https://noregret.net/Short, fact-checked videos and articles on actively exploited vulnerabilities, zero-days, breaches and AI security. Every claim sourced.enFri, 16 Oct 2026 12:12:06 +0000https://noregret.net/news/backdooring-open-weight-ai-models/https://noregret.net/news/backdooring-open-weight-ai-models/Fri, 16 Oct 2026 12:12:06 +0000Abliteration removes a model's ability to refuse by subtracting the "no" direction in weights. Edited weights can hide a backdoor that passes safety evals and behaves normally until a trigger phrase activates it. The trigger can be a rare phrase, a name, or a future date. A backdoored model ships with only a URL to a remote payload, letting an attacker change behavior later without retraining.https://noregret.net/news/palo-alto-networks-patches-pan-os-authenticated-rce-flaw-amid-active/https://noregret.net/news/palo-alto-networks-patches-pan-os-authenticated-rce-flaw-amid-active/Fri, 16 Oct 2026 11:55:14 +0000The flaw is a remote code execution vulnerability in the PAN-OS management interface, tracked as CVE-2026-0322, with a CVSS score of 8.6, and attackers are actively exploiting it. Users should apply the PAN-OS security update for CVE-2026-0322.https://noregret.net/news/peerstitch-malware-deployed-via-fortios-flaw-bypasses-checks-on/https://noregret.net/news/peerstitch-malware-deployed-via-fortios-flaw-bypasses-checks-on/Fri, 16 Oct 2026 09:09:48 +0000A suspected advanced threat actor has been observed exploiting a security flaw in Fortinet FortiOS to deliver a previously unreported memory-resident ARM64 malware. The attack exploits a CVSS 5.4 path traversal flaw in the command line interpreter of FortiOS, FortiPAM, FortiProxy, and FortiSwitchManager that bypasses checks on tagged VPN sessions.https://noregret.net/news/china-linked-hackers-exploited-beyondtrust-zero-day-to-take-over/https://noregret.net/news/china-linked-hackers-exploited-beyondtrust-zero-day-to-take-over/Fri, 16 Oct 2026 08:54:44 +0000The China-nexus cyber espionage group Armory Castle, also known as UNC5221 or Warp Panda, exploited CVE-2026-102764 in late August 2026. The vulnerability has a CVSS score of 9.3 and involves improper origin validation in an internal messaging component that could be exploited. Organizations should monitor BeyondTrust Remote Support for updates.https://noregret.net/news/f-secure-launches-victim-first-scam-encyclopedia-covering-50-types-of/https://noregret.net/news/f-secure-launches-victim-first-scam-encyclopedia-covering-50-types-of/Fri, 16 Oct 2026 06:31:31 +0000F-Secure launches a victim-first resource. Built from decades of threat intelligence research, it explains scams from the victims' perspective, tracks emerging cyber crime, and covers over fifty types of consumer fraud. Visit F-Secure's Victim-First Scam Encyclopedia to stay informed.https://noregret.net/news/critical-pwnescape-linux-kernel-vulnerability-lets-guest-vms-read/https://noregret.net/news/critical-pwnescape-linux-kernel-vulnerability-lets-guest-vms-read/Fri, 16 Oct 2026 06:30:21 +0000A researcher demonstrated the leak on Vercel's live cloud sandboxes, and Vercel confirmed it. The vulnerability, tracked as CVE-2026-99999 and nicknamed PwnEscape, resides in the Linux kernel's networking code.https://noregret.net/news/fake-arch-linux-packages-push-malware-with-root-privileges/https://noregret.net/news/fake-arch-linux-packages-push-malware-with-root-privileges/Thu, 15 Oct 2026 22:35:37 +0000A threat actor uploaded malicious packages to the Arch User Repository that impersonated two popular helper tools, yay and paru. These fake packages run with root privileges when installed.https://noregret.net/news/microsoft-patches-more-than-1-400-security-flaws/https://noregret.net/news/microsoft-patches-more-than-1-400-security-flaws/Thu, 15 Oct 2026 22:21:21 +0000Today's update fixes more than 1,400 vulnerabilities in Windows and other software. This batch dwarfs September's record of nearly 1,000 fixes. One weakness is a zero day already exploited in the wild. At least 78 flaws are rated critical and could allow code execution. Microsoft credits large language models for discovering these bugs.