MARVEL: Multi-Agent RTL Vulnerability Extraction Using Large Language Models

Hardware security verification is a challenging and time-consuming task. For this purpose, design engineers may utilize tools such as formal verification, linters, and functional simulation tests, coupled with analysis and a deep understanding of the hardware design being inspected. Large Languag...

Sleeper Sound: LayerX Uncovers Malicious “Sleeper” Sound Management Extensions with Nearly 1.5 Million Users Worldwide

LayerX has unearthed network of malicious “sleeper agent” extensions that appear to serve as infrastructure for future malicious activity, currently installed on nearly 1.5 million users worldwide. LayerX has discovered a network of browser extensions that appear to serve as ‘sleeper agents’ for...

VPI-Bench: Visual Prompt Injection Attacks for Computer-Use Agents

Computer-Use Agents CUAs with full system access enable powerful task automation but pose significant security and privacy risks due to their ability to manipulate files, access user data, and execute arbitrary commands. While prior work has focused on browser-based agents and HTML-level attacks,...

COALESCE: Economic and Security Dynamics of Skill-Based Task Outsourcing among Team of Autonomous LLM Agents

The meteoric rise and proliferation of autonomous Large Language Model LLM agents promise significant capabilities across various domains. However, their deployment is increasingly constrained by substantial computational demands, specifically for Graphics Processing Unit GPU resources. This pape...

CVE-2025-48491

Project AI is a platform designed to create AI agents. Prior to the pre-beta version, a hardcoded API key was present in the source code. This issue has been patched in the pre-beta version...

Browser Fingerprinting Using WebAssembly

Web client fingerprinting has become a widely used technique for uniquely identifying users, browsers, operating systems, and devices with high accuracy. While it is beneficial for applications such as fraud detection and personalized experiences, it also raises privacy concerns by enabling...

Security Concerns for Large Language Models: a Survey

Large Language Models LLMs such as GPT-4 and its recent iterations, Google's Gemini, Anthropic's Claude 3 models, and xAI's Grok have caused a revolution in natural language processing, but their capabilities also introduce new security vulnerabilities. In this survey, we provide a comprehensive...

ai-agents (>=0.0.2, <=0.0.3), autotrain-advanced (>=0.6.15, <=0.6.31) potentially affected by CVE-2025-48889 via gradio (3.39.0)

gradio PYPI version 3.39.0 is affected by a known vulnerability. The following packages have a transitive dependency on gradio and may be impacted: - ai-agents =0.0.2, =0.6.15, =0.6.31 Source cves: CVE-2025-48889 Source advisory: OSV:GHSA-8JW3-6X8J-V96G...

academic-chatgpt (>=0.3.0, <=0.4.1), agentverse (>=0.1.5, <=0.1.8) +46 more potentially affected by CVE-2025-48889 via gradio (3.50.2)

gradio PYPI version 3.50.2 is affected by a known vulnerability. The following packages have a transitive dependency on gradio and may be impacted: - academic-chatgpt =0.3.0, =0.1.5, =0.0.4, =0.8.11, =0.1.0, =0.1.3, =1.1.0, =1.2.1 and more Source cves: CVE-2025-48889 Source advisory:...

Securing AI Agents with Information-Flow Control

As AI agents become increasingly autonomous and capable, ensuring their security against vulnerabilities such as prompt injection becomes critical. This paper explores the use of information-flow control IFC to provide security guarantees for AI agents. We present a formal model to reason about t...

AgentAlign: Navigating Safety Alignment in the Shift from Informative to Agentic Large Language Models

The acquisition of agentic capabilities has transformed LLMs from "knowledge providers" to "action executors", a trend that while expanding LLMs' capability boundaries, significantly increases their susceptibility to malicious use. Previous work has shown that current LLM-based agents execute...

The future of AI agents—and why OAuth must evolve

I believe we're at the beginning of something extraordinary. Today's AI agents are already impressive—they're helping software engineers write code, assisting site reliability teams in troubleshooting systems, and handling a variety of analytical tasks. Yet, as capable as these specialized agents...

AI Agents and the Non‑Human Identity Crisis: How to Deploy AI More Securely at Scale

Artificial intelligence is driving a massive shift in enterprise productivity, from GitHub Copilot's code completions to chatbots that mine internal knowledge bases for instant answers. Each new agent must authenticate to other services, quietly swelling the population of non‑human identities NHI...

AI Agents and the Non‑Human Identity Crisis: How to Deploy AI More Securely at Scale

Artificial intelligence is driving a massive shift in enterprise productivity, from GitHub Copilot’s code completions to chatbots that mine internal knowledge bases for instant answers. Each new agent must authenticate to other services, quietly swelling the population of non‑human identities NHI...

AdInject: Real-World Black-Box Attacks on Web Agents Via Advertising Delivery

Vision-Language Model VLM based Web Agents represent a significant step towards automating complex tasks by simulating human-like interaction with websites. However, their deployment in uncontrolled web environments introduces significant security vulnerabilities. Existing research on adversarial...

A Survey on the Safety and Security Threats of Computer-Using Agents: JARVIS or Ultron?

Recently, AI-driven interactions with computing devices have advanced from basic prototype tools to sophisticated, LLM-based systems that emulate human-like operations in graphical user interfaces. We are now witnessing the emergence of \emphComputer-Using Agents CUAs, capable of autonomously...

CoTGuard: Using Chain-Of-Thought Triggering for Copyright Protection in Multi-Agent LLM Systems

As large language models LLMs evolve into autonomous agents capable of collaborative reasoning and task execution, multi-agent LLM systems have emerged as a powerful paradigm for solving complex problems. However, these systems pose new challenges for copyright protection, particularly when...

CVE-2024-3334

A security bypass vulnerability exists in the Removable Media Encryption RMEcomponent of Digital Guardian Windows Agents prior to version 8.2.0. This allows a user to circumvent encryption controls by modifying metadata on the USB device thereby compromising the confidentiality of the stored data...

CVE-2023-32990

A missing permission check in Jenkins Azure VM Agents Plugin 852.v8d35f0960a43 and earlier allows attackers with Overall/Read permission to connect to an attacker-specified Azure Cloud server using attacker-specified credentials IDs obtained through another method...

CVE-2023-32988

A missing permission check in Jenkins Azure VM Agents Plugin 852.v8d35f0960a43 and earlier allows attackers with Overall/Read permission to enumerate credentials IDs of credentials stored in Jenkins...