AI Testing Theory · ISTQB CT-AI

AI doesn't break like normal software.

So testing it is a different craft, and this is where you learn the theory of it: how these systems actually go wrong - bad data going in, behaviour that quietly drifts once it's live - and how you test for that. The knowledge is the point. The ISTQB CT-AI certificate is the bonus: learn this properly and you can pass the exam without cramming, or skip it entirely - the skill is yours either way.

Start the course Try 12 lectures, free
How it works

Closer to a game than a textbook.

A lecture in fifteen minutes, about thirty with the quiz - and the questions get mean on purpose.

Step 1 · Learn

Watch the lecture

One idea, five to fifteen minutes, explained the way you'd explain it to a colleague over lunch. No ML degree assumed.

Step 2 · Practice

Answer a few questions

Straight after, while it is still fresh. Get one wrong and it tells you why, not just "nope, try again".

Step 3 · Ask Kai

Your AI mentor, 24/7

Stuck at 11pm on what "fairness" even means for your model? Ask Kai in Telegram. He answers, any hour.

Bonus · Exam

Sit the cert, if you want

There is a full mock exam to drill against. You book the real ISTQB one yourself - the paper is a nice bonus on top of knowing your stuff.

Curriculum

Eight weeks.

Thirteen modules: an intro, eleven that follow the official ISTQB CT-AI syllabus, and a final assessment. It opens with what these systems actually are and where they go wrong, then moves into how you test for that. One lecture a day; the mock exam at the end runs under real timed conditions.

Full syllabus →
  • M-00IntroductionStart here
  • M-01Introduction to AI TestingFoundations
  • M-02Quality characteristics for AI-based systemsISO 25059
  • M-03Machine learning - overviewRead a model card
  • M-04Machine learning - dataLeakage, splits
  • M-05ML functional performance metricsConfusion matrix
  • M-06Neural networks and testingAdversarial
  • M-07Testing AI-based systems - overviewStrategy
  • M-08Testing AI-specific quality characteristicsBias, drift
  • M-09Methods & techniques for testing AIMetamorphic, A/B
  • M-10Test environments for AI-based systemsData, isolation
  • M-11Using AI for testingTooling
  • M-12Final assessmentMock exam
Your AI mentor

Meet Kai.

Not a chatbot bolted on. A mentor who remembers you, answers out loud, and writes first.

A real question, a real answer
You

explain metamorphic testing in 3 lines, simple

Kai

You can't always know the exact right output, but you know how it should change. Rotate a photo and a good image classifier should still say "cat" - same label, new input. When it doesn't, you found a bug without ever needing a "correct" answer. We drill this in Module 9.

  • He keeps track. Finished lecture 23 on Monday? On Thursday he picks up where you left off - your progress, quiz scores and weak spots ride along, no re-explaining.
  • He speaks. Ask out loud - voice in, voice out, in his own voice, any hour.
  • He writes first. About to break a study streak? He nudges before it snaps, not after.
The "What Kai remembers" page: every memory Kai keeps about a student, each one deletable
What Kai remembers - and you can delete any of it. Nothing hidden.
Hear him · 15 sec

AI-synthesized voice - a clone built from a real voice sample.

Pricing · simple

Two doors.

- Free taster

The first twelve.

free
No card required · start in 30 seconds
  • ◆The first 12 lectures, end to end
  • ◆Practice questions for each
  • ◆Ask Kai, the AI mentor, 3 questions a day
  • ◆A taste of the mock exam
Start free
If it is not for you · zero cost
Focus group · public posts

What the testers said publicly.

Forty-six QA folks went through the free Module 1 before launch. These are the ones who wrote about it on LinkedIn - names link to the original post.

Independent reviews of the ISTQB CT-AI course from the pre-launch focus group of 46 QA engineers who completed the free first module on testing AI systems - covering bias, data drift, adversarial inputs, and the test oracle problem. Each reviewer's name links to their public LinkedIn post about the AI testing course.

Maxim Stelmakh, LinkedIn post about CT-AI course

Maxim Stelmakh

QA · CS Teacher

LinkedIn
 Kateryna Karpenko, LinkedIn post about CT-AI course

Kateryna Karpenko

QA Manual · Health Apps

LinkedIn
 Yevheniia Nikiforova, LinkedIn post about CT-AI course

Yevheniia Nikiforova

QA Engineer

LinkedIn
 Olena Kharkova, LinkedIn post about CT-AI course

Olena Kharkova

Manual QA

LinkedIn
 Oksana Gryshyna, LinkedIn post about CT-AI course

Oksana Gryshyna

QA · ISTQB Certified

LinkedIn
 Anna Stanishevska, LinkedIn post about CT-AI course

Anna Stanishevska

ISTQB · Manual QA

LinkedIn
Anastasia, course author
Who is teaching

Anastasia, QA for 12 years.

I'm a QA lead, not an AI researcher. Someone handed me "the new AI thing" and I had to work out how to test it.

Twelve years in QA, ISTQB Foundation since 2018, and these days I work alongside about twenty other testers. When AI landed on my desk I started building the course I wish someone had handed me. It took two years and around €35,000.

It is theory first: what these systems actually are, and what words like fairness and drift come to mean once the thing is learning on its own. The CT-AI certificate sits on top, for when you want it. The theory is the part that makes people in an AI standup stop and actually listen to you.

- AnastasiaCourse author · QA Tech Lead · ISTQB

AI doesn't break like normal software.

Start the course Try 12 lectures, free