EUVD-2025-35358

Connection desynchronization between an HTTP proxy and the model backend. The fixes were rolled out for all proxies in front of impacted models by 2025-09-28. Users do not need to take any action...

PT-2025-43062

Name of the Vulnerable Software and Affected Versions versions prior to 2025-09-28 Description A desynchronization issue exists between an HTTP proxy and the model backend. This affects communication within the model backend. Approximately 10,000 devices worldwide are potentially affected. The...

Google Cloud Vertex AI API 安全漏洞

Google Cloud Vertex AI API is a set of interfaces for machine learning platforms from Google, Inc USA. A security vulnerability exists in the Google Cloud Vertex AI API that stems from an unsynchronized connection between the HTTP proxy and the model backend...

PT-2025-43330

Name of the Vulnerable Software and Affected Versions D-Link DIR-823G A1 version 1.0.2B05 Description The D-Link DIR-823G A1 device version 1.0.2B05 contains a buffer overflow in the FillMacCloneMac parameter of the /EXCU SHELL API endpoint. A crafted input to this parameter can lead to a Denial ...

You can poison AI with just 250 dodgy documents

Researchers have shown how you can corrupt an AI and make it talk gibberish by tampering with just 250 documents. The attack, which involves poisoning the data that an AI trains on, is the latest in a long line of research that has uncovered vulnerabilities in AI models. Anthropic which produces...

EUVD-2018-21605

GeoVision embedded IP devices, confirmed on GV-BX1500 and GV-MFD1501, contain a remote command injection vulnerability via /PictureCatch.cgi that enables an attacker to execute arbitrary commands on the device. VulnCheck has observed this vulnerability being exploited in the wild as of 2025-10-19...

Evaluating Large Language Models in Detecting Secrets in Android Apps

Mobile apps often embed authentication secrets, such as API keys, tokens, and client IDs, to integrate with cloud services. However, developers often hardcode these credentials into Android apps, exposing them to extraction through reverse engineering. Once compromised, adversaries can exploit...

The Attribution Story of WhisperGate: An Academic Perspective

This paper explores the challenges of cyberattack attribution, specifically APTs, applying the case study approach for the WhisperGate cyber operation of January 2022 executed by the Russian military intelligence service GRU and targeting Ukrainian government entities. The study provides a detail...

Prompting the Priorities: A First Look at Evaluating LLMs for Vulnerability Triage and Prioritization

Security analysts face increasing pressure to triage large and complex vulnerability backlogs. Large Language Models LLMs offer a potential aid by automating parts of the interpretation process. We evaluate four models ChatGPT, Claude, Gemini, and DeepSeek across twelve prompting techniques to...

EUVD-2025-35078

Reolink Video Doorbell WiFi DB566128M5MPW allows root shell access through an unsecured UART/serial console. An attacker with physical access can connect to the exposed interface and execute arbitrary commands with root privileges...

CVE-2025-3465

The CVE-2025-3465 entry concerns an improper limitation of a pathname to a restricted directory (path traversal) affecting ABB CoreSense HM and CoreSense M10. Public-quoted details from multiple feeds specify the affected products and versions: CoreSense HM up to 2.3.1 and CoreSense M10 up to 1.4...

CVE-2025-60856

Reolink Video Doorbell WiFi DB566128M5MPW allows root shell access through an unsecured UART/serial console. An attacker with physical access can connect to the exposed interface and execute arbitrary commands with root privileges. NOTE: this is disputed by the Supplier because of "certain...

MINI-H8FV-R22M-9768

Bulletin has no description...

MINI-G4Q5-38VQ-M34W

Bulletin has no description...

131 Chrome Extensions Caught Hijacking WhatsApp Web for Massive Spam Campaign

Cybersecurity researchers have uncovered a coordinated campaign that leveraged 131 rebranded clones of a WhatsApp Web automation extension for Google Chrome to spam Brazilian users at scale. The 131 spamware extensions share the same codebase, design patterns, and infrastructure, according to...

oatpp-mcp 安全特征问题漏洞

oatpp-mcp is an Oat++ open source implementation of the Model Context Protocol. A security signature issue vulnerability exists in oatpp-mcp that stems from an MCP SSE endpoint returning an instance pointer as a session ID, which could lead to a session hijacking attack...

Multimodal Safety Is Asymmetric: Cross-Modal Exploits Unlock Black-Box MLLMs Jailbreaks

Multimodal large language models MLLMs have demonstrated significant utility across diverse real-world applications. But MLLMs remain vulnerable to jailbreaks, where adversarial inputs can collapse their safety constraints and trigger unethical responses. In this work, we investigate jailbreaks i...

JLSEC-2025-109 FFmpeg through 4.3 has a heap-based buffer overflow in `avio_get_str` in `libavformat/aviobuf.c`...

FFmpeg through 4.3 has a heap-based buffer overflow in aviogetstr in libavformat/aviobuf.c because dnnbackendnative.c calls ffdnnloadmodelnative and a certain index check is omitted...

When AI Takes the Wheel: Security Analysis of Framework-Constrained Program Generation

In recent years, the AI wave has grown rapidly in software development. Even novice developers can now design and generate complex framework-constrained software systems based on their high-level requirements with the help of Large Language Models LLMs. However, when LLMs gradually "take the whee...

Toward Understanding Security Issues in the Model Context Protocol Ecosystem

The Model Context Protocol MCP is an emerging open standard that enables AI-powered applications to interact with external tools through structured metadata. A rapidly growing ecosystem has formed around MCP, including a wide range of MCP hosts i.e., Cursor, Windsurf, Claude Desktop, and Cline, M...