Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

Arctic Wolf has launched Aurora MDR Connect, a new tier of Aurora Managed Detection and Response (MDR) built exclusively for managed service providers (MSPs).

Anthropic CEO Dario Amodei published an essay this month called “We Must Pace the Frontier.” His argument is straightforward, calling out the reality that AI capabilities are advancing faster than the industry’s ability to align and safeguard them, and frontier labs need to slow the rate of capability growth long enough for safety work, alignment, and interpretability to catch up. He points to two developments behind his concern.

CVE-2026-76461 is a pre-authentication SQL injection vulnerability in the email parsing logic of Cisco Secure Email Gateway (AsyncOS). This vulnerability enables unauthenticated remote threat actors to execute arbitrary code as root by sending crafted, malicious emails. This grants threat actors full control over the operating system, allowing data exfiltration, email surveillance, persistent access, and potential network pivoting, all without user interaction or credentials.

Patches have now been released addressing two critical vulnerabilities: CVE-2026-85102, which affects Security Gateway and Spark Firewall, and CVE-2026-85103, which affects Security Gateway, Management Server, and Spark Firewall.

A critical security weakness (CVE-2026-84869) has been identified in the ConnectWise ScreenConnect client (prior to version 26.6.5), where missing authorization controls and improper privilege management allow file transfers and execution through active remote sessions without host confirmation.

There’s been a lot of buzz lately about using AI to automate Tier 1 and Tier 2 SOC work. Anthropic, using Claude as the base, has been leading with some of the most specific “how to” research.

OpenAI published its assessment of its newest GPT model, Astra, and found it to be the first of their models to reach a critical level of cybersecurity capability, meaning that given the right tools and access, it could autonomously exploit previously unknown vulnerabilities. As a result, OpenAI has restricted Astra’s most advanced cybersecurity capabilities to trusted partners before a public rollout.

A maximum-severity (CVSS 10.0) vulnerability CVE-2026-86218 has been discovered in N-able N-central prior to build 2026.3.1.14. This flaw allows unauthenticated attackers to execute arbitrary code on the N-central server before authentication enabling remote takeover of the platform. The vulnerability is classified as static code injection (consistent with CWE-96) in a public-facing application endpoint.

On September 3, 2026, a security researcher known as Nightmare Eclipse/Chaotic Eclipse publicly disclosed a zero-day dubbed ‘FalconFlank’ which abuses the Office malicious macro remediation workflow in CrowdStrike Falcon Sensor. The attack leverages a time-of-check to time-of-use (TOCTOU) race condition, allowing an attacker with code execution on a vulnerable system to hijack the Falcon macro remediation routine. This results in DLL side-loading and execution as NT AUTHORITY\SYSTEM.