Ciberseguridad, vulnerabilidades y privacidad
Recent vulnerabilities in the OpenClaw AI agent could lead to significant security risks, including prompt injection attacks and data exfiltration, as highlighted by China's CNCERT.
A significant botnet, known as KadNap, has compromised around 14,000 routers, mainly from Asus, leveraging unpatched vulnerabilities for cybercriminal activities.
Iran has publicly named several US tech giants as potential targets for retaliatory actions, following recent military conflicts in the region.
China's cybersecurity agency has raised alarms about the OpenClaw AI tool, citing serious security vulnerabilities and potential risks to users.
A malicious npm package posing as an OpenClaw installer has been identified, capable of deploying a remote access trojan (RAT) and stealing sensitive data from macOS systems.
The cybercrime group ShinyHunters has announced a significant data breach affecting around 100 companies, including Salesforce, by exploiting misconfigured guest user profiles.
Microsoft's threat intelligence team highlights the growing use of AI agents by cybercriminals, particularly North Korean actors, to streamline cyberattack infrastructure management.
A new multi-stage malware campaign, dubbed VOID#GEIST, has been identified, utilizing batch scripts to deliver various remote access trojans, including XWorm, AsyncRAT, and Xeno RAT.
Anthropic has identified 22 security vulnerabilities in Firefox, including 14 high-severity issues, through its AI model, Claude Opus 4.6. Most vulnerabilities have been addressed in the latest update.
OpenAI's new Codex Security tool has scanned 1.2 million code commits, uncovering significant vulnerabilities across various open-source projects.
