Headlines

• Headline
  Course Orientation

  Understand the syllabus structure, how questions are written, and how to study for scenario-style CT-AI items.

  • Sub-headline
    Exam Expectations
• Headline
  1. AI Landscape and Terminology

  Define AI terms used by ISTQB, recognize AI system types, and understand why AI creates new testing problems.

  • Sub-headline
    1.1 Defining AI and the AI Effect
  • Sub-headline
    1.2 AI-Based vs Conventional Software
  • Sub-headline
    1.3 Delivery Models
• Headline
  2. AI-Specific Quality Characteristics

  Translate AI risks into test objectives: bias, explainability, transparency, autonomy, adaptability, and non-determinism.

  • Sub-headline
    2.1 Non-Determinism and Its Testing Impact
  • Sub-headline
    2.2 Transparency vs Explainability
  • Sub-headline
    2.3 Autonomy, Adaptability, Self-Learning
  • Sub-headline
    2.4 Automation Bias and Human Decision Quality
• Headline
  3. Machine Learning Fundamentals

  Understand learning types, model behavior, and why training choices create specific test risks.

  • Sub-headline
    3.1 Learning Paradigms and Testing Impact
  • Sub-headline
    3.2 Underfitting vs Overfitting
• Headline
  4. Data for AI Systems

  Test data sourcing, preparation, labeling, and drift controls because data drives model behavior.

  • Sub-headline
    4.1 Data Sourcing and Representativeness
  • Sub-headline
    4.2 Data Preparation and Leakage Prevention
  • Sub-headline
    4.3 Data Labeling Controls
  • Sub-headline
    4.4 Drift and Data Refresh Strategy
• Headline
  5. ML Functional Performance Metrics

  Select metrics that match risk, class balance, and decision cost; do not treat validation metrics as proof of real quality.

  • Sub-headline
    5.1 Classification Metrics
  • Sub-headline
    5.2 Benchmarks and Model Comparison
  • Sub-headline
    5.3 Metrics in Production
• Headline
  6. Neural Networks and Testing Challenges

  Understand why deep models create transparency/explainability gaps and different robustness risks.

  • Sub-headline
    6.1 Transparency Checks (Often Without Execution)
  • Sub-headline
    6.2 Explainability Requires Execution
• Headline
  7. Testing AI-Based Systems

  Plan test objectives, identify AI-specific testing challenges, and choose appropriate test levels and acceptance criteria.

  • Sub-headline
    7.1 Test Oracle Problem
  • Sub-headline
    7.2 Human Factors and Decision Quality
  • Sub-headline
    7.3 Concept Drift
  • Sub-headline
    7.4 Acceptance Criteria and Test Strategy
• Headline
  8. Testing AI-Specific Quality Characteristics

  Apply tests for bias, robustness, autonomy, explainability, and complexity-driven risks.

  • Sub-headline
    8.1 Bias Testing
  • Sub-headline
    8.2 Complexity as a Testing Challenge
  • Sub-headline
    8.3 Autonomy Testing
• Headline
  9. Test Techniques for AI-Based Systems

  Use AI-relevant test techniques: adversarial testing, pairwise, back-to-back, metamorphic testing, exploratory data analysis, and data pipeline testing.

  • Sub-headline
    9.1 Adversarial Testing
  • Sub-headline
    9.2 Pairwise Testing for Large Parameter Spaces
  • Sub-headline
    9.3 Back-to-Back Testing (Pseudo-Oracle)
  • Sub-headline
    9.4 A/B Testing
  • Sub-headline
    9.5 Metamorphic Testing
  • Sub-headline
    9.6 Exploratory Data Analysis (EDA) as Testing
• Headline
  10. Test Environments for AI-Based Systems

  Build environments that support simulation, large datasets, reproducibility, and explainability outputs.

  • Sub-headline
    10.1 Virtual Test Environments
  • Sub-headline
    10.2 Explainability in the Environment
• Headline
  11. Using AI to Support Testing

  Apply AI-assisted testing responsibly: test generation, defect prediction, regression optimization, and visual testing.

  • Sub-headline
    11.1 AI-Based Test Generation
  • Sub-headline
    11.2 Regression Test Optimization
  • Sub-headline
    11.3 Defect Prediction
  • Sub-headline
    11.4 Visual Testing