13514 matches found
VisualDAN: Exposing Vulnerabilities in VLMs with Visual-Driven DAN Commands
Vision-Language Models VLMs have garnered significant attention for their remarkable ability to interpret and generate multimodal content. However, securing these models against jailbreak attacks continues to be a substantial challenge. Unlike text-only models, VLMs integrate additional modalitie...
New Machine Learning Approaches for Intrusion Detection in ADS-B
With the growing reliance on the vulnerable Automatic Dependent Surveillance-Broadcast ADS-B protocol in air traffic management ATM, ensuring security is critical. This study investigates emerging machine learning models and training strategies to improve AI-based intrusion detection systems IDS...
Bloodroot: When Watermarking Turns Poisonous for Stealthy Backdoor
Backdoor data poisoning is a crucial technique for ownership protection and defending against malicious attacks. Embedding hidden triggers in training data can manipulate model outputs, enabling provenance verification, and deterring unauthorized use. However, current audio backdoor methods are...
Pattern Enhanced Multi-Turn Jailbreaking: Exploiting Structural Vulnerabilities in Large Language Models
Large language models LLMs remain vulnerable to multi-turn jailbreaking attacks that exploit conversational context to bypass safety constraints gradually. These attacks target different harm categories like malware generation, harassment, or fraud through distinct conversational approaches...
PT-2025-41422
Name of the Vulnerable Software and Affected Versions Juniper Networks Junos Space versions prior to 24.1R4 Description An Improper Neutralization of Input During Web Page Generation 'Cross-site Scripting' issue exists in Juniper Networks Junos Space. This allows an attacker to inject script tags...
Severe Framelink Figma MCP Vulnerability Lets Hackers Execute Code Remotely
Cybersecurity researchers have disclosed details of a now-patched vulnerability in the popular figma-developer-mcp Model Context Protocol MCP server that could allow attackers to achieve code execution. The vulnerability, tracked as CVE-2025-53967 CVSS score: 7.5, is a command injection bug...
CVE-2025-61785
Deno is a JavaScript, TypeScript, and WebAssembly runtime. In versions prior to 2.5.3 and 2.2.15, Deno.FsFile.prototype.utime and Deno.FsFile.prototype.utimeSync are not limited by the permission model check --deny-write=./. It's possible to change to change the access atime and modification mtim...
EMPalm: Exfiltrating Palm Biometric Data Via Electromagnetic Side-Channels
Palm recognition has emerged as a dominant biometric authentication technology in critical infrastructure. These systems operate in either single-modal form, using palmprint or palmvein individually, or dual-modal form, fusing the two modalities. Despite this diversity, they share similar hardwar...
PT-2025-41209
Name of the Vulnerable Software and Affected Versions Deno versions prior to 2.5.3 Deno versions prior to 2.2.15 Description Deno is a JavaScript, TypeScript, and WebAssembly runtime. The Deno.FsFile.prototype.stat and Deno.FsFile.prototype.statSync functions do not enforce the --deny-read=./...
The vulnerability of the Open Asset Import Library’s 3D model import library, due to the lack of checks for division by zero, allows a hacker to trigger a service failure.
The vulnerability of the Open Asset Import Library Assimp for importing 3D models is related to the lack of checks for division by zero. Exploiting this vulnerability can allow attackers to cause service failures...
Deno 安全漏洞
Deno is a simple, modern and secure JavaScript and TypeScript runtime environment from Deno Open Source. A security vulnerability exists in Deno versions prior to 2.5.3 and prior to 2.2.15, which stems from insufficient permission model checking and could lead to bypassing the permission model...
Deno 安全漏洞
Deno is a simple, modern and secure JavaScript and TypeScript runtime environment from Deno Open Source. A security vulnerability exists in Deno versions prior to 2.5.3 and prior to 2.2.15, which stems from the fact that the utime and utimeSync methods are not constrained by the privilege model,...
Another Critical RCE Discovered in a Popular MCP Server
Artificial Intelligence development is moving faster than secure coding practices, and attackers are taking notice. Imperva Threat Research recently uncovered and disclosed a critical Remote Code Execution RCE vulnerability CVE-2025-53967 in the Framelink Figma MCP Server. This is just one exampl...
CVE-2023-53667
In the Linux kernel, the following vulnerability has been resolved: net: cdcncm: Deal with too low values of dwNtbOutMaxSize Currently in cdcncmchecktxmax, if dwNtbOutMaxSize is lower than the calculated "min" value, but greater than zero, the logic sets txmax to dwNtbOutMaxSize. This is then use...
EUVD-2025-32853
vLLM is an inference and serving engine for large language models LLMs. Before version 0.11.0rc2, the API key support in vLLM performs validation using a method that was vulnerable to a timing attack. API key validation uses a string comparison that takes longer the more characters the provided A...
CVE-2025-59425 vLLM vulnerable to timing attack at bearer auth
vLLM is an inference and serving engine for large language models LLMs. Before version 0.11.0rc2, the API key support in vLLM performs validation using a method that was vulnerable to a timing attack. API key validation uses a string comparison that takes longer the more characters the provided A...
EUVD-2014-6999
Malware in sbrugna...
EUVD-2021-0934
Malware in sbrugna...