SoK: Are Watermarks in LLMs Ready for Deployment?

Large Language Models LLMs have transformed natural language processing, demonstrating impressive capabilities across diverse tasks. However, deploying these models introduces critical risks related to intellectual property violations and potential misuse, particularly as adversaries can imitate...

CVE-2025-5601

A buffer overflow vulnerability was found in Wireshark. This vulnerability is triggered when a user views a specifically malformed packet or a pcap file with such a malformed packet. Mitigation Mitigation for this issue is either not available or the currently available options do not meet the Re...

CVE-2025-20286

A vulnerability in Amazon Web Services AWS, Microsoft Azure, and Oracle Cloud Infrastructure OCI cloud deployments of Cisco Identity Services Engine ISE could allow an unauthenticated, remote attacker to access sensitive data, execute limited administrative operations, modify system configuration...

CVE-2025-48710

kro Kube Resource Orchestrator 0.1.0 before 0.2.1 allows users with permission to create or modify ResourceGraphDefinition resources to supply arbitrary container images. This can lead to a confused-deputy scenario where kro's controllers deploy and run attacker-controlled images, resulting in...

CVE-2025-48944

A denial of service flaw was found in vLLM. This flaw allows a remote attacker with access to /v1/chat/completions OpenAPI endpoint to submit malformed data in the "pattern" and "type" fields to crash the vLLM instance. Mitigation Mitigation for this issue is either not available or the currently...

CVE-2025-48887

A regular expression denial of service ReDoS flaw was found in vLLM. This vulnerability allows an attacker with prompt access to degrade or deny service in a vLLM instance. Mitigation Mitigation for this issue is either not available or the currently available options do not meet the Red Hat...

How to deploy AI safely

In this blog you will hear directly from Corporate Vice President and Deputy Chief Information Security Officer CISO for AI, Yonatan Zunger, about how to build a plan to deploy AI safely. This blog is part of a new ongoing series where our Deputy CISOs share their thoughts on what is most importa...

DragonForce Exploits SimpleHelp Flaws to Deploy Ransomware Across Customer Endpoints

The threat actors behind the DragonForce ransomware gained access to an unnamed Managed Service Provider's MSP SimpleHelp remote monitoring and management RMM tool, and then leveraged it to exfiltrate data and drop the locker on multiple endpoints. It's believed that the attackers exploited a tri...

Security Bulletin: SPSS Collaboration and Deployment Services is affected by vulnerability in Apache Commons BCEL

Summary SPSS Collaboration and Deployment Services is affected by vulnerability in Apache Commons BCEL. This has been addressed in the remediation section. Vulnerability Details CVEID:CVE-2022-42920 DESCRIPTION: Apache Commons BCEL could allow a remote attacker to bypass security restrictions,...

Exploit for Exposure of Sensitive Information to an Unauthorized Actor in Microsoft

CVE-2025-24071 PoC SMB + TAR Extraction Method This is a Po...

A Flyby on the CFO's Inbox: Spear-Phishing Campaign Targeting Financial Executives with NetBird Deployment

A Flyby on the CFO's Inbox: Spear-Phishing Campaign Targeting Financial Executives with NetBird Deployment By Trellix · May 28, 2025 This blog was written by Srini Seethapathy Trellix wants to acknowledge thequick response from NetBird following our initial findings. NetBird acted immediately to...

Zero-Trust Foundation Models: a New Paradigm for Secure and Collaborative Artificial Intelligence for Internet of Things

This paper focuses on Zero-Trust Foundation Models ZTFMs, a novel paradigm that embeds zero-trust security principles into the lifecycle of foundation models FMs for Internet of Things IoT systems. By integrating core tenets, such as continuous verification, least privilege access LPA, data...

CVE-2024-46942

In OpenDaylight Model-Driven Service Abstraction Layer MD-SAL through 13.0.1, a controller with a follower role can configure flow entries in an OpenDaylight clustering deployment...

CVE-2024-37293

The AWS Deployment Framework ADF is a framework to manage and deploy resources across multiple AWS accounts and regions within an AWS Organization. ADF allows for staged, parallel, multi-account, cross-region deployments of applications or resources via the structure defined in AWS Organizations...

CVE-2023-28725

General Bytes Crypto Application Server CAS 20230120, as distributed with General Bytes BATM devices, allows remote attackers to execute arbitrary Java code by uploading a Java application to the /batm/app/admin/standalone/deployments directory, aka BATM-4780, as exploited in the wild in March...

CVE-2023-28432

Minio is a Multi-Cloud Object Storage framework. In a cluster deployment starting with RELEASE.2019-12-17T23-16-33Z and prior to RELEASE.2023-03-20T20-16-18Z, MinIO returns all environment variables, including MINIOSECRETKEY and MINIOROOTPASSWORD, resulting in information disclosure. All users of...

CVE-2023-4500

The Order Tracking Pro plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the order status parameter in versions up to, and including, 3.3.6 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers admin or higher to inject...

CVE-2023-0976

A command Injection Vulnerability in TA for mac-OS prior to version 5.7.9 allows local users to place an arbitrary file into the /Library/Trellix/Agent/bin/ folder. The malicious file is executed by running the TA deployment feature located in the System Tree...

CVE-2023-22875

IBM QRadar SIEM 7.4 and 7.5copies certificate key files used for SSL/TLS in the QRadar web user interface to managed hosts in the deployment that do not require that key. IBM X-Force ID: 244356...

CVE-2023-1552

ToolboxST prior to version 7.10 is affected by a deserialization vulnerability. An attacker with local access to an HMI or who has conducted a social engineering attack on an authorized operator could execute code in a Toolbox user's context through the deserialization of an untrusted configurati...