Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

On April 4, 2026, Fortinet released a hotfix for a critical vulnerability in FortiClient EMS (CVE-2026-35616) that allows unauthenticated remote threat actors to execute unauthorized code or commands via crafted requests. The flaw stems from improper access control in the API authentication. Fortinet has confirmed observing exploitation of CVE-2026-35616 in the wild. The vulnerability was responsibly disclosed by Defused, which had observed exploitation prior to Fortinet’s official disclosure.

On March 10, 2026, Progress ShareFile released fixes for two critical severity vulnerabilities in Progress ShareFile Storage Zones Controller (SZC) 5.x, tracked as CVE-2026-2699 and CVE-2026-2701.

At RSAC 2026, Arctic Wolf set the agenda for the future of cybersecurity and AI. Throughout the week, we were at the center of the industry dialogue, shaping how the market is approaching agentic AI in cybersecurity and setting clear expectations for where the industry is headed next. The launches of the Aurora Superintelligence Platform and the Aurora Agentic SOC raised the bar for the industry.

Security operations teams face an overwhelming challenge: making sense of massive volumes of telemetry. Even well-resourced organizations struggle to apply this data effectively. Traditional SIEM platforms require tuning, maintenance, and constant care. Meanwhile, some managed detection and response (MDR) solutions often deliver findings but may not provide accessible ways to dig deeper into the underlying telemetry.

Security teams are under pressure to demonstrate measurable progress against an increasingly complex cybersecurity landscape. Framework expectations evolve, insurance requirements tighten, and executive stakeholders demand defensible evidence that investments are improving risk posture. Yet most organizations still rely on static assessments — point-in‑-‑time documents that provide limited visibility and quickly lose relevance as environments change.

The recent leak related to Claude Mythos has offered a rare and revealing look inside the real capabilities of frontier AI models. The details of the leak underscore a reality that cybersecurity leaders need to understand clearly: Advances in model capability do not automatically translate into advances in cybersecurity, nor do they translate into better security outcomes without the right platform to apply them.

On March 28, 2026, F5 updated its security advisory for a vulnerability impacting BIG-IP APM that was originally disclosed in October 2025 (CVE-2025-53521). The vulnerability was initially classified as a medium-severity denial-of-service (DoS) issue but has been reclassified as a critical remote code execution (RCE) vulnerability. F5 has stated CVE-2025-53521 is being exploited by unauthenticated remote threat actors to deploy web shells.

The widely used Axios npm package, a JavaScript library that enables applications to make HTTP/S requests and is included as a dependency in millions of applications, was compromised in a supply chain attack on March 31, 2026 (UTC).

Arctic Wolf has recently observed a phishing campaign targeting Microsoft 365 that abuses the OAuth device code flow to trick victims into providing authentication codes. Threat actors use Railway’s Platform-as-a-Service (PaaS) infrastructure (a trusted cloud platform with valid IP addresses) to host attack components, allowing the activity to blend in with normal traffic. This enables threat actors to steal valid access and refresh tokens and bypass multi‑factor authentication protections.

Customers do not experience AI as architecture. They experience it as outcomes. They experience it in the quality of the signal they receive, the speed of the investigation, the confidence behind the recommendation, and the amount of time their teams can spend being proactive instead of buried in noise. That is why the most important question in cybersecurity today is not whether a vendor has AI. It is whether that AI produces better outcomes. Security teams are not buying AI for its own sake.