ARTEX Abuse Expands: Cybercriminals Target the Financial Sector with Agentic AI

"Prevention is cheaper than a breach"

Since October 2026, the Cambodia-based threat actor Storm-3358 has been actively targeting financial institutions across Southeast Asia and globally. The group utilizes ‘ARTEX,’ an agentic AI offensive framework, to accelerate every stage of the attack chain, including reconnaissance, exploitation, and lateral movement.

Attack Methodology

Storm-3358 leverages ARTEX in tandem with conventional automated tools to scan internet-facing systems, exploit vulnerabilities like SSRF and anonymous file uploads, and deploy open-source RATs such as WISP, vshell, and AdaptixC2. By utilizing agentic AI, the attackers can perform tasks at machine speed, including brute-force attacks, API abuse, and data exfiltration through compromised cloud identities.

Mitigation and Defensive Strategy

Defenders are encouraged to shift from static defenses to AI-enabled investigations. Key defensive recommendations include:

  • Deception Technology: Deploy honeypots, decoys, and tarpits via tools like Microsoft Sentinel Deception to distract AI agents, consume their resources, and generate high-confidence signals.
  • Behavioral Analysis: Monitor for high volumes of rapid, machine-speed requests and the use of standard penetration testing utilities (Nmap, curl, Playwright) that may carry default user-agent signatures.
  • Unified Security Posture: Utilize the ‘Security Dashboard for AI’ to aggregate risks across Microsoft Defender, Entra, and Purview.

Detection and Hunting

Microsoft has integrated protections across its security stack, including:

  • Identity Protection: Microsoft Entra ID Protection now flags traffic from known agentic framework infrastructure as high-risk.
  • Endpoint & Network Monitoring: Microsoft Defender for Endpoint provides specific detections for suspicious processes, command-line executions, and shell activity commonly associated with ARTEX.
  • Threat Hunting: Advanced hunting queries for Microsoft Defender XDR and Sentinel are available to help security teams identify the use of tools like ‘curl’ for file downloads and to track known malicious IP indicators.

Indicators of Compromise (IOCs)

While threat actors frequently rotate IP addresses, the following IPs have been linked to ARTEX and C2 infrastructure as of November 2026:

  • 199.19.95.66, 199.19.95.67, 199.19.95.68 (WISP C2)
  • 91.92.240.238, 85.11.168.45, 178.16.54.129, 31.77.229.35 (ARTEX servers)
  • 85.11.167.52 (General C2 server)

Organizations are encouraged to leverage Microsoft Security Copilot to summarize incidents, analyze malicious scripts, and expedite hunting efforts against this rapidly evolving threat.

Leave A Comment

Name*
Message*

Scroll to top