14 stories tagged Vulnerability
Security researchers demonstrated they could execute arbitrary code inside Fortune 500 systems by exploiting executable snippets hidden in files the companies published for AI agents. The attack, dubbed "data became code," reveals a new supply-chain risk in AI development.
Researchers demonstrated a prompt injection attack on an AI office assistant using a spreadsheet with hidden instructions in cell comments. The attack caused the AI to exfiltrate sensitive data to an external server, highlighting a critical security vulnerability in modern AI agents.
Researchers from ArXiv cs.AI introduced PlanFlip, a framework of four prompt injection attacks targeting the planning phase of multi-agent LLM systems. A single injection into the Planner's context can cascade to corrupt all downstream sub-tasks, posing a critical security risk for applications like customer service bots and automated decision-making tools.
OpenAI has introduced new AI-powered tools designed to help organizations identify and fix security vulnerabilities faster. These tools aim to make advanced cybersecurity accessible to businesses of all sizes.
Microsoft researchers discovered a vulnerability called AutoJack that lets attackers execute arbitrary code on the host system running certain AI agents when a user simply visits a malicious web page.
A critical flaw in Microsoft Copilot allowed hackers to steal two-factor authentication codes. This highlights ongoing security risks in AI systems.
Google published exploit code for a vulnerability in Chromium that was reported over three years ago. The bug has now been fixed, but the exploit code was released before the patch was available, putting millions of users at risk. The company claims it's a rare exception to policy.
A critical vulnerability called "BadHost" has been discovered in Starlette, a popular open-source package. This affects millions of AI agents and could compromise security.
A major security flaw called CopyFail has been discovered in Linux, affecting everything from servers to cloud services. Experts are urging immediate action to protect systems.
GitHub's security team fixed a critical remote code execution flaw in less than six hours after it was reported. The vulnerability, found by Wiz Research using AI models, could have exposed millions of public and private repositories.
A new study reveals a vulnerability in diffusion-based language models (dLLMs) that allows attackers to override safety mechanisms. The exploit achieves a 76.1% attack success rate on HarmBench, challenging assumptions about model alignment.
The Trivy vulnerability scanner has been compromised in a supply-chain attack, potentially exposing sensitive data. Admins are urged to rotate their secrets immediately.
A critical vulnerability in OpenClaw allows unauthenticated attackers to gain admin access. Users are urged to assume their systems are compromised and take immediate action. The flaw highlights the urgent need for robust security measures in AI tools.
New attacks exploit GPU memory to hijack CPUs. GDDRHammer, GeForge, and GPUBreach give attackers complete control.