AI Security
Today
Practice
Library
Exam
Certificate
Search
⌘K
ខ្មែរ
◐
Sign in
Today
Practice
Library
Profile
Today
Track 415
Prompt Injection & Jailbreaks
0 / 4 lessons
Start track
1
Direct Prompt Injection
When the user is the attacker
Intermediate
2
Indirect Prompt Injection via Untrusted Content
The attack that never talks to your model directly
Intermediate
3
Jailbreak Techniques and Why They Work
Reframing, not brute force, breaks most guardrails
Intermediate
4
Defending System Prompts
Layered defenses beat a longer instruction list
Advanced
Prompt Injection & Jailbreaks