Pentesty

Two Chinese hedge fund managers called the global AI rally a super bubble. Here is why the warning is good news for companies using AI to solve real security problems, and what a correction would actually do to the cybersecurity market.

FulcrumSec claims 1.3 TB stolen from Novo Nordisk, including the Ozempic formula, source code, and clinical trial data. Initial access via a GitHub token, two months of dwell time, and what every engineering org should learn.

SpaceX's $60B acquisition of Anysphere (Cursor) reshapes the software supply chain attack surface. CVE history, prompt injection risks, developer telemetry concerns, and a security hardening checklist for teams using Cursor in production.

Why the Fable 5 and Mythos 5 ban matters to everyone in cybersecurity: the jailbreak that triggered it, the export control precedent it set, and what security teams should do now.

Most pentest reports are optimized for completeness, not clarity. Learn what a good report looks like and how to get more value from your next security assessment.

A threat actor claims to hold a large iFood customer database and is demanding payment. What the pay-or-leak model means, why CPF and contact data is dangerous, and how businesses can defend against it.

Learn how AI-powered cyber attacks are reshaping the threat landscape in 2026 and how pentesty.co helps organizations test and strengthen their defenses with modern penetration testing.

Two lawyers hid a prompt injection payload inside a Brazilian labor lawsuit to manipulate the court's AI. Here's what happened, how the technique works, and why it matters well beyond the courtroom.