Learn to spot an AI failure and explain a release decision — free.
No signup and no programming prerequisite. Three complete lessons introduce the language in plain terms, let you examine a concrete failure, and show how the same reasoning applies to security.
Start with the decision. Code is an optional extension for readers who also want to implement the checks.
Start here
Read in your browser. No statistics, evaluation, machine-learning, or Python background is required.
Before you start
Learn why an AI answer can look convincing and still fail, what evidence means in this context, and how a judge, evaluation, and gate support a decision. Technical terms are introduced from zero.
Read the primer → 0.1 · THE METHODStart from a failure you can recognize
Compare several AI answers, identify what went wrong, and decide which failure should stop a release. An optional implementation section then shows how the check can run repeatedly.
Read lesson 0.1 → SEC.1 · THE METHOD, APPLIEDThe threat map
The same method pointed at security: map an LLM app's attack surface to the OWASP Top 10 for LLM Applications (OWASP-LLM) and spot the blind spots — the first move of the Securing LLM Apps track.
Read lesson SEC.1 →Want to make the full decision?
The full line is 8 tracks · 58 lessons · 57 runnable, offline labs · a research brief per track — a downloadable kit you own, not a stream. Explore the syllabus while launch details are finalized.
See the full syllabus →