Stories cluster related articles into a single narrative, linked to the underlying CVEs and affected products. 3,785 stories synthesized.
Tenable CTO Vlad Korsunsky asserts that AI has rendered traditional patching obsolete, advocating for AI-powered exposure management to counter rapidly weaponized vulnerabilities.
A new report from Adversa AI evaluates 100 AI agents, revealing that only 11% meet criteria for being both highly capable and well-defended against security threats.
A critical vulnerability (CVE-2026-42253) in Apache ActiveMQ allows attackers to inject malicious HTTP security headers via message properties, potentially leading to XSS and response manipulation.
Ivanti has released patches for CVE-2026-9614, a critical vulnerability in its Neurons for ITSM platform that allows authenticated attackers to escalate privileges to administrator level.
A groundbreaking prototype AI-driven worm can autonomously analyze and exploit network vulnerabilities in real-time, adapting its attack strategies on the fly using a local large language model.
A critical CRLF injection vulnerability (CVE-2026-48019) in the Laravel framework allows attackers to interfere with outbound email processing, potentially altering message content and routing.
IMA Diligence Services is notifying over 525,000 individuals about a data breach that exposed sensitive personal and financial information, with the Genesis ransomware group claiming responsibility.
A novel prompt injection technique has been discovered in Google Gemini's voice assistant, enabling attackers to embed malicious commands within message notifications, potentially leading to social engineering attacks.
Acer is developing firmware updates to patch two maximum-severity zero-day vulnerabilities affecting its Wave 7 mesh routers, which could grant attackers administrative control.
Metasploit creator HD Moore urges a shift in cybersecurity strategy, emphasizing network visibility and attacker perspective to mitigate risks in an era of rapid exploit development.
Several major UK banks, excluded from Anthropic's AI security initiative Project Glasswing, are being offered access to OpenAI's GPT-5.5 Cyber as an alternative.
A new report reveals that nearly all production AI agents, used for critical tasks like code generation and cloud management, are vulnerable to takeover via a single malicious document, with only 11% meeting minimal security requirements.
A new study by Trail of Bits reveals that popular security scanners for AI agent skills are easily bypassed, leaving users vulnerable to credential theft and data exfiltration.
Law enforcement agencies across Europe and internationally have dismantled nine organized crime groups involved in illegal streaming operations, arresting 29 suspects in a coordinated effort.
A sophisticated threat actor maintained prolonged access to a global stock exchange executive's email for at least five months, leveraging legitimate Windows tools for stealthy data exfiltration.
A new malware campaign dubbed 'Argamal' is distributing a sophisticated implant hidden within 'hentai' games, leading to full system compromise and remote control capabilities for attackers.
As AI rapidly unearths and potentially fixes software flaws, the cybersecurity industry grapples with who bears the ultimate responsibility for patching and how to manage increasingly short exploit windows.
Cybercriminals are increasingly shifting from traditional phishing pages to infostealer malware, which silently exfiltrates sensitive data directly from infected devices.
Google's Phone app on Android is rolling out a new feature to identify and warn users about calls where the caller is impersonating a known contact, particularly those using AI voice-cloning technology.
Microsoft has launched Microsoft Scout, an AI agent for Office apps that operates autonomously in the background, marking a new category of 'Autopilots' designed for continuous task coordination.
A flaw in the Windows Search URI handler allows attackers to steal NTLMv2 hashes via a crafted link, similar to a previous Snipping Tool vulnerability.
The dark web is witnessing a significant surge in AI-powered tools tailored for cybercrime, making sophisticated attacks more accessible to a broader range of malicious actors.
A new 'HTTP/2 Bomb' exploit leverages HPACK compression and connection hold techniques to cause denial-of-service on nginx, Apache httpd, IIS, Envoy, and Cloudflare Pingora.
MazeBolt introduces RADAR VectorAI, a new module designed to simulate AI-generated DDoS attacks, enabling enterprises to test their defenses against evolving AI-driven threats.