ISTQB® Certified Tester - AI Testing (CT-AI)

Price
Net €
VAT €

Price
Price on Request

Duration
3 days

For companies and job seekers:
this course is 100% fundable!
 

Location

Course Language
English

Training Solutions
Online Live

Request a consultation Add to favorites Print as PDF

Intelligent systems make decisions based on data. Quality assurance must therefore systematically test models, training processes, and dynamic behavior. Current expertise in the field of AI testing combines classic testing methodology with data science and modern software architecture.

Key topics

  • Special features of AI systems in a testing context.
  • Verification and validation of machine learning models.
  • Test data management and data integrity.
  • Explainability, fairness, and bias control.
  • Testing strategy for adaptive and self-learning systems.
  • Governance, compliance, and risk management.

Prerequisites
Experience in software testing or quality assurance; understanding of basic development processes and test concepts.

Target audience
QA professionals, test analysts, IT project managers, development teams, data analysts, and those responsible for AI-supported products.

Conclusion
: Structured testing procedures for AI create reliability, traceability, and trust in data-based applications—essential for organizations that want to shape innovation responsibly.

Print as PDF
Course content
  • Definition and AI effect
  • Weak, strong, and superior AI
  • Comparison of traditional and AI-supported systems
  • Technological approaches to AI
  • Development platforms for AI
  • Technical equipment for AI systems
  • Artificial intelligence as a service (AIaaS)
  • Pre-built AI models
  • Guidelines, standards, and regulation of AI
  • Adaptability and versatility
  • Independence
  • Progress
  • Prejudice
  • Morality
  • Risks and manipulation of incentives
  • Openness, traceability, and interpretation
  • Protection and artificial intelligence
  • Types of machine learning
  • Machine learning process
  • Choosing the right approach
  • Factors for algorithm selection
  • Overfitting and underfitting
  • Data processing in the ML process
  • Division into training, validation, and test data
  • Challenges in data quality
  • Impact of data quality on models
  • Labeling data for supervised learning
  • Error matrix
  • Advanced performance measures for classification, regression, and clustering
  • Limitations of performance measures
  • Selecting suitable performance measures
  • Reference sets for ML
  • Artificial networks
  • Key figures on network coverage
  • Requirements for AI systems
  • Levels in AI testing
  • Data basis for AI testing
  • Bias testing in AI automation
  • Evidence for AI modules
  • Checking for concept changes
  • Choice of ML testing strategy
  • Testing learning systems
  • Testing autonomous systems
  • Testing for bias and randomness
  • Testing uncertain systems
  • Testing complex systems
  • Testing transparency and explanation
  • Oracle for AI testing
  • Define goals and criteria
  • Attacks through manipulation and false data
  • Comparative tests in pairs
  • Comparative tests
  • Variant tests A versus B
  • Metamorphic test methods
  • Practical tests for systems with AI
  • Selection of suitable test methods for AI systems
  • Test platforms for intelligent systems
  • Digital environments for testing smart systems
  • Intelligent systems in testing
  • AI-supported evaluation of error reports
  • AI-based creation of test cases
  • AI optimization of regression tests
  • AI methods for error prediction
  • AI testing of graphical user interfaces

Frequently asked questions

  • AI testing refers to the systematic testing of AI-based systems and machine learning models. The focus is on data quality, model behavior, bias, traceability, risk analysis, and special testing methods for learning systems.
  • Key topics include the fundamentals of AI and machine learning, testing strategies for AI systems, data validation, bias detection, model evaluation, testing environments, and risks and quality assurance for AI-based applications.
  • Basic knowledge of software testing is recommended, as well as ISTQB Foundation certification, ideally. Basic knowledge of data analysis, machine learning, or AI concepts facilitates understanding of the test methods.
  • This topic is relevant for software testers, test managers, QA engineers, data scientists, AI engineers, developers, and those responsible for quality assurance in AI or data-driven projects.
  • AI systems make autonomous decisions and are based on training data. Testing procedures must therefore evaluate data quality, model behavior, fairness, robustness, and transparency in order to reduce risks, wrong decisions, and bias.
  • The focus is on data dependency, non-deterministic results, and learning models. Classic testing methods are expanded to include procedures for data validation, model evaluation, bias analysis, and continuous quality assurance.
  • Certification documents expertise in testing strategies for AI-based systems. Organizations benefit from structured quality assurance, while professionals demonstrate competence in a rapidly growing technology field.
  • Areas of application include machine learning, such as automated decision-making processes, image recognition, speech processing, and predictive analytics. The goal is reliable quality, secure systems, and traceable AI results.
Berlin
Bremen
Dortmund
Dusseldorf
Frankfurt
Hamburg
Hanover
Cologne
Leipzig
Munich
Nuremberg
Stuttgart

Do you have any further questions? Please contact us.