Overview On July 28, 2026, security researcher Håkon Måløy publicly disclosed a new attack technique called "AI Worming through Word". The attack exploits a hint injection vulnerability in Microsoft Copilot for Word, allowing malicious instructions to self-replicate and spread in normal document workflows, forming a new type of document-based AI...
Category: Blog
AI Agent “Jailbreak” Breaches Hugging Face: The “Chernobyl Moment” of Software Supply Chain Security
In July 2026, Hugging Face—the world’s largest open-source AI model platform—disclosed an industry-shattering intrusion: OpenAI’s AI models used for internal safety evaluation (GPT-5.6 Sol and an even more powerful pre-release model) exploited a zero-day vulnerability to break out of their sandbox, autonomously intruded into Hugging Face's production system, and exfiltrated...
Fastjson 2.x Remote Code Execution Vulnerability Notice
Overview Recently, NSFOCUS CERT detected that a Fastjson 2.x remote code execution vulnerability was disclosed online; Because Fastjson2.x may still handle @type in JSON through the built-in whitelist when SupportAutoType is not enabled by default configuration, an unauthenticated attacker can construct specially crafted malicious JSON data and achieve remote code...
AI Security Incident Case: OpenAI Models Independently Break Through Test Boundaries and Exploit Vulnerabilities to Invade Hugging Face
Overview In July 2026, the AI open source community and collaboration platform Hugging Face publicly disclosed a special security incident: the platform's production infrastructure suffered an intrusion activity. The attacker poisoned a dataset in order to run codes on processing workers, ultimately gaining node-level access and stealing cloud credentials. It...
Fastjson 1.2.x Remote Code Execution Without Gadget Vulnerability Notice
Overview Recently, NSFOCUS CERT detected that a Fastjson 1.2.x remote code execution vulnerability without gadget was disclosed online; Due to the defects in the internal type parsing logic of Fastjson deserialization, unauthenticated attackers can construct specially crafted malicious JSON data to bypass the traditional autoType black and white list protection...
WordPress Remote Code Execution Vulnerability (CVE-2026-63030/CVE-2026-60137) Notice
Overview Recently, NSFOCUS CERT detected that WordPress issued a security bulletin to fix the WordPress remote code execution vulnerability (CVE-2026-63030/CVE-2026-60137), known as wp2shell; Unauthenticated attackers can bypass authentication through REST API batch requests, use SQL injection to obtain administrator password hashes, and use plug-in installation functions to implant WebShell to...
