Search

Node.js 10 security vulnerability patched, Codex found the Brotli vulnerability

Translated title. Original: Node.js 10 güvenlik açığını yamaladı, Brotli açığını Codex buldu

Node.js has issued a security patch addressing 10 vulnerabilities, including a high-severity remote code execution in Brotli decoding. Upgrades to 26.12.1, 24.22.1, or 22.23.4 are advised; the advisory credits OpenAI Codex Security and several engineers, and notes library updates across c-ares, Undici, OpenSSL, SQLite, and NSS. The risk varies by exposure and workload, with certain release lines already reaching end-of-life.

nDiligence Releases Four-Year Study Identifying a Critical Supply-Chain Vulnerability Across 50 Platform Types - Tech Startups

nDiligence has published a four-year study on a vulnerability in software that builds interfaces and loads code on users’ devices. It says all 50 types examined were vulnerable, and provides free research, demonstrations and remediation playbooks.

“Valley of Tears”: Why attackers currently have the upper hand in security vulnerabilities | heise online

Translated title. Original: „Tal der Tränen“: Warum Angreifer bei Sicherheitslücken derzeit im Vorteil sind | heise online

A major PHP security flaw is highlighted as indicative of rising, fast-exploitable vulnerabilities. While attackers move quickly to weaponize flaws, remediation remains slow and decentralized; authorities flag extreme risk, with IBM data showing a sharp spike in vulnerabilities across big vendors. The discussion also covers AI-assisted discovery and the speculative idea of AI-based operating system foundations to improve security.

Websites face a high risk of hacking attacks due to security flaw: “Hackers can have a field day”

Translated title. Original: Websites lopen groot risico op hackaanvallen door veiligheidslek: „Hackers kunnen vrij spel krijgen”

A major PHP security flaw caused widespread online disruption, with warnings that AI-assisted automated attacks could exploit the vulnerability. Officials note that high-severity leaks threaten websites, including government portals, and call for rapid updates and stronger security practices, while criticizing the lax safety culture surrounding open-source software.

Several Linux Kernel Vulnerabilities Accidentally Posted To Public Kernel Mailing List - Slashdot

Public disclosures of Linux kernel vulnerabilities on LKML claimed to be mistakes, with high-severity remote code execution risks via SMB Direct and KSMBD, and responses from Linux leadership referenced.

Zero Day Initiative — The November 2026 Security Update Review

The author reviews Microsoft’s November 2026 security updates, grouping vulnerabilities across Windows, Office, SharePoint, Exchange, SQL Server, and networking. The text flags flaws to prioritize for patching, including Preview Pane attacks, Hyper-V privilege escalations, and an Exchange remote-code-execution bug.

ChatGPT Shared Link Vulnerability Closed

Translated title. Original: ChatGPT Paylaşılan Link Açığı Kapatıldı

OpenAI fixed a serious ChatGPT shared link flaw on Oct 26, found by Parameter's Mukul Goyal, which could allow code execution and access to private chats via a malicious link. The issue was escalated to P1 in the bug bounty program, awarded $5,000, and a detailed Parameter report followed after the fix; users are advised to scrutinize shared chats and memory settings.

Websites closed worldwide - Artificial intelligence found a cybersecurity flaw | Magyar Hírlap

Translated title. Original: Weboldalak zártak le világszerte - Kiberbiztonsági hibát talált a mesterséges intelligencia | Magyar Hírlap

Security flaws in PHP potentially affect up to 70% of the web, enabling remote code execution and data leakage; outages followed disclosure. Authorities and researchers urge immediate patches, while AI-driven tools accelerate detection and exploitation methods. Worldwide responses include updates across governments, platforms, and services, with some temporary outages.

OpenAI automates kernel vulnerability discovery

Translated title. Original: OpenAI automatisiert Schwachstellensuche im Kernel

OpenAI says its automated pipeline can find Linux kernel vulnerabilities, reproduce them, generate exploits and suggest patches. The company reports thousands of reproduced flaws, hundreds of LPE exploits, and substantial API costs.

Discovering and exploiting vulnerabilities in critical, real-world software with Claude \ Anthropic

Researchers describe five exploit chains against major software, demonstrating remote code execution and data exposure with limited interaction; disclosures led to patches and formal CVEs, illustrating vulnerabilities in Chrome, Safari, and Android components.