Skip to main content

Threat model an AI system

Draw the data flow of an AI system, find threats to each part with STRIDE and MITRE ATLAS, rank them, and turn the top ones into controls and tests.

Pro 1 · Quality engineerabout 3 hours5 sections

Quick answer

Threat modelling an AI system means drawing how data flows through it, listing what can go wrong at each part with STRIDE and MITRE ATLAS, ranking the risks and giving each a control and a test.

Why it matters

Most AI security failures are design failures that a one-hour threat model would have caught. Security teams expect AI engineers to bring one to every design review, and interviewers ask you to do one live.

Sign up free to read the whole lesson

The rest of this lesson: 5 parts, video lessons, the best course, practice and a quick check. It is free, takes a few seconds and needs no card.

Accounts are for learners aged 18 and over. Under 18? The quick answer and the first part of every lesson stay open: browse lessons by topic.

On the roadmap: AI Security Engineer

SecurityPro 1: Quality engineerVideo lessonsThreat modellingSTRIDEMITRE ATLASPrompt injectionData exfiltrationLeast privilege

Track

Today's AI turns are used

Lessons, roadmaps, videos, practice cards and the code playground stay free with your account, with no limit.