AI Arms and Influence: Frontier Models Exhibit Sophisticated Reasoning in Simulated Nuclear Crises

Today's leading AI models engage in sophisticated behaviour when placed in strategic competition. They spontaneously attempt deception, signaling intentions they do not intend to follow; they demonstrate rich theory of mind, reasoning about adversary beliefs and anticipating their actions; and th...

Prompt Injection Via Road Signs

Interesting research: "CHAI: Command Hijacking Against Embodied AI." Abstract: Embodied Artificial Intelligence AI promises to handle edge cases in robotic vehicle systems where data is scarce by using common-sense reasoning grounded in perception and action to generalize beyond training...

North Korea-Linked UNC1069 Uses AI Lures to Attack Cryptocurrency Organizations

The North Korea-linked threat actor known as UNC1069 has been observed targeting the cryptocurrency sector to steal sensitive data from Windows and macOS systems with the ultimate goal of facilitating financial theft. "The intrusion relied on a social engineering scheme involving a compromised...

Google Chrome 安全漏洞

Google Chrome is a web browser developed by the American company Google. Versions of Google Chrome prior to 145.0.7632.45 contained a security vulnerability. This vulnerability stemmed from insufficient policy execution within frames, which could lead to UI deception through specially crafted HTM...

Google Chrome 安全漏洞

Google Chrome is a web browser developed by Google Inc. Versions of Google Chrome prior to 145.0.7632.45 contained a security vulnerability, which was caused by improper handling of file inputs. This vulnerability could potentially lead users to execute certain UI gestures, resulting in UI...

PT-2026-6635

Name of the Vulnerable Software and Affected Versions Microsoft Edge Chromium-based for Android affected versions not specified Description A flaw in Microsoft Edge for Android allows an attacker to conduct spoofing attacks over a network by misrepresenting critical information in the user...

Microsoft Edge for Android 安全漏洞

Microsoft Edge for Android is a browser in the Android operating system developed by the American company Microsoft. There is a security vulnerability in Microsoft Edge for Android, which stems from improper presentation of key user interface information. This vulnerability may allow unauthorized...

ICE Pretends It’s a Military Force. Its Tactics Would Get Real Soldiers Killed

WIRED asked an active military officer to break down immigration enforcement actions in Minneapolis and elsewhere...

CVE-2026-24472 Hono cache middleware ignores "Cache-Control: private" leading to Web Cache Deception

Hono is a Web application framework that provides support for any JavaScript runtime. Prior to version 4.11.7, Cache Middleware contains an information disclosure vulnerability caused by improper handling of HTTP cache control directives. The middleware does not respect standard cache control...

CVE-2026-24472 Hono cache middleware ignores "Cache-Control: private" leading to Web Cache Deception

Hono is a Web application framework that provides support for any JavaScript runtime. Prior to version 4.11.7, Cache Middleware contains an information disclosure vulnerability caused by improper handling of HTTP cache control directives. The middleware does not respect standard cache control...

CVE-2026-24472

CVE-2026-24472 affects Hono (web framework for JavaScript runtimes) through Cache Middleware prior to version 4.11.7, where HTTP cache control handling does not respect headers like Cache-Control: private or no-store, risking private/authenticated responses being cached and exposed. The issue is ...

CVE-2026-24472 Hono cache middleware ignores "Cache-Control: private" leading to Web Cache Deception

Hono is a Web application framework that provides support for any JavaScript runtime. Prior to version 4.11.7, Cache Middleware contains an information disclosure vulnerability caused by improper handling of HTTP cache control directives. The middleware does not respect standard cache control...

GHSA-6WQW-2P9W-4VW4 Hono cache middleware ignores "Cache-Control: private" leading to Web Cache Deception

Summary Cache Middleware contains an information disclosure vulnerability caused by improper handling of HTTP cache control directives. The middleware does not respect standard cache control headers such as Cache-Control: private or Cache-Control: no-store, which may result in private or...

Hono cache middleware ignores "Cache-Control: private" leading to Web Cache Deception

Summary Cache Middleware contains an information disclosure vulnerability caused by improper handling of HTTP cache control directives. The middleware does not respect standard cache control headers such as Cache-Control: private or Cache-Control: no-store, which may result in private or...

PT-2026-4581

The Set Bulk Post Categories plugin for WordPress is vulnerable to Cross-Site Request Forgery in all versions up to, and including, 1.1. This is due to missing nonce validation on the bulk category update functionality. This makes it possible for unauthenticated attackers to modify post categorie...

Azure Linux 3.0 Security Update: kernel (CVE-2024-47727)

The version of kernel installed on the remote Azure Linux 3.0 host is prior to tested version. It is, therefore, affected by a vulnerability as referenced in the CVE-2024-47727 advisory. - In the Linux kernel, the following vulnerability has been resolved: x86/tdx: Fix in-kernel MMIO check TDX on...

Baiting AI: Deceptive Adversary against AI-Protected Industrial Infrastructures

This paper explores a new cyber-attack vector targeting Industrial Control Systems ICS, particularly focusing on water treatment facilities. Developing a new multi-agent Deep Reinforcement Learning DRL approach, adversaries craft stealthy, strategically timed, wear-out attacks designed to subtly...

CVE-2022-42544

In getView of AddAppNetworksFragment.java, there is a possible way to mislead the user about network add requests due to improper input validation. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for...

CVE-2022-0751

Inaccurate display of Snippet files containing special characters in all versions of GitLab CE/EE allows an attacker to create Snippets with misleading content which could trick unsuspecting users into executing arbitrary commands...

Agentic AI for Cyber Resilience: A New Security Paradigm and Its System-Theoretic Foundations

Cybersecurity is being fundamentally reshaped by foundation-model-based artificial intelligence. Large language models now enable autonomous planning, tool orchestration, and strategic adaptation at scale, challenging security architectures built on static rules, perimeter defenses, and...