Storm-0940 Exploits AI Agents in Sweeping ConnectWise Breach

A state-aligned hacking group from China has leveraged autonomous artificial intelligence tools to execute a sophisticated supply chain attack against ConnectWise. This offensive, tracked by Microsoft as Storm-0940, utilized ARTEX, an open-source framework designed for penetration testing, to infiltrate a multi-tenant application. Once inside, the group attempted to sign into more than 2,250 Microsoft 365 tenants. The operation was notable for its speed, with some attacks reaching privileged access levels in less than three hours after the initial targeting began.

The scale of the intrusion was massive and largely automated. The attackers employed password spraying, a technique involving automated guessing, which targeted 327 accounts across eight organizations. This resulted in over 151,000 failed logon attempts, though none were successful. The use of ARTEX allowed the team of language model driven agents to run autonomous tests with minimal human interaction. This setup sped up scanning, reconnaissance, and the abuse of OAuth applications, leaving defenders with very little time to detect and respond to the threats.

Storm-0940 has been active since 2021 and is identified as a China-state aligned actor with no known industry aliases. The campaign specifically targeted industries including aerospace and defense, semiconductors, government agencies, and think tanks, with fourteen organizations listed in total. Microsoft clarified that the activity did not stem from a vulnerability in its own systems or services. Instead, the group exploited stolen credentials from ConnectWise apps, allowing them to authenticate without interactive user sign-ins.

The intensity of the attack peaked on October 11, when password spraying reached 2,900 attempts per minute. That day saw more than 111,000 failed attempts, with 48 accounts experiencing over 1,000 failures each. Almost 98% of this activity was locked out. However, the AI agents managed to sign into customer tenants from the same IP range using stolen application credentials. In one instance, more than 200 sign-ins were executed in under four minutes using just four application credentials. Within these tenants, the agents performed directory operations, reconnaissance, and changes to identities and privileged roles.

The stakes for affected businesses were high, as the attackers planted new secrets set to expire in 2028. Huntress, a security firm, observed an app adding itself to the top administrator role in one tenant on October 15. By October 16, unauthorized sign-ins were confirmed for 162 tenants. The following day, nearly 1,000 Entra tenants showed sign-in activity from the attacking IP, with more than 300 succeeding. The systematic sweep through tenants suggested a script was working from a predefined target list.

This incident highlights a shift toward using open-source AI frameworks, which can utilize various AI models. Public reporting this month linked similar technology to breaches at South Korean financial institutions. In response, ConnectWise suspended several services including Cloud Backup and SaaS Security as a precaution. The company has begun a phased restoration process, with recovery expected to take several days. There are no current indicators that partner backup data was compromised, and Microsoft advises customers to keep affected app connections disabled until formal restoration instructions are issued.

The developer of ARTEX, known by the handle Autumn-27, announced on GitHub that the project will no longer be updated and will become closed source. This move follows reports that the framework was used in other regional attacks, potentially involving Anthropic's Claude Code. For New England businesses and other enterprises relying on cloud management tools, the incident underscores the growing speed and autonomy of modern cyber threats.