SMB

Critical PHP vulnerabilities put millions of websites worldwide at risk, but there are still no reports of mass breaches. We discuss the attacks on PAN-OS, a flaw in Brotli, and the steps businesses need to take beyond patching.

How to apply Secure by Design ideas when rolling out AI agents.

Spammers automatically register hundreds of near-identical domains to fool security solutions. We can block those domains automatically, too.

PHP developers have released patches for two critical vulnerabilities that could allow an attacker to remotely execute arbitrary code on the server without authentication.

People, processes, and technology: if a company has gaps in its defenses, they most likely come from one of these three areas. We walk through a real-world breach, and explain how to stand up to attackers without hiring a huge team or needing to vastly increase the security budget.

Microsoft’s October Patch Tuesday bulletin is breaking records again — it addresses 1,414 vulnerabilities. Here’s what you need to know.

We look at an unusual eavesdropping method that leverages the fundamental physical properties of electronic components.

In April, our experts investigated the PAYLOAD campaign — where the leverage wasn’t data encryption but a malicious AD group policy object.

CVE-2026-87902, a new critical vulnerability in the WordPress core, is already being actively exploited in real-world attacks.

What tokenomics is, and how the value of AI tokens has become a cybersecurity issue.

We believe that investing in cybersecurity, like any other investment, should pay off. This post explains what a small company needs to make its security setup work for the business instead of staying a formality.

We break down the file formats that can be unfamiliar to some users, and that aren’t always scanned by security solutions but can still pose cyberthreats.

Attackers are crafting files that devices can read as two different document types at once, and using these “nesting dolls” to smuggle in malware. How to detect and neutralize these two-faced files?

How attackers distribute ScreenConnect under the guise of free software to deploy AsyncRAT.

We analyze the first-ever real-world incidents where attackers targeted AI agents deployed in corporate environments.

A new variation of ClickFix allows attackers to gain access to Microsoft 365 accounts. We break down how this technique works, and what threat it poses to organizations.

How attackers gain access to corporate services without stealing passwords or cookies: we’re analyzing the Shadow Token via Remote Debug technique used in ToddyCat APT attacks.

Microsoft has addressed approximately 600 vulnerabilities in its products, affecting its entire product line — including even Minecraft Server and Age of Empires II. How can organizations handle such a volume?

A GReAT study has identified ~250,000 potential security issues in publicly accessible GitHub Actions.

Austrian researchers have uncovered a bizarre new way hackers could steal sensitive data.

How to detect and block unauthorized AI tools in an organization.