How Storm-0940 Used AI to Hack Hundreds of Companies in Hours via ConnectWise
Executive Summary
Beginning September 2026, Storm-0940, a China-state aligned threat actor active since 2021, weaponized ARTEX—an open-source agentic AI penetration-testing framework—to automate large-scale attacks against hundreds of organizations across aerospace, defense, financial services, government, and technology sectors globally. The actor exploited a supply chain compromise of ConnectWise's multi-tenant cloud backup application, using ARTEX to conduct password spraying at 2,900 attempts per minute, OAuth token theft, and Microsoft Graph reconnaissance. In several confirmed intrusions, privileged cloud access was achieved in under three hours, with follow-on activity including application credential modification, privileged role self-assignment, and persistent backdoor establishment across victim Microsoft 365 tenants.
This incident marks a watershed moment in nation-state offensive operations: AI-driven automation has collapsed the attacker dwell-time advantage that defenders historically relied upon. As agentic AI frameworks become commoditized and open-source, the barrier to executing sophisticated, multi-stage supply chain compromises at machine speed is dramatically lowered, forcing organizations to rethink detection and response timelines and elevate cloud identity governance to a board-level priority.
Why This Matters Now
Agentic AI frameworks like ARTEX are now openly available and actively used by nation-state actors to compress attack timelines from days to hours. The ConnectWise supply chain vector demonstrates that trusted third-party OAuth applications can serve as silent entry points into thousands of downstream tenants simultaneously—making traditional perimeter and signature-based defenses insufficient and demanding immediate zero-trust segmentation, continuous Graph API monitoring, and least-privilege enforcement for cloud application permissions.
Attack Path Analysis
Storm-0940, a China-state aligned threat actor, leveraged the ARTEX agentic AI framework to conduct a large-scale supply chain compromise beginning with automated password spraying and exploitation of a ConnectWise multi-tenant OAuth application to gain initial access to hundreds of downstream Microsoft 365 environments. The actor then used stolen application credentials to escalate privileges by assigning high-privilege cloud roles, performed discovery and lateral movement across tenant environments via Microsoft Graph API, maintained command and control through legitimate OAuth application channels, and exfiltrated email and directory data — all within compressed timeframes enabled by AI-driven automation operating at machine speed.
Kill Chain Progression
This analysis maps confirmed threat intelligence to the full cloud kill chain to show where defensive gaps would emerge as an attack progresses.
Initial Compromise
Description
Storm-0940 conducted automated, AI-accelerated password spraying campaigns against Microsoft 365 and Exchange Online accounts from infrastructure at 144.172.64.99, generating over 151,000 failed logon events across 327 accounts in eight organizations between October 10-14, 2026. Concurrently, the actor exploited the ConnectWise supply chain compromise to steal OAuth application credentials for three multi-tenant ConnectWise applications, bypassing the need for interactive user sign-in. ARTEX agents automated reconnaissance and scanning of cloud resources to identify viable targets. The Cloud Native Security Fabric (CNSF) with inline enforcement and real-time inspection of agentic AI traffic could have detected the anomalous automation patterns, repeated malformed authentication requests, and suspicious automation signatures characteristic of ARTEX-driven activity at this stage.
MITRE ATT&CK® Techniques
Active Scanning
Supply Chain Compromise
Valid Accounts: Cloud Accounts
Steal Application Access Token
Account Manipulation: Additional Cloud Credentials
Account Manipulation: Additional Cloud Roles
Account Discovery: Cloud Account
Cloud Infrastructure Discovery
Potential Compliance Exposure
Mapping incident impact across multiple compliance frameworks.
