Jailbreaking Leaves a Trace: Understanding and Detecting Jailbreak Attacks from Internal Representations of Large Language Models

Jailbreaking large language models LLMs has emerged as a critical security challenge with the widespread deployment of conversational AI systems. Adversarial users exploit these models through carefully crafted prompts to elicit restricted or unsafe outputs, a phenomenon commonly referred to as...

Warning: Two Dangerous Ransomware Are Back – Protect Your Computers

Ransomware has been around for a few years but has become an albatross around everyone's neck—from big businesses and financial institutions to hospitals and individuals worldwide—with cyber criminals making millions of dollars. In just past few months, we saw a scary strain of ransomware attacks...

Mamba Ransomware Encrypts Hard Drives Rather Than Files

Just when we thought ransomware’s evolution had peaked, a new strain has been discovered that forgoes the encryption of individual files, and instead encrypts a machine’s hard drive. The malware, called Mamba, has been found on machines in Brazil, the United States and India, according to...

Dating for everyone – Mamba! - Customized SSL, Dangerous filesystem permissions, Redefined SSL Common Names verifier vulnerabilities

HackApp vulnerability scanner discovered that application Dating for everyone – Mamba! published at the 'play' market has multiple vulnerabilities...