AttackEval: A Systematic Empirical Study of Prompt Injection Attack Effectiveness against Large Language Models

Prompt injection has emerged as a critical vulnerability in large language model LLM deployments, yet existing research is heavily weighted toward defenses. The attack side -- specifically, which injection strategies are most effective and why -- remains insufficiently studied.We address this gap...

Romance scammer deepfakes Mark Ruffalo to con elderly artist

Deepfakes have settled into a groove, as most scam techniques do. It seems most deepfakers have decided to make as much cash as possible from unsuspecting victims instead of doing anything particularly earth-shattering with their technology. One curious twist we may not have seen coming is the...

How Instagram scammers talk users out of their accounts

If youve dealt with a scammer, youll know that making up stories is their bread and butter. Think about it: Just when you thought youd heard all the infamous 419 scam backstories, scammers surprise you with a "stuck astronaut" scam, something so utterly hilarious, nonsensical, and otherworldly th...

Party Like a Russian, Carder’s Edition

"It takes a certain kind of man with a certain reputation To alleviate the cash from a whole entire nation…" KrebsOnSecurity has seen some creative yet truly bizarre ads for dodgy services in the cybercrime underground, but the following animated advertisement for a popular credit card fraud shop...

Social engineering attacks: What makes you susceptible?

We now live in a world where holding the door open for someone balancing a tray of steaming hot coffee—she can’t seem to get her access card out to place it near the reader—is something we need to think twice about. Courtesy isn’t dead, mind you, but in this case, you'd almost wish it were. Becau...