NIST SP 800-53 Rev. 5 – Account Management, Least Privilege, Service Identification and Authentication
Control ID: AC-2, AC-6, IA-9
CISA Zero Trust Maturity Model 2.0 – Continuous Identity Validation and Automated Threat Response
Control ID: Identity Pillar — Visibility & Analytics, Automation & Orchestration
NIS2 Directive (EU) 2022/0383 – Risk Management — Supply Chain Security, Access Control, and Incident Handling
Control ID: Article 21(2)(a)(b)(e)
DORA (EU) 2022/2554 – Third-Party ICT Provider Risk and Resilience Obligations
Control ID: Article 28 — ICT Third-Party Risk Management; Article 45 — Information Sharing
PCI DSS 4.0 – System and Application Account Management; Third-Party Service Provider Risk Management
Control ID: Requirement 8.6, 12.8
NYDFS 23 NYCRR Part 500 – Access Privileges, Monitoring, and Incident Response Planning
Control ID: Section 500.7, 500.14(b), 500.16
Sector Implications
Industry-specific impact of the vulnerabilities, including operational, regulatory, and cloud security risks.
Information Technology/IT
AI-driven ARTEX supply chain compromise of ConnectWise OAuth applications exposed multi-tenant cloud infrastructure, enabling Storm-0940 to escalate privileges and exfiltrate data via Microsoft Graph at machine speed.
Financial Services
Explicitly targeted by Storm-0940's ARTEX campaign, financial services face critical OAuth credential theft risks with PCI.4.0 compliance violations and lateral movement enabling unauthorized cloud account discovery and data exfiltration.
Defense/Space
Aerospace and defense organizations directly named as Storm-0940 targets; nation-state supply chain compromise via ARTEX enables privileged cloud persistence threatening classified data under NIST.800-53 compliance frameworks.
Government Administration
Government entities targeted by China-state actor Storm-0940 face ARTEX-accelerated password spraying and OAuth application abuse, risking persistent backdoor access through credential modification violating ZTMM and NIST.CSF controls.
Sources
- AI-driven, large-scale nation-state campaign using ARTEX enables supply chain compromisehttps://www.microsoft.com/en-us/security/blog/2026/10/21/storm-0940-uses-artex-agentic-ai-connectwise-supply-chain-compromise/Verified
- ConnectWise Trust Center - Cloud Backup Security Advisorieshttps://www.connectwise.com/company/trust/advisoriesVerified
- ConnectWise (SkyKick) Vendor Credentials Abused in M365 Attacks - Huntresshttps://www.huntress.com/blog/connectwise-skykick-vendor-credentials-abused-m365-attacksVerified
- CISA Known Exploited Vulnerabilities Cataloghttps://www.cisa.gov/known-exploited-vulnerabilities-catalogVerified
- MITRE ATT&CK - Supply Chain Compromise (T1195)https://attack.mitre.org/techniques/T1195/Verified
- MITRE ATT&CK - Steal Application Access Token (T1528)https://attack.mitre.org/techniques/T1528/Verified
- Microsoft Threat Intelligence - Nation-State Threat Actor Profileshttps://www.microsoft.com/en-us/security/blog/topic/threat-intelligence/Verified
Frequently Asked Questions
Cloud Native Security Fabric Mitigations and ControlsCNSF
Based on the attack progression modeled above, these are the defensive controls that would constrain each stage.
Aviatrix Zero Trust CNSF is directly relevant to this incident because Storm-0940 relied on uncontrolled cross-tenant API reachability, unconstrained outbound egress paths, and visibility gaps that allowed ARTEX-driven automation to operate at machine speed across hundreds of environments. By enforcing segmentation, identity-scoped access boundaries, and east-west traffic controls, the blast radius of the supply chain compromise could likely have been significantly reduced.
Control: Cloud Native Security Fabric (CNSF)
Mitigation: Inline enforcement and real-time inspection of agentic AI traffic patterns could likely have surfaced the anomalous automation signatures and high-volume malformed authentication requests characteristic of ARTEX-driven activity, reducing the window of undetected initial access and constraining the actor's ability to quietly enumerate viable targets at scale.
Control: Zero Trust Segmentation
Mitigation: Identity-based policy enforcement and least-privilege segmentation controls could likely have constrained unauthorized role assignment activity, reducing the actor's ability to self-elevate from application-level access to Global Administrator and limiting the scope of privilege escalation across tenant environments.
Control: East-West Traffic Security
Mitigation: East-west traffic controls monitoring workload-to-workload and service-to-service communication flows could likely have detected anomalous cross-tenant API pivoting patterns, reducing the actor's ability to freely enumerate directory data across thousands of tenant environments and constraining the lateral reach enabled by the supply chain trust relationship.
