The ISTQB® Certified Tester AI Testing v2.0 (CT-AI) certification provides a comprehensive introduction to testing AI-based systems, with a primary focus on machine learning and including key concepts and testing approaches for generative AI.

It extends core software testing knowledge to address the unique challenges of AI, including non-deterministic behavior, data dependency, probabilistic outcomes, and continuously evolving models. The certification equips professionals with the skills needed to design, execute, and evaluate tests for AI systems in modern development environments.

Version 2.0 replaces the previous CT-AI v1.0 syllabus. The v1.0 certification will be phased out:

• English version: available until 21 April 2027
• Non-English versions: available until 21 October 2027

Candidates are encouraged to pursue v2.0 to align with current industry practices and evolving AI technologies.

Professionals interested in testing using generative AI systems should consider the ISTQB® Certified Tester – Testing with Generative AI (CT-GenAI) certification.

Contents

The ISTQB® CT-AI v2.0 certification covers the following key areas:

Introduction to Artificial Intelligence

• AI-based vs conventional systems
• Types of AI (narrow, general, super AI)
• Overview of AI technologies, including generative AI
• AI development, hosting, and regulatory considerations

Quality Characteristics for AI-Based Systems

• AI-specific quality characteristics (ISO/IEC 25059)
• Functional correctness and adaptability
• Transparency, robustness, and controllability
• Safety and ethical considerations

Machine Learning

• Supervised, unsupervised, and reinforcement learning
• ML workflow and lifecycle
• Data preparation and dataset management
• ML functional performance metrics
• Neural networks fundamentals

Testing AI-Based Systems

• Challenges in testing AI systems
• Statistical and risk-based testing approaches
• Testing generative AI and large language models
• Test levels for machine learning systems

Input Data Testing

• Data quality risks and mitigation
• Bias detection and representativeness
• Data pipeline and dataset validation

Model Testing

• ML model risks and performance testing
• Adversarial and metamorphic testing
• Overfitting, underfitting, and drift
• A/B and back-to-back testing

Machine Learning Development Testing

• Testing during development and deployment
• Monitoring and maintaining model performance in production

Who is the CT-AI v2.0 certification for?

This certification is suitable for:

• Testers, test analysts, and test engineers
• Software developers and data scientists
• Test managers and QA professionals
• Business analysts and project managers
• IT leaders and consultants
• Anyone seeking a foundation in testing AI-based systems

How does the CT-AI v2.0 certification benefit me?

After completing this certification, candidates will be able to:

• Understand the current state of AI, including generative AI.
• Experience the implementation and testing of machine learning models
• Understand the working and testing of simple neural networks.
• Understand the specific AI quality characteristics defined by ISO/IEC 25059.
• Calculate and interpret ML functional performance metrics for machine learning models.
• Recognize the scope and importance of the two test levels that are specific to the testing of machine learning systems.
• Contribute to the development of an effective test strategy for a machine learning system.
• Design and execute test cases for machine learning systems.

Exam Information

• Number of Questions: 40
• Duration: 60 minutes
• Pass Mark: 29 out of 44 points
• Format: Multiple-choice questions

Prerequisites

• Mandatory: ISTQB® Certified Tester Foundation Level (CTFL)
• Recommended: Basic experience in software testing, development, or data-related roles and approximately six months of practical experience

How do I obtain the ISTQB® Certified Tester AI Testing (CT-AI) v2.0 certification?

To be certified, you must pass the ISTQB® Certified Tester AI Testing (CT-AI) v2.0 certification exam.

You can prepare for the exam by:

• Attending an accredited training course for ISTQB® CT-AI v2.0
• Self-study using the official syllabus, sample exams (see “Downloads” section), and the ISTQB® Glossary

You can also test your knowledge without obligation and free of charge by taking an online mock exam before attempting the certification exam.

Frequently Asked Questions

What is the ISTQB® CT-AI v2.0 certification?

The ISTQB® Certified Tester AI Testing v2.0 (CT-AI) certification validates knowledge in testing AI-based systems, with a focus on machine learning, data quality, and AI-specific testing techniques.

Is ISTQB® CT-AI v2.0 focused on generative AI?

No. CT-AI primarily focuses on machine learning systems and AI testing fundamentals. Generative AI is included as part of the broader AI landscape. For a dedicated focus on using GenAI in testing, refer to the ISTQB® Certified Tester – Testing with Generative AI (CT-GenAI) certification.

What is the difference between ISTQB® CT-A v2.0 and ISTQB® CT-GenAI?

CT-AI v2.0 focuses on testing AI-based systems, especially machine learning systems, including data, model, and AI-quality-related testing. CT-GenAI focuses on using generative AI in software testing, including applying LLMs, prompt engineering, and GenAI-supported testing approaches across the test process.

What are the prerequisites for CT-AI v2.0?

• Mandatory: ISTQB® Certified Tester Foundation Level (CTFL) certification ( ISTQB® Certified Tester Foundation Level (CTFL 4) or a previous version)
• Recommended: A minimal background in software development or software testing and approximately six months of relevant practical experience

Do I need AI or programming experience to take the ISTQB® CT-AI v2.0 certification exam?

No prior AI or programming experience is required. However, a basic understanding of software testing or development concepts is beneficial.

How is AI testing different from traditional software testing?

AI testing introduces additional challenges, including:

• Non-deterministic outputs
• Dependence on training data quality
• Lack of clear test oracles
• Continuous model evolution

How long does it take to prepare for the ISTQB® CT-AI exam?

The syllabus is designed for approximately 19.5 hours of training, with additional self-study recommended depending on experience. If you are self-studying, then use the learning time per chapter as a guide.

What topics are covered in the ISTQB® CT-AI v2.0 certification?

The certification covers:

• AI fundamentals and machine learning
• Data preparation and model evaluation
• AI-specific quality characteristics
• Testing techniques for AI systems
• Model and data testing approaches

Is ISTQB® CT-AI v2.0 relevant for real-world AI projects?

Yes. The certification provides practical knowledge for testing machine learning models, data pipelines, and AI system behavior, which can be applied in real-world projects.

Can I take the ISTQB® CT-AI v2.0 exam without training?

Yes. You can prepare through self-study using the syllabus and sample exams (including the online sample), although accredited training can help structure learning and improve exam readiness.

What other certifications can you recommend?

ISTQB® Certifications:

• ISTQB® CTFL 4 (CTFL is a prerequisite for CT-AI, v4.0 or a previous version if you are already certificated)
• ISTQB® Certified Tester - Testing with Generative AI (CT-GenAI)

General AI Certifications:

• A4Q AI Essentials
• A4Q AI Foundation
• A4Q AI Compliance
• IREB Microcredential AI4RE