De-identification 한계를 넘는 Foundation Model의 Memorization 취약점 분석
De-Identified and Still Exposed
De-Identified and Still Exposed
When AI Reads Blueprints: The Hidden Attack Surface of Multimodal Engineering Intelligence
Automate LLM Red Team Campaigns with PyRIT
How I Reduced Prompt Injection Attacks by 86% With My Own Framework (And What Went Wrong the First Time)
How AI Penetration Testing Helps Prevent Adversarial Attacks and Data Poisoning
Why CAPTCHA and Traditional Verification Methods Are Failing
Just like phishing for gullible humans, prompt injecting AIs is here to stay
SynthID 워터마크의 1비트 신호 분석 및 역공학 기반 제거 시도
I Built a Tool to Detect Hidden Prompt Injections in PDFs. Here's What I Learned.