InInTheCyber PostsbyMarco Pedrinazzi·Aug 24Prompt, Publish, Phish: In-the-Wild Abuse of the Framer AI Website BuilderTL;DR AI website builders are legitimate tools that make creating and publishing websites faster and more accessible, but the same…
Suresh Shekar·Aug 18What If the AI Evaluating the AI Can Be Manipulated?Thinking about evaluators, critics, and a different kind of AI attack surface.
InSecurityBreak ResearchbyMarco Pedrinazzi·Jul 16The Payload Is in the HeaderA New Prompt Injection Blind SpotA response icon1
InInTheCyber PostsbyMarco Pedrinazzi·Jul 6Tales of an Ollama Honeypot: LLMjacking for SSH Credential Recon (Part 4)At the end of March, I built a honeypot that emulates an Ollama server and deployed it on a VPS. This is the fourth article in the series…
InInTheCyber PostsbyMarco Pedrinazzi·Jun 8Tales of an Ollama Honeypot (Part 3): More Traffic, More FindingsAt the end of March, I built a honeypot that emulates an Ollama server and deployed it on a VPS. This is the third article in the series…
