Welcome screen
Teacher pre-flight
Lesson plan
Resources
Projector mode
Adversarial Testing
Break your bot, then fix it β attack JARVIS with adversarial prompts and patch the cracks.
Today's run
1
Adversarial Testing
Classify real, attributed jailbreak prompts into the OWASP LLM Top-10 2025 attack buckets against a per-attack countdown, scoring speed times accuracy on a live leaderboard, so students activate the schema they already hold before being taught how the attacks work.
2
Adversarial Testing
3
Red-Team the Agent
Learn to attack your own agent to find its failure modes, to harden your penpal before strangers do, in the context of producing the bug log for your build.
4
The Hardened Agent
How do you design novel attacks that expose failure modes in your own agent, then harden it before users find the cracks?
Start lesson
β
KIT
0/0
Seoul AI 2026
Home
Calendar
Course
Projects
Reading List
About
Library
Theming
Search
βK
Atompunk
