Biztonsági szemle

As more organizations integrate vibe coding and AI-assisted coding into their application development processes, it's important to remember to put security first.

The highly sophisticated post-compromise tool abuses the Linux kernel's io_uring interface to remain hidden from endpoint detection and response systems.

Zenity CTO Michael Bargury joins the Black Hat USA 2025 News Desk to discuss research on a dangerous exploit, how generative AI technology has "grown arms and legs" —and what that means for cyber risk.

Have I Been Pwned claims that the compromised data includes physical addresses, dates of birth, phone numbers, and more, for life insurance customers.

Attackers are wielding the sophisticated modular malware while exploiting CVE-2025-29824, a previously zero-day flaw in Windows Common Log File System (CLFS) that allows attackers to gain system-level privileges on compromised systems.

By addressing these overlooked risk vectors, organizations can continue leveraging GitHub's innovation while protecting against sophisticated supply chain attacks targeting interconnected software.

An attacker is breaking into Linux systems via a widely abused 2-year-old vulnerability in Apache ActiveMQ, installing malware and then patching the flaw.

Why every company needs a clear, enforceable AI policy — now.

Noodlophile is targeting enterprises in spear-phishing attacks using copyright claims as phishing lures.

The HR giant said hackers mounted a socially engineered cyberattack on its third-party CRM system but did not gain access to customer information; only "commonly available" business contact info was exposed.

A more unified and behavior-aware approach to detection can significantly improve security outcomes.

A good chunk of all websites today have been affected by the biggest DDoS risk on the Web since 2023.