Sophos AI to present on how to defang malicious AI models at Black Hat Europe
“LLMbotomy” research reveals how Trojans can be injected into Large Language Models, and how to disarm them.
Similar Posts
The day after XSS.is forum was seized, it struggles to come back online — but is it really them?
Bys sEver since law enforcement announced the arrest of an administrator of the XSS.is forum, forum members watched threads disappear from the site, and then a seizure notice splash screen appeared. No administrator or moderator had made any statement about the arrest or situation despite pleas from forum members for some clarification, and attempts to discuss…
France’s Waltio faces ransom threat from notorious hacker collective
Bys sCaroline Alvarez reports: Waltio, a French crypto tax platform, is under siege from ShinyHunters, a notorious ransomware group claiming to hold the personal data of nearly 50,000 users. ShinyHunters, known for high-profile crypto and corporate hacks, is threatening to leak users’ 2024 tax reports unless a ransom is paid. Waltio says its services and production……
New TeleMessage SGNL Flaw Is Actively Being Exploited by Attackers
Bys sWaqas reports: TeleMessage SGNL, a made-in-Israel clone of the Signal app used by US government agencies and regulated businesses, has been found running with an outdated configuration that exposes sensitive internal data to the internet, no login required. The main cause of the problem is how some deployments of TeleMessage SGNL are using older versions…
Tech Overtakes Gaming as Top DDoS Attack Target, New Gcore Radar Report Finds
Bys sThe latest Gcore Radar report analyzing attack data from Q1–Q2 2025, reveals a 41% year-on-year increase in total attack volume. The largest attack peaked at 2.2 Tbps, surpassing the 2 Tbps record in late 2024. Attacks are growing not only in scale but in sophistication, with longer durations, multi-layered strategies, and a shift in target…
Grupo NGO gains 24/7 security visibility without expanding its IT team
Bys sTags: Case Study, MDR, Retail
PLAYFULGHOST Delivered via Phishing and SEO Poisoning in Trojanized VPN Apps
Cybersecurity researchers have flagged a new malware called PLAYFULGHOST that comes with a wide range of information-gathering features like keylogging, screen capture, audio capture, remote shell, and file transfer/execution. The backdoor, according to Google’s Managed Defense team, shares functional overlaps with a known remote administration tool referred to as Gh0st RAT, which had its source