Learn how attackers manipulate AI models through malicious inputs, and how to stop them.

This module covers one of the most prevalent AI attack classes: prompt injection. Learners begin with direct injection techniques before moving on to jailbreaking and instruction smuggling via external content. The module then shifts to the defensive side, covering hardening techniques including filters, guards, and template isolation. Two challenge rooms provide hands-on red experience, reinforcing both attack recognition and mitigation strategies.

Prompt Injection

Continue your AI security journey by learning about prompt injection.

Jailbreaking

Explore jailbreaking techniques to bypass an AI model’s safety and policy restrictions.

Prompt Defence

Learn defence measures that can be taken against attacks like prompt injections and jailbreaking.

LLMborghini

Put your indirect prompt injection skills to the test in this AI security challenge.

White Rabbit

Hello, Mr Anderson. Care to put your prompt injection skills to the test?

Topic Rewind Recap

Lock in what you learned with a recap. Earn points and keep your streak.

Next steps

AI Supply Chain Security

Explore how AI model artefacts and supply chains can be compromised and how to defend them.

What are modules?

A learning pathway is made up of modules, and a module is made of bite-sized rooms (think of a room like a mini security lab).

Hierarchical diagram showing how learning pathways contain modules, which contain individual rooms.