Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

Arctic Wolf Labs has observed a recurring social engineering pattern in hundreds of cases where threat actors have used Microsoft Teams’ external chat creation feature for impersonation scams, or to deliver malicious files. A consistent pattern of attacker behavior has emerged across these incidents: the use of the same throwaway tenants, the same “IT Support (Internal)” display names, the same remote-access tools, and increasingly, the same payload domains.

AI is changing how vulnerabilities are discovered and exploited for defenders and attackers alike, and the practical effect is that the time organizations have between a flaw being found and a flaw being exploited keeps shrinking. Today, we have more known vulnerabilities than at any point in history, and they’re being exploited faster than ever.

On October 21, 2026, the PHP development team released emergency security updates addressing two critical vulnerabilities, CVE-2026-109618 and CVE-2026-109619, collectively referred to by researchers as PHP2Shell. Both vulnerabilities carry a CVSS v4.0 score of 9.2 (Critical Severity) and can allow an unauthenticated attacker to execute code with the privileges of the affected PHP worker process.

On Tuesday, October 20, threat actors gained advance visibility into two critical PHP vulnerabilities after details leaked ahead of the planned disclosure, accelerating the race between defenders and attackers. The flaws are tracked as Common Vulnerabilities and Exposures (CVE) identifiers CVE-2026-109618 and CVE-2026-109619. Together they’re known as PHP2Shell. Both carry a 9.2 rating on the Common Vulnerability Scoring System (CVSS) version 4.0 scale.

In the recently released 2026 SANS AI in Cybersecurity Survey, nearly two out of three security professionals said they have been given AI-generated guidance that later turned out to be wrong. Not from an experimental tool off in a sandbox, but from systems built into the security stack they were told to rely on. Security teams should not have to live with that.

AI now plays a central role in defending organizations from the modern threat landscape. It triages alerts, investigates security events, and recommends what to do next. AI is critical to defending at machine speed, but security leaders can’t see how it is making decisions on their behalf. This visibility challenge creates multiple downstream problems for security leaders that need to justify AI decisions that they can’t fully explain to their board, regulatory bodies, and insurance partners.

Imagine this: There’s an active attack but no alert. Nothing looks wrong. Every login is valid. Every tool is trusted. Every command is one your team has run before. This invisible attack chain is how many attackers work today, and they can take control of a domain in minutes and encrypt an environment while evading traditional detections. It’s what happens between the login and the ransom note that’s the modern cyber battlefield.

A lot of the current conversation about agentic SOCs is about triage. Faster detection, faster investigation, alerts resolved in minutes instead of hours. When we set out to build the Aurora Agentic SOC, we didn’t just want to make triage faster. We fundamentally re-engineered our SOC to strengthen every job it does, and enhancing the Concierge Experience for our customers was a key part of that.

New training on using AI safely and spotting AI-powered threats arrives this Cybersecurity Awareness Month Attackers are moving faster than ever, weaponizing newly discovered vulnerabilities before defenders can patch them and building new types of malware that signature-based defenses can’t catch.

On September 27th 2026, Citrix disclosed multiple vulnerabilities affecting NetScaler ADC and NetScaler Gateway. Arctic Wolf tracking indicates that CVE-2026-88771 and CVE-2026-88772 have been exploited in attacks against NetScaler devices as zero-days. Additional CVEs are also disclosed in the Citrix Security Bulletin. CVE-2026-88771 is an unauthenticated remote code execution vulnerability caused by improper input validation that can allow arbitrary command execution.