ISTQB CT-AI v2.0 Practice Tests: Test Models & GenAI

Categories: Software Testing
Share

About Course

This course contains the use of artificial intelligence.

A strong aggregate score can hide subgroup failures, and unchanged application code can still produce different decisions after a model replacement. These ISTQB AI Testing CT-AI v2.0 practice tests help you identify what the evidence establishes and which test addresses the remaining risk.

Test the complete AI-based system

Practise AI quality, safety and acceptance criteria; training and test data; metrics, bias, representativeness and preprocessing; strategy, oracles and generated-output evaluation. Model and deployment cases connect adversarial inputs, changing artifacts, surrounding applications and operational behaviour. Questions ask you to trace versions and conditions rather than treat one score as a complete assurance claim.

Every option has feedback, with an overall explanation and study guidance. Use these to distinguish a model issue from a data or application issue, select a meaningful transformation or comparison, and explain why another test would leave the stated failure unexamined.

The 300 questions comprise four focused banks and two 40-question mocks with 60-minute timers. Begin with a weak testing stage, review guesses alongside mistakes, then attempt a mixed mock without notes after revisiting the relevant principle.

The scope is CT-AI v2.0. CTFL knowledge is recommended and the official exam has its own Foundation prerequisite. These written questions do not include a machine-learning laboratory or hands-on training activities. The equal-weight 75% study target does not reproduce official weighted scoring. This independent resource contains no official examination questions or ISTQB endorsement.

Show More

TRY BEFORE YOU CHOOSE

Three questions. A clearer next step.

Explore original questions and explanations. No account or payment required.

What Will You Learn?

  • Distinguish AI technologies and evaluate model, hardware and hosting choices against the intended use.
  • Translate AI quality and safety requirements into observable acceptance conditions.
  • Interpret ML workflows, data partitions, preprocessing and neural-network coverage evidence.
  • Calculate and select accuracy, precision, recall and F1 using the correct populations and denominators.
  • Evaluate input-data usefulness, bias, representativeness, constraints and label correctness.
  • Select test oracles and evidence for adaptive, probabilistic and generative AI systems.
  • Choose controlled red-team inputs, adversarial tests and valid metamorphic follow-ups.
  • Interpret drift, fitting, model comparisons and deployment recovery evidence.

Course Content

Practice tests

  • Foundations and AI Quality | Focused Practice
  • ML, Metrics and Input Data | Focused Practice
  • Test Strategy and GenAI | Focused Practice
  • Model and Deployment Testing | Focused Practice
  • Mixed Mock 1 | 40 Questions | 60 Minutes
  • Mixed Mock 2 | 40 Questions | 60 Minutes