Control: Multicloud Visibility & Control
Mitigation: Multicloud visibility controls aggregating and correlating API interaction patterns could likely have surfaced the anomalous combination of machine-speed Graph API calls, Python automation signatures, and activity originating from threat-intelligence-flagged IP ranges, reducing the actor's ability to blend C2 traffic with normal application behavior and enabling earlier detection and response.
Control: Egress Security & Policy Enforcement
Mitigation: Egress policy enforcement with FQDN filtering and outbound traffic controls could likely have constrained unauthorized data flows to attacker-controlled infrastructure, reducing the volume of email and organizational data accessible through the compromised OAuth application channels and limiting the exfiltration paths available to ARTEX-driven agents operating with high-privilege application credentials.
If earlier CNSF controls had constrained privilege escalation scope and limited lateral reachability, the number of tenant environments retaining active backdoor credentials after initial remediation would likely have been smaller, reducing the investigative and containment burden and narrowing the window of residual unauthorized access risk across the affected supply chain environment.
Impact at a Glance
Affected Business Functions
- Cloud Infrastructure and Identity Management (Microsoft 365, Entra ID, Microsoft Graph)
- Multi-Tenant SaaS Application Trust and OAuth Authorization
- Cloud Backup and Disaster Recovery Operations (ConnectWise Cloud Backup)
- Enterprise Email and Collaboration Services (Exchange Online, Microsoft Teams)
Estimated downtime: 14 days
Estimated loss: N/A
Confirmed reconnaissance and credential modification activity across hundreds of downstream Microsoft 365 tenant environments spanning aerospace and defense, financial services, government, energy, semiconductor, technology, military, and legal sectors. In multiple affected environments, AI agents conducted directory enumeration, application credential additions, privileged role self-assignments including Global Administrator, and identity and group modifications. Email and SharePoint data were at elevated risk of access and exfiltration via Microsoft Graph API. The breadth of targeting across 2,250 tenants with privileged cloud persistence achieved at multiple organizations represents a significant potential for sensitive corporate data, government information, intellectual property, and regulated customer data to have been accessed or staged for exfiltration. The presence of successful authentication and follow-on Graph activity in confirmed environments indicates material data exposure risk even where exfiltration was not directly confirmed.
Recommended Actions
Key Takeaways & Next Steps
- • Deploy Cloud Native Security Fabric (CNSF) with real-time agentic AI traffic inspection to detect and block automated, machine-speed attack patterns characteristic of ARTEX-driven campaigns — including anomalous automation signatures, repeated malformed authentication requests, and suspicious API call volumes that exceed human-possible interaction rates, directly addressing the initial compromise vector.
- • Enforce Zero Trust Segmentation with identity-based least-privilege policies across all multi-tenant OAuth application permissions and service principal role assignments, ensuring that no application can self-assign Global Administrator or other high-privilege roles without explicit authorized approval workflows, eliminating the privilege escalation path exploited by Storm-0940.
- • Activate Multicloud Visibility and Control with centralized anomaly detection and baselining across Microsoft 365 tenant Graph API activity to surface east-west cross-tenant pivoting, unusual service principal authentication patterns, and high-volume directory enumeration from known threat infrastructure IPs — enabling defenders to detect lateral movement and C2 activity before privileged access is fully established.
- • Implement Egress Security and Policy Enforcement with FQDN filtering and data loss prevention to restrict unauthorized outbound data flows from cloud workloads to attacker-controlled destinations, combined with Inline IPS (Suricata) signature detection for Python-based automation tooling user agents and known malicious payload patterns associated with ARTEX infrastructure.
- • Apply Encrypted Traffic (HPE) controls with data-in-transit enforcement to ensure all inter-tenant and application-to-cloud communications are encrypted and inspectable, preventing passive interception of OAuth tokens and application credentials in transit, while establishing strict egress monitoring to detect and alert on bulk mail access, SharePoint enumeration, and unauthorized data export activity consistent with supply chain exfiltration TTPs.
