Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications

Large Language Models LLMs have become the predominant paradigm in NLP, advancing both research and industry. As model sizes and pretraining data grow, concerns about Pretraining Data Exposure PDE increase due to the scale and opacity of training datasets. PDE refers to determining whether specif...

OpenRT: An Open-Source Red Teaming Framework for Multimodal LLMs

The rapid integration of Multimodal Large Language Models MLLMs into critical applications is increasingly hindered by persistent safety vulnerabilities. However, existing red-teaming benchmarks are often fragmented, limited to single-turn text interactions, and lack the scalability required for...

Adios 2025, you won’t be missed

Welcome to this week's edition of the Threat Source newsletter. For us in America, we're in the holiday doldrums and things slow and/or shut down until the new year. At Cisco, we shut down the last week of the year to reset and recharge, and I've grown to be quite fond of it. I've worked plenty o...

EUVD-2019-1138

Malware in sbrugna...

EUVD-2005-2891

Malware in sbrugna...

EUVD-2009-3554

Malware in sbrugna...

EUVD-2025-7193

Malicious code in bioql PyPI...

Linux Distros Unpatched Vulnerability : CVE-2025-50078

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - Vulnerability in the MySQL Server product of Oracle MySQL component: Server: DML. Supported versions that are affected are 8.0.0-8.0.42, 8.4.0-8.4.5 and...

Enhancing Jailbreak Attacks on LLMs Via Persona Prompts

Jailbreak attacks aim to exploit large language models LLMs by inducing them to generate harmful content, thereby revealing their vulnerabilities. Understanding and addressing these attacks is crucial for advancing the field of LLM safety. Previous jailbreak approaches have mainly focused on dire...

CVE-2025-6948 Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') in GitLab

An issue has been discovered in GitLab CE/EE affecting all versions from 17.11 before 17.11.6, 18.0 before 18.0.4, and 18.1 before 18.1.2 that, under certain conditions, could have allowed a successful attacker to execute actions on behalf of users by injecting malicious content...

CVE-2025-40652 Cross-Site Scripting (XSS) in CoverManager

Stored Cross-Site Scripting XSS vulnerability in the CoverManager booking software. This allows an attacker to inject malicious scripts into the application, which are permanently stored on the server. The malicious scripts are executed in the browser of any user visiting the affected page withou...

PandaGuard: Systematic Evaluation of LLM Safety against Jailbreaking Attacks

Large language models LLMs have achieved remarkable capabilities but remain vulnerable to adversarial prompts known as jailbreaks, which can bypass safety alignment and elicit harmful outputs. Despite growing efforts in LLM safety research, existing evaluations are often fragmented, focused on...

Benchmarking Poisoning Attacks against Retrieval-Augmented Generation

Retrieval-Augmented Generation RAG has proven effective in mitigating hallucinations in large language models by incorporating external knowledge during inference. However, this integration introduces new security vulnerabilities, particularly to poisoning attacks. Although prior work has explore...

SoK: Stealing Cars since Remote Keyless Entry Introduction and How to Defend from It

Remote Keyless Entry RKE systems have been the target of thieves since their introduction in automotive industry. Robberies targeting vehicles and their remote entry systems are booming again without a significant advancement from the industrial sector being able to protect against them...

📄 Ruby on Rails Cross Site Request Forgery

Ruby on Rails appears to include a one time pad for cross site request forgery protections to the user, making it possible to forge valid tokens. Good morning. All current versions and all versions since the 2022/2023 "fix" to the Rails cross-site request forgery CSRF protections continue to be...

QR codes sent in attachments are the new favorite for phishers

Recently we’ve been seeing quite a few phishing campaigns using QR codes in email attachments. The lure and the targets are varied, but the use of a QR code to get someone to visit the phishing site is fast becoming a preferred method for cybercriminals. There are several reasons why cybercrimina...

Linux Distros Unpatched Vulnerability : CVE-2022-34301

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - A flaw was found in CryptoPro Secure Disk bootloaders before 2022-06-01. An attacker may use this bootloader to bypass or tamper with Secure Boot protections. I...

Linux Distros Unpatched Vulnerability : CVE-2018-10845

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - It was found that the GnuTLS implementation of HMAC-SHA-384 was vulnerable to a Lucky thirteen style attack. Remote attackers could use this flaw to conduct...

3 New State-Backed Gangs Target Govt Sectors with HEAT Attack Methods

Global cyber gangs are evolving rapidly, wielding advanced techniques and enjoying state sponsorship. Menlo Security’s latest report exposes…...

New Fickle Stealer Exploits Software Flaws to Steal Crypto, Browser Data

Fortinet's FortiGuard Labs exposes the Fickle Stealer, a malware using multiple attack methods to steal logins, financial details, and more. Learn how to protect yourself from this evolving threat...