The Pulse
Storm-3358 began targeting organizations in the financial sector with ARTEX in October 2026, Microsoft says
Storm-3358 began targeting organizations in the financial sector in October 2026 using the agentic AI offensive framework ARTEX, Microsoft says. Microsoft observed Storm-3358 use ARTEX to scan systems connected to the public internet, explo
AI.info Team · Written with AI · fact-checked ·
Storm-3358 began targeting organizations in the financial sector with ARTEX in October 2026, Microsoft reported
Storm-3358 began targeting organizations in the financial sector in October 2026 using the agentic AI offensive framework ARTEX, Microsoft Threat Intelligence says. Microsoft Threat Intelligence tracks Storm-3358 as a Cambodia-based intrusion cluster that primarily targets financial organizations in Southeast Asia, in addition to a smaller number of large financial organizations globally. Microsoft observed Storm-3358 use ARTEX in combination with conventional automated tools to scan systems connected to the public internet and exploit security flaws. Those flaws included anonymous file upload and server-side request forgery, or SSRF.
Recent ARTEX-related attacks involve web applications, credentials and cloud identities
In recent ARTEX-related attacks, threat actors have used ARTEX agents to scan and exploit vulnerabilities in web applications and use web shells for persistence on compromised web servers. The agents also compromised credentials from public sources and compromised systems, and used compromised cloud identities for persistence and data exfiltration. ARTEX agents can perform brute-force password attacks and password spraying, enumerate and abuse APIs, and dump databases, Microsoft says. Agentic tools like ARTEX can leverage compromised credentials and systems to move laterally, escalate privileges and exfiltrate data at machine speed, the company says.
ARTEX uses penetration-testing tools by default
Storm-3358 has used the WISP, vshell and AdaptixC2 remote access trojans for persistence and lateral movement, Microsoft says. Microsoft observed ARTEX agents using netcat and curl to download those remote access trojans to compromised systems. ARTEX by default uses Nmap, curl, wget, Python HTTP clients, Norma SDK WebFetch and headless Chromium through Playwright, among others, Microsoft says. A wide variety of requests in a brief time, retries of variations on the same technique, or movement to next-stage actions at machine speed can help organizations determine that they are being targeted by agentic AI frameworks, according to Microsoft.
Microsoft points defenders to decoys and detection signals
The tools used by ARTEX may send network traffic with distinctive characteristics, including default user agent strings, Microsoft says. Blocking traffic with those characteristics will not stop an attack by an AI agent, the company says, because the agent will quickly change them; identifying the signals can still alert defenders. Defensive honeypots, decoys, tarpits and fake wins can pull autonomous AI attacks away from legitimate targets, Microsoft says. Honeypots and decoy surfaces offer attractive but worthless targets, while tarpits consume an agent’s time and budget.
Fake wins can present an agent with convincing evidence that its objective is complete, potentially causing it to stop on a false success. In Microsoft’s testing, the controls helped reduce the attack traffic reaching real assets and produced high-confidence, attribution-quality signal. Microsoft’s research into threats involving agentic frameworks informs security detections and investigative context across its products, and some Microsoft Defender alerts may trigger on unrelated activity and should be investigated to determine whether they relate to the activity outlined in the blog.