AI Governance ConferenceDec. 10, 12-3pm ETSave your seat →
Valid credential
Threat Modeling AI Systems badge

Earn this badge

Threat Modeling AI Systems is a free course from AI Security University.

Take the course

This certifies that

Kirill Kreicberg Stepanov

has successfully completed

Threat Modeling AI Systems

Threat Modeling AI Systems teaches how to find design flaws in LLM, RAG and agent systems before they become incidents. It follows Adam Shostack's four threat-modeling questions and one test: assume the model is wrong, and ask what keeps the system safe. Badge earners can map an AI system and mark where trust changes, with a component diagram, a request trace and a trust-boundary diagram. They can tell locks (checks in code that hold when the model is wrong) from extras that only make mistakes less likely, and trace whose authority each action carries. They can threat-model a complete system, prove the controls hold with the model on the attacker's side, and record what risk is left. This course is taught by Arian Gharedaghi.

Skills

AI SecurityThreat ModelingSecure by DesignLarge Language Models (LLM)RAG SecurityAI Agent SecurityApplication SecurityIdentity and Access Control

Issued

November 17, 2026

Training hours

4

Requirement

Complete all course lessons and pass the final assessment with 80% or higher

Final assessment

Passed

Credential ID

77571cfa-4003-481a-ad3c-a14093979cf4

Is this your certificate?

We’ll send it to the email this certificate is on file for.