Microsoft Links ConnectWise Compromise to AI-Assisted Storm-0940 Campaign

Microsoft has attributed the recently disclosed ConnectWise supply-chain compromise to Storm-0940, a group it describes as aligned with the Chinese state. Its threat researchers say the attackers used the agentic AI penetration-testing framework ARTEX to accelerate a campaign targeting organizations across multiple industries.

The attribution is Microsoft’s assessment. Reuters reported that ConnectWise had previously acknowledged investigating unauthorized access associated with its Cloud Backup environment but did not immediately respond to questions about Microsoft’s conclusions.

Trusted applications became an access route

Microsoft says compromised application credentials were used to authenticate to customers’ Microsoft 365 environments through connections that already had permission. In some affected environments, researchers observed follow-on activity including reconnaissance and changes to credentials or privileges.

This is different from saying Microsoft’s own corporate systems were breached. Microsoft says the reported activity did not result from a compromise or vulnerability in its corporate systems, products or services.

The company reports that ARTEX helped the attackers reach more than 2,250 tenants, but that figure is not a count of confirmed organizations with stolen data. Microsoft explicitly cautions that an affected application, or even a successful authentication, does not necessarily establish unauthorized data access or exfiltration. Most targeting in the wider campaign was unsuccessful.

Rapid access can leave lasting changes

In several incidents, Microsoft observed privileged cloud access within less than three hours of initial targeting. Researchers warn that changes to other credentials or identities could preserve access even after the original compromised connection is contained.

Its defensive guidance emphasizes reviewing relevant application sign-ins and unexpected privilege or credential changes, preserving evidence, and following ConnectWise’s formal containment and restoration instructions. It advises against deleting enterprise applications as an initial response before their configuration and evidence have been preserved.

The broader issue is how automation can compress defenders’ response time while abusing trust already granted to a third-party application. The findings do not justify treating every ConnectWise customer as compromised or every AI-assisted security tool as malicious.

Sources

• Microsoft Threat Intelligence: campaign assessment and defensive guidance
• Reuters: reporting and company-response context