Adversarial AI Researcher.
My internationally recognized work has informed information security policy for governments and companies in both hemispheres.
Speaker at Hackers on Planet Earth 16 in Queens, NY in 2025.
Credited discoverer of 7 unique exploits that function across 12+ commercial LLMs.
@thomasfuchs Heh. Last month I actually hacked OpenAIs customer service support agent via email.
It's all half melted cheese, in terms of its ability to resist attacks.
At some point I should probably write up a thing on Topic Butchering and User Alignment and how it effectively renders any LLM defenseless over enough of an interaction window.
Today on Adversarial AI Researcher Jeopardy for $500: US, Guyana, UK, Australia, Canada, and New Zealand cybersecurity posture all benefitted from this researcher's classification of emergent property based vulnerabilities in Large Language Model AI systems.
The recording of the day I spoke at #HOPE_16 at St. John's University in NYC: https://www.youtube.com/live/6mI-8ias7Dw?si=DjOhx2wcvEg2YqOv
There's a lot of excitement about updates to Gemini. I can't help but mention to folks that Gemini is terrible at safety. Remember the time @Zigula and I hacked it via Fortnite using just voice chat? We called it the Kyber #jailbreak and it produced instructions for napalm production, baking key like pie, and counting cards. #ai #LLMs #cybersecurity
Stjep