Security Insights
Latest threat analysis, industry news, and security best practices from our expert team.
Detecting Azure AD Enumeration: ROADrecon, AADInternals & Closing Visibility Gaps
Introduction For years, security operations centers (SOCs) have fought a battle of shadows in the cloud. While we excel at detecting on-prem...
Icarus Threat Actor: Detecting Klue OAuth Breach and Salesforce Data Exfiltration
Introduction Security Arsenal is actively tracking an ongoing extortion campaign attributed to the 'Icarus' threat actor, which targets orga...
Anatomy of the 2026 Attack Surface: Defending Against AI Abuse, NastyC2, and OAuth Device Codes
Introduction The internet didn't break this week; it simply functioned exactly as threat actors designed it to. The latest ThreatsDay Bullet...
Orphaned AI Agents: Detecting and Remediating Shadow AI Access Risks
Orphaned AI Agents: Detecting and Remediating Shadow AI Access Risks Introduction The rapid adoption of generative AI and autonomous agents ...
Crypto Clipper Campaign: Defending Against AI-Driven Social Engineering and Fake Repositories
Introduction Security analysts are currently facing a sophisticated evolution in social engineering tactics. A new campaign, identified by C...
NarwhalRAT Delivery via Fake Microsoft Account Alerts — Detection and Incident Response
NarwhalRAT Delivery via Fake Microsoft Account Alerts — Detection and Incident Response Introduction Security Arsenal has detected a concern...
Winning the 72-Minute Race: AI-Driven SOC Defense and Automation Strategies
Winning the 72-Minute Race: AI-Driven SOC Defense and Automation Strategies Introduction The metrics for incident response have shifted irre...
Maine Data Breach Portal Disruption: Defending Public Forms Against Automated Abuse
Introduction The recent shutdown of the Maine Attorney General's online data breach notification portal serves as a stark warning to securit...
Anthropic Fable 5 & Mythos 5 Takedown: Managing AI Vendor Risk Under New Export Controls
Introduction Anthropic has abruptly taken its Fable 5 and Mythos 5 large language models (LLMs) offline. This action is in direct response t...