Category: GBHackers Security | #1 Globally Trusted Cyber Security News Platform

GPT-5 Compromised Using Echo Chamber and Storytelling Exploits

Cybersecurity researchers have successfully demonstrated a new jailbreaking technique that compromises OpenAI’s GPT-5 model by combining “Echo Chamber” algorithms with narrative-driven manipulation, raising fresh concerns about the vulnerability of advanced AI systems to sophisticated exploitation methods. Novel Attack Vector Emerges…

ClickFix macOS Malware Targets User Login Credentials

Security researchers have identified a new malware campaign targeting macOS users through a sophisticated ClickFix technique that combines phishing and social engineering to steal cryptocurrency wallet details, browser credentials, and sensitive personal data. The Odyssey Stealer malware, discovered by X-Labs…

Over 5,000 Fake Online Pharmacies Caught Selling Counterfeit Medicines

Researchers at Gen have uncovered a vast network of over 5,000 fraudulent online pharmacy domains operated by a single cybercriminal entity dubbed MediPhantom. This elaborate PharmaFraud operation exploits advanced techniques including domain hijacking, search engine optimization manipulation, and AI-generated content…

Axis Camera Server Vulnerabilities Expose Thousands of Organizations to Attack

Claroty’s Team82 research unit has unveiled four vulnerabilities affecting Axis Communications’ widely deployed video surveillance ecosystem, potentially endangering thousands of organizations worldwide. These flaws, centered on the proprietary Axis.Remoting communication protocol, enable pre-authentication remote code execution (RCE) on key components…

Windows UAC Bypass Exploits Character Map Tool for Privilege Escalation

Cybersecurity researchers have uncovered a new technique that allows attackers to bypass Windows User Account Control (UAC) protections by exploiting an unexpected vulnerability in the system’s Private Character Editor tool, potentially granting unauthorized administrative privileges without user consent. The exploit…