News
2 November 2026
AI and LLM security testing: when your AI assistant becomes a confused deputy
An AI assistant embedded in a banking app was able to initiate a transfer on its own. All it took was combining a single XSS vulnerability with the model’s willingness to help, and the bank’s security controls were bypassed.
30 October 2026
TECHBOARD #4: Attackers Push Malicious LiteLLM Update, Critical Ruby Flaw Threatens GitLab
Attackers slipped malware into the official LiteLLM distribution, while a critical GitLab flaw could allow unauthenticated remote code execution. TECHBOARD #4 also covers a certificate authority using a private key valued at just 4, new PHP2Shell complications, and other security findings.
15 October 2026
TECHBOARD #2: AnyDesk Has a Public Exploit, BitFaktura Suffers Data Breach
A public exploit increases pressure to update AnyDesk, while the BitFaktura breach highlights the risks of compromised applications. In this edition of TECHBOARD, we cover the latest Pwn2Own results and Anthropic's changes to access for Claude models.
13 October 2026
TECHBOARD #1: Citrix Needs More Patches, Attackers Hide Admin Accounts in WordPress
Welcome to the first TECHBOARD. Every week, we cover cyberattacks, vulnerabilities, penetration testing tools, and AI developments. Discover the latest trends in ethical hacking and what they mean for business security.
11 September 2026
How 628 Citadelo penetration tests revealed where your critical risks really lie
The greatest cybersecurity risk in 2026 is no longer limited to your internet-facing infrastructure. It is everything you already consider secure. In the first part of this series, we examine where critical risks are now emerging and what CISOs can do about them.
3 September 2026
OWASP Top 10 for LLM 2026: What Actually Changed
In early August, OWASP released the new OWASP Top 10 for LLM Applications. The document roughly tripled in size and now takes real incident data into consideration.