Storm-0940 Leverages AI Agents to Breach ConnectWise and Sweep Microsoft 365 Tenants

A hacking group known as Storm-0940 has launched a significant supply chain attack by deploying AI-driven agents against the ConnectWise platform, according to recent analysis from Microsoft. This China-state-aligned actor utilized ARTEX, an open-source framework designed for autonomous penetration testing, to execute a large-scale intrusion. The group’s primary objective was to exploit unauthorized access to the multi-tenant application, allowing them to authenticate as the software itself rather than as individual human users. This approach bypassed standard interactive sign-in procedures, enabling the attackers to move rapidly through the network with minimal friction.

The scale of the operation was substantial, with the attackers attempting sign-ins to more than 2,250 Microsoft 365 tenants. By leveraging stolen credentials associated with ConnectWise applications, the AI agents could access systems without the need for traditional user interaction. The speed of the intrusion was particularly alarming; in several instances, the group achieved privileged access to tenant environments in less than three hours after initial targeting. This rapid escalation highlights how automated tools can compress the timeline for a breach, leaving security teams with very little time to detect and react to the unfolding incident.

Beyond the direct application abuse, the attackers conducted extensive password spraying campaigns, targeting 327 accounts across eight organizations. This phase resulted in over 151,000 failed login attempts, with the majority of the activity occurring on October 11. At its peak, the automated system generated 2,900 failed attempts per minute. While nearly 98% of these attempts were locked out by security controls, the volume of traffic demonstrated the capacity of the AI agents to work continuously with little to no human oversight. The use of ARTEX, which employs large language model-driven agents, accelerated reconnaissance and scanning processes far beyond the pace of manual testing.

The attack targeted specific industries, including aerospace and defense, semiconductors, government entities, and think tanks. Microsoft clarified that the breach did not stem from a vulnerability in its own systems but rather from the compromise of the ConnectWise application. Security firm Huntress observed specific behaviors in one tenant on October 15, where an application added itself to the top administrator role, a pattern that triggered Microsoft Defender alerts. By October 17, Huntress had confirmed unauthorized sign-ins for nearly 1,000 tenants, with more than 300 of those attempts succeeding. The systematic, sorted order of the tenant sweep suggested the attackers were working from a pre-defined list using automated scripts.

The deployment of ARTEX marks a shift in how threat actors conduct operations, as the framework can utilize various AI models and is being used by groups in multiple regions. In a related development, public reports have linked the same framework to breaches at South Korean financial institutions, where the suspected attacker allegedly used Anthropic’s Claude Code. The developer of ARTEX, identified by the handle Autumn-27, recently announced that the project would become closed source and would no longer receive updates. This move may impact the availability of the tool for other actors, though the immediate threat remains high.

In response to the incident, ConnectWise suspended several of its services, including Cloud Backup and SaaS Security, as a precautionary measure. The company has begun a phased restoration process, with full recovery expected to take several days. While there are currently no indicators that partner backup data was compromised, Microsoft advises customers to keep affected application connections disabled until formal restoration instructions are issued. Organizations are encouraged to review their activity logs from this month to assess the extent of the exposure and ensure that any unauthorized access is properly mitigated.