Six courses, from the fundamentals through to adversarial machine learning. Each one teaches the material first, then assesses it. Free, marked on the spot, and a certificate with a verifiable serial when you pass.
Proves a working grasp of the AI attack surface, core risks, and the baseline controls every AI deployment needs.
6 modules 43 min read 16 questions 70% to pass 25 min limit
Proves the ability to design LLM and agent systems that stay safe when the model itself is successfully manipulated.
8 modules 60 min read 18 questions 70% to pass 30 min limit
Proves the ability to govern personal and confidential data across training, retrieval and inference without relying on privacy theatre.
7 modules 49 min read 18 questions 75% to pass 30 min limit
Proves you can design, build and ship LLM-backed systems that stay safe under adversarial pressure.
8 modules 50 min read 19 questions 75% to pass 35 min limit
Proves you can build AI governance that produces real assurance evidence, not paperwork theatre.
7 modules 51 min read 18 questions 75% to pass 35 min limit
The hardest exam in the ladder: proves expert judgement on adversarial attacks, defence evaluation and red team practice.
8 modules 54 min read 20 questions 80% to pass 40 min limit
Each course opens with taught material: concepts, worked scenarios and the traps people actually fall into. Read it or skip it, your choice.
Drawn from exactly that material. Questions and answers are shuffled every sitting, and marking is exact: on multi-answer questions you need the whole set.
Your score straight away, and which questions you got right. Pass, and you get the full reasoning plus a certificate with a verifiable serial you can share.
There is nothing to sign up for and no account to create. If you pass and want the certificate, we ask for your name, which is printed on it, and an email address, which is where we send it. Nothing before that point, and nothing at all if you would rather not.