  • 28Lecture 3: Dataset Quality IssuesQuiz
  • 29Lecture 4: Data Quality and its Effect on the ML ModelQuiz
  • 30Lecture 5: Data Labeling for Supervised LearningQuiz
M-05
ML Functional Performance Metrics Reading a model honestly: the confusion matrix, the metrics for classification, regression and clustering, and where those metrics lie to you.
5 lectures
  • 31Lecture 1: Confusion MatrixQuiz
  • 32Lecture 2: Add ML Functional Performance Metrics for Classification, Regression and ClusteringQuiz
  • 33Lecture 3: Limitations of ML Functional Performance MetricsQuiz
  • 34Lecture 4: Selecting ML Functional Performance MetricsQuiz
  • 35Lecture 5: Benchmark Suites for ML PerformanceQuiz
M-06
ML Neural Networks and Testing Neural networks and the coverage measures used to test them.
2 lectures
  • 36Lecture 1: Neural NetworksQuiz
  • 37Lecture 2: Coverage Measures for Neural NetworksQuiz
M-07
Testing AI-Based Systems - Overview Putting it into practice: specification, test levels, test data, automation bias, documenting a component, and concept drift.
7 lectures
  • 38Lecture 1: Specification of AI-Based SystemsQuiz
  • 39Lecture 2: Test Levels for AI-Based SystemsQuiz
  • 40Lecture 3: Test Data for Testing AI-Based SystemsQuiz
  • 41Lecture 4: Testing for Automation Bias in AI-Based SystemsQuiz
  • 42Lecture 5: Documenting an AI ComponentQuiz
  • 43Lecture 6: Testing for Concept DriftQuiz
  • 44Lecture 7: Selecting a Test Approach for an ML SystemQuiz
M-08
Testing AI-Specific Quality Characteristics The hard part: testing self-learning, autonomous, probabilistic systems, plus bias, explainability and the oracle problem.
8 lectures
  • 45Lecture 1: Challenges Testing Self-Learning SystemsQuiz
  • 46Lecture 2: Testing Autonomous Self-Learning SystemsQuiz
  • 47Lecture 3: Testing for Algorithmic, Sample and Inappropriate BiasQuiz
  • 48Lecture 4: Challenges Testing Probabilistic and Non-Deterministic AI-Based SystemsQuiz
  • 49Lecture 5: Challenges Testing Complex AI-Based SystemsQuiz
  • 50Lecture 6: Testing Transparency, Interpretability and Explainability of AI-Based SystemsQuiz
  • 51Lecture 7: Test Oracles for AI-Based SystemsQuiz
  • 52Lecture 8: Test Objectives and Acceptance CriteriaQuiz
M-09
Methods and Techniques for the Testing of AI-Based Systems The toolkit: adversarial attacks and data poisoning, pairwise, back-to-back, A/B, metamorphic, and experience-based testing.
7 lectures
  • 53Lecture 1: Adversarial Attacks and Data PoisoningQuiz
  • 54Lecture 2: Pairwise TestingQuiz
  • 55Lecture 3: Back-to-Back TestingQuiz
  • 56Lecture 4: A/B TestingQuiz
  • 57Lecture 5: Metamorphic Testing (MT)Quiz
  • 58Lecture 6: Experience-Based Testing of AI-Based SystemsQuiz
  • 59Lecture 7: Selecting Test Techniques for AI-Based SystemQuiz
M-10
Test Environments for AI-Based Systems Setting up the test and virtual environments these systems need.
1 lecture
  • 60Lectures 1: Test and Virtual Environments for AI-Based SystemsQuiz
M-11
Using AI for Testing Turning it around: the AI technologies that help you test, from defect analysis to interface automation.
2 lectures
  • 61Lecture 1: AI Technologies for TestingQuiz
  • 62Lectures 2: Using AI for Testing: From Defect Analysis to Interface AutomationQuiz
M-12
Final Assessment A comprehensive mock exam covering all course material, run under real timed conditions.
1 lecture
  • 63Final Mock Exam - ISTQB CT-AI · 40 questions · 65% to passExam
§
Ready when you are

Start with the first twelve.

They are free, end to end, no card. If the way it is taught works for you, the other 51 lectures and the full mock exam are one payment away.

Start the course Try 12 free
14-day money-back · the ISTQB exam is booked + paid separately, with ISTQB