Defensive
Evolution of LLM Defense
Every defense is a reflex of the prior attack and at every step we keep rediscovering classical systems security. The agentic frontier is where the stakes are now highest.
2026-06 Read slides
u9up:~$ whoami
We break it, we fix it, we write about it.
U9UP /9up/ verb
An independent AI security community from Malaysia. We publishes hands-on research, raising awareness, and sharing what we learn.
Research
Latest from the lab
Deep dives into AI security blind spots, written by people who actually break them.
Offensive
Breaking AI: An Offensive Perspective
AI attacks have evolved alongside the evolution of AI itself, from early chatbots to today’s autonomous AI agents.
2026-05 Read slides
What We Do
Two sides of the same coin
AI security isn't just breaking things or defending them, it's both. We research across the full spectrum.
Offensive Security
Finding the cracks before the bad actors do. We research attack vectors, test guardrails, and publish what we find so the community can learn.
Prompt injectionJailbreaksAdversarial MLModel extractionSupply Chain attacksAgent-In-The-Middle
Defensive Security
Building resilience into AI systems. We study detection, guardrail architectures, and monitoring strategies that actually work in production.
Guardrail designInput validationAI observabilityThreat detectionModel hardeningCanary Token
Gallery
Moments & meetups
Workshops, talks, and the occasional makan session.
7 photos
Get Involved
u9up:~$ ./join --community
Come hang with us.
Whether you're deep into adversarial ML or just getting curious about AI security — pull up a chair. No gatekeeping.
Research & write-ups
Community discussions
No gatekeeping, just good vibes