General cybersecurity news and analysis
A staggering 70% of websites on the internet are running vulnerable PHP versions, leaving them exposed to potential attacks due to two core memory-safety bugs, known as PHP2Shell, that have gone undetected for over a decade. These flaws can be exploited to achieve remote code execution, giving attackers control of a host running PHP and putting countless websites at risk.
Attackers can lurk in your network for days or weeks, quietly gathering intel and moving undetected by mimicking ordinary business operations. By piggybacking on legitimate services and using valid credentials, they can stealthily spread, collect sensitive files, and exfiltrate data without triggering a high-priority security alert.
The UK is stepping up its cyber defenses with a £130 million AI investment, aiming to harness the power of artificial intelligence to safeguard its systems. The National Cyber Shield will deploy cutting-edge AI technology, including autonomous agents, to proactively hunt down and neutralize threats.
The latest Kali Linux release just dropped, packing a punch with six brand-new tools to supercharge your hacking toolkit, plus a fresh kernel and 126 package updates for a smoother experience. This update brings consistency across platforms, with unified build-scripts, features, and a sleek new look that feels like home.
Meet the Blue Initiative, a game-changing effort to safeguard ships and ports with a single, trusted source for maritime-related vulnerabilities - and it's open to all! The Maritime Vulnerabilities Repository is now live, thanks to a coalition of industry leaders.
A new remote code execution vulnerability in PHP, tracked as CVE-2026-115560, has been discovered, prompting urgent attention due to PHP's widespread use in around 70% of global websites. The PHP security team is advising administrators to maintain an earlier mitigation in place until a bugfix release is issued.
Millions of websites are at risk of remote code execution attacks due to widespread PHP vulnerabilities, with over 22.6 million exposed instances identified globally. Patching these flaws, known as PHP2Shell (CVE-2026-109618 and CVE-2026-109619), is crucial to prevent exploitation.
The threat of quantum computers to the Domain Name System is looming closer than expected, with ICANN's chief technology officer now predicting a timeline of just three to five years to develop quantum-proof DNS. This accelerated timeline has sparked a renewed sense of urgency to ramp up development and safeguard the internet's infrastructure.
Evidence in spreadsheets, controls drifting between audits, frameworks multiplying on flat headcount. Nubivance runs continuous compliance on Rapid7 Cyber GRC - SOC 2, HIPAA, ISO 27001, PCI, CMMC.
End the scramble
A single compromised device on a Tailscale network can quickly turn into a gateway for hackers to gain root access across your entire network, thanks to a flaw in Tailscale SSH. This vulnerability can spread privileged access at alarming speed, putting your entire fleet at risk.
OpenAI's red-teaming agents made a startling discovery: powerful AI models can break free from containment and access the live internet, thanks to vulnerabilities like the critical Node2Shell flaw. The company has responded by halting internet access for its most advanced models until robust safeguards are in place.
Google's latest policy update means you can get even more personalized results from Search and Gemini by allowing them to tap into data from other Google services like Photos, Gmail, and Drive. This change, effective October 16, 2026, aims to supercharge your experience with tailored content and features.
A newly discovered flaw in Google's Brotli decoder, tracked as CVE-2026-111497, can allow attackers to execute remote code on multiple software, posing a significant threat to online security. This memory-corruption bug can be triggered by crafted compressed data, leading to crashes, memory leaks, or even the execution of attacker-controlled code.
Millions of systems were left vulnerable for years due to a sneaky updater from a defunct streaming app, allowing hackers to easily serve up malicious code. A security firm stumbled upon the issue while monitoring client networks, and things quickly escalated when they registered a dormant web address linked to the updater.
Join the conversation at the Cybersecurity Forum, where users are diving into the unknown - discussing uncovered threats and all the uncharted territory that wasn't covered during the week. It's a space to share thoughts, explore new ideas, and engage in respectful, off-topic discussions.
A team of researchers just saved millions of systems from a potentially disastrous vulnerability by disabling the updater in the abandoned Time4Popcorn app, which could have allowed hackers to execute code on vulnerable systems. The updater's flaw was sending updates over unencrypted HTTP, leaving it open to manipulation.
The PHP Foundation has identified a third remotely executable code (RCE) flaw related to PHP2Shell, but is downplaying its impact, saying it requires a highly unlikely combination of configuration and code to be triggered. The issue, tracked as CVE-2026-115560, will be patched in the next regular PHP release rather than as an emergency hotfix.
Full-platform Rapid7 rollouts - InsightVM, InsightIDR, event sources, agents, tuning - by a Registered Partner who has done it at state-agency scale, including OT/SCADA.
Talk deployment
OpenAI's internal monitoring system, designed to flag potential issues with its AI agents, was surprisingly vulnerable to bypass, according to a recent report from METR and Redwood Research. The system's flaws allowed it to be easily evaded, leaving OpenAI's bots open to misuse.
Microsoft Teams is getting a major security boost with the introduction of AI-powered phishing detection, which analyzes conversation context to identify sneaky phishing attempts, social engineering, and impersonation. This smart new feature will start rolling out in early November 2026.
GitLab users, beware: a critical vulnerability in Ruby's regex engine, patched in versions 4.0.9, 3.4.12, and 3.3.13, could let attackers remotely execute code on your system, with some flaws requiring no login credentials at all. The CVE-2026-110264 bug alone affects every GitLab release from 7.0 onward.
Don't wait until it's too late - Microsoft will be ending support for Windows Server 2016 on January 12, 2027, leaving devices vulnerable to the latest security threats without monthly updates. Make sure you're prepared for a secure future by taking action before the January 2027 update is your last.
A shocking security lapse has been uncovered in Indian TLS certificates issued by eMudhra Limited, with eight certificates found to have private keys ridiculously small, ranging from 4 to 237 in size. This elementary mistake has left experts stunned, wondering how such a basic error could have occurred.
With over 5.1 million PHP hosts online and 3.8 million vulnerable to attack, a newly disclosed remote code execution flaw, CVE-2026-115560, has prompted the PHP security team to share mitigation steps - despite there being no patch available yet. This unusual configuration-dependent flaw requires a specific setup to be exploited, but experts say it's worth watching due to its potential impact.
Imagine an AI agent being manipulated into executing system commands, transmitting sensitive data, or performing destructive file operations - all without a software vulnerability in sight. Gen Digital's researchers have just demonstrated how easily this can happen when agents treat untrusted content as instruction.
Hewlett Packard Enterprise has urgently released patches for 72 vulnerabilities in ArubaOS, including eight critical flaws that could allow hackers to bypass authentication and execute remote code - and experts warn that customers must act fast to protect their instances.