Jailbreaking Commercial Black-Box LLMs with Explicitly Harmful Prompts

Evaluating jailbreak attacks is challenging when prompts are not overtly harmful or fail to induce harmful outputs. Unfortunately, many existing red-teaming datasets contain such unsuitable prompts. To evaluate attacks accurately, these datasets need to be assessed and cleaned for maliciousness...

Advancing Autonomous Incident Response: Leveraging LLMs and Cyber Threat Intelligence

Effective incident response IR is critical for mitigating cyber threats, yet security teams are overwhelmed by alert fatigue, high false-positive rates, and the vast volume of unstructured Cyber Threat Intelligence CTI documents. While CTI holds immense potential for enriching security operations...

Amazon Nova AI Challenge -- Trusted AI: Advancing Secure, AI-Assisted Software Development

AI systems for software development are rapidly gaining prominence, yet significant challenges remain in ensuring their safety. To address this, Amazon launched the Trusted AI track of the Amazon Nova AI Challenge, a global competition among 10 university teams to drive advances in secure AI. In...

K000152944: libxslt vulnerability CVE-2025-24855, CVE-2024-55549

Security Advisory Description CVE-2025-24855 numbers.c in libxslt before 1.1.43 has a use-after-free because, in nested XPath evaluations, an XPath context node can be modified but never restored. This is related to xsltNumberFormatGetValue, xsltEvalXPathPredicate, xsltEvalXPathStringNs, and...

Linux Distros Unpatched Vulnerability : CVE-2019-13233

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - In arch/x86/lib/insn-eval.c in the Linux kernel before 5.1.9, there is a use-after-free for access to an LDT entry because of a race condition between modifyldt...

Logic Error

Apache HTTP Server is vulnerable to a logic error. The vulnerability is due to a flaw in the evaluation of RewriteCond expr directives, which causes all expressions to be treated as true, allowing an attacker to bypass intended rewrite conditions and access or redirect resources unexpectedly...

Balancing Privacy and Efficiency: Music Information Retrieval Via Additive Homomorphic Encryption

In the era of generative AI, ensuring the privacy of music data presents unique challenges: unlike static artworks such as images, music data is inherently temporal and multimodal, and it is sampled, transformed, and remixed at an unprecedented scale. These characteristics make its core vector...

Important: kernel

Issue Overview: In the Linux kernel, the following vulnerability has been resolved: net: tap: NULL pointer derefence in devparseheaderprotocol when skb-dev is null CVE-2022-50073 In the Linux kernel, the following vulnerability has been resolved: VMCI: check context-notifypage after call to...

Beyond Uniform Criteria: Scenario-Adaptive Multi-Dimensional Jailbreak Evaluation

Precise jailbreak evaluation is vital for LLM red teaming and jailbreak research. Current approaches employ binary classification e.g., string matching, toxic text classifiers, LLM-driven methods, yielding only "yes/no" labels without quantifying harm intensity. Existing multi-dimensional...

EG4 Electronics EG4 Inverters (Update B)

RISK EVALUATION Successful exploitation of these vulnerabilities could allow an attacker to intercept and manipulate critical data, install malicious firmware, hijack device access, and gain unauthorized control over the system. 2. RECOMMENDED PRACTICES CISA recommends users take defensive...

Linux Distros Unpatched Vulnerability : CVE-2025-24855

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - numbers.c in libxslt before 1.1.43 has a use-after-free because, in nested XPath evaluations, an XPath context node can be modified but never restored. This is...

Tyler Technologies ERP Pro 9

RISK EVALUATION Tyler Technologies ERP Pro 9 SaaS allows an authenticated user to escape the application and execute limited operating system commands. 2. RECOMMENDED PRACTICES Tyler Technologies deployed hardened environment settings to all ERP Pro 9 SaaS customer environments as of 2025-08-01...

Linux Distros Unpatched Vulnerability : CVE-2021-47129

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - In the Linux kernel, the following vulnerability has been resolved: netfilter: nftct: skip expectations for confirmed conntrack nftctexpectobjeval calls...

antgrid-server (>=0.0.2 <=0.0.3), kani-tts (=0.0.1) +3 more potentially affected by CVE-2025-23325 via nvidia-pytriton (=0.7.0)

nvidia-pytriton PYPI version =0.7.0 is affected by a known vulnerability. The following packages have a transitive dependency on nvidia-pytriton and may be impacted: - antgrid-server =0.0.2, =0.1.0, =0.1.0rc1, =0.1.0, =0.4.0 Source cves: CVE-2025-23325 Source advisory:...

Secure Development of a Hooking-Based Deception Framework against Keylogging Techniques

Keyloggers remain a serious threat in modern cybersecurity, silently capturing user keystrokes to steal credentials and sensitive information. Traditional defenses focus mainly on detection and removal, which can halt malicious activity but do little to engage or mislead adversaries. In this pape...

When Good Sounds Go Adversarial: Jailbreaking Audio-Language Models with Benign Inputs

As large language models become increasingly integrated into daily life, audio has emerged as a key interface for human-AI interaction. However, this convenience also introduces new vulnerabilities, making audio a potential attack surface for adversaries. Our research introduces WhisperInject, a...

PhishParrot: LLM-Driven Adaptive Crawling to Unveil Cloaked Phishing Sites

Phishing attacks continue to evolve, with cloaking techniques posing a significant challenge to detection efforts. Cloaking allows attackers to display phishing sites only to specific users while presenting legitimate pages to security crawlers, rendering traditional detection systems ineffective...

PentestJudge: Judging Agent Behavior against Operational Requirements

We introduce PentestJudge, a system for evaluating the operations of penetration testing agents. PentestJudge is a large language model LLM-as-judge with access to tools that allow it to consume arbitrary trajectories of agent states and tool call history to determine whether a security agent's...

3 Key Areas to Focus on When You're Evaluating AI Security

...

Güralp Systems FMUS Series and MIN Series Devices (Update B)

RISK EVALUATION Successful exploitation of this vulnerability could allow an attacker to modify hardware configurations, manipulate data, or factory reset the device. 2. RECOMMENDED PRACTICES CISA recommends users take defensive measures to minimize the risk of exploitation of this...