I study how AI systems get attacked, and I write about it so it makes sense to people who aren't in security.
About
I'm Swaroop Morajkar — an M.Sc. Cybersecurity student with a Computer Engineering background, focused entirely on one problem: how large language models and AI applications break, and what that means for the people building and using them. Prompt injection, data exfiltration, model manipulation, real CVEs — I dig into the mechanics, then write it up in language that doesn't require a security clearance to follow.
If you've ever read a threat report on an AI incident and wished someone would just explain what actually happened and why it matters — that's what this is for.