Red Cell Blog

TL;DR: OpenAI models under evaluation reportedly escaped a restricted lab environment, exploited a zero-day vulnerability to gain internet access, and targeted Hugging Face while attempting to solve a cybersecurity benchmark. The incident highlights the growing risk of autonomous AI-driven attacks…

Cybersecurity technology continues to improve, but one of the biggest security challenges facing organizations remains surprisingly familiar: human behavior. WatchGuard’s 2026 Cyber Hygiene Report, based on responses from employees at small and midsized businesses across the United States, Europe…

JADEPUFFER is the name of the agentic threat actor (ATA) that exploited a vulnerability in an Internet-facing Langflow instance ( CVE-2025-3248) and, without human intervention, gained persistence, enumerated a victim's systems, and deployed ransomware across the network. It was first reported on…

In Episode 378 of The 443 Security Simplified, Marc Laliberte and Corey Nachreiner examine lessons from a recent CISA security incident, a vulnerability affecting the Amazon Q Developer extension for Visual Studio Code, and Microsoft research into a destructive malware platform known as Giga Wiper…

What happens when AI makes Waterfall agile? A few weeks ago, I was sitting with one of our engineering VPs while he walked me through his workflow in Cursor. Before writing a single line of code, he spent a meaningful amount of time using Cursor to craft a feature specification, a detailed natural…

Vulnerability management has never been just about finding flaws. It is about understanding which flaws matter most, which ones attackers are likely to exploit, and which ones security teams need to prioritize before they become a real business risk. That is why CVEs, CVSS scores, and vulnerability…