As AI assistants and agents do more, they bring new security risks. These lessons explain LLM security in approachable language, prompt injection, jailbreaks, data leakage, and the defenses that keep AI systems trustworthy. Best for intermediate learners and grown-ups who want to understand AI risk.
What Is Sensitive Information?
Private info to share only with safe, trusted people.
Read the lesson →
What Is Data Exfiltration?
When sneaky attackers steal data and send it out without permission.
Read the lesson →
🧪 Interactive
What Is RAG?
How the robot looks things up before it answers.
Read the lesson →
What Are AI Agents?
AI that can take actions and use tools to get things done.
Read the lesson →
What Is Hallucination?
When an AI sounds sure but makes something up.
Read the lesson →
What Are Guardrails?
The rules and filters that keep an AI's answers safe.
Read the lesson →
What Is Red Teaming?
Friendly testers find weak spots so we can fix them.
Read the lesson →
What Is AI Model Evaluation?
How we test whether an AI is helpful, safe, and working as we want.
Read the lesson →
What Is Misinformation?
Wrong information that spreads, by accident or on purpose.
Read the lesson →
🧪 Interactive
What Is Prompt Injection?
When sneaky instructions try to boss the AI around.
Read the lesson →
Indirect Prompt Injection
When sneaky hidden instructions in a page or file trick an AI.
Read the lesson →
LLM Jailbreaking
When someone tries to trick an AI past its safety rules.
Read the lesson →