Microsoft has disclosed the details of how it disrupted a large-scale business email compromise (BEC) infrastructure hosted across multiple Web services, a strategy that allowed attackers to fly under the radar.
The wave of recent high-profile ransomware attacks may be top of mind for business leaders, but BEC remains a prolific — and expensive — enterprise problem. The FBI's Internet Crime Complaint Center (IC3) reported BEC scams numbered 19,369 and cost approximately $1.8 billion in 2020, during a year when total losses from cybercrime exceeded $4.1 billion overall.
Part of the reason why BEC campaigns are successful is their stealthy nature, wrote researchers with the Microsoft 365 Defender Research Team in a blog post on the BEC campaign disruption. These attacks have a small footprint, create low signals that don't top defenders' alert lists, and usually blend in with the typical noise of corporate network traffic.
"The attackers performed discrete activities for different IPs and timeframes, making it harder for researchers to correlate seemingly disparate activities as a single operation," they said of the challenges in analyzing this particular operation.
Researchers traced this campaign to a phishing attack in which criminals stole user credentials to log in to target mailboxes and create forwarding rules that would give them access to emails regarding financial transactions. Before forwarding rules were created, the target mailboxes received a phishing email with a voice message lure and an HTML attachment. These emails came from an external cloud provider's address space, researchers noted.
Throughout their investigation, researchers saw hundreds of compromised mailboxes in multiple businesses. All forwarding rules were configured to send emails to one of two attacker-controlled accounts if the messages had "invoice," "payment," or "statement." Attackers also added rules to delete the forwarded emails from the victim's mailbox.
BEC Criminals Look to Cloud
Microsoft's analysis revealed this campaign was run on a "robust" cloud-based infrastructure that was used to automate attacker operations, which included finding high-value targets, adding forwarding rules, monitoring target inboxes, and handling the emails they were after.
In this case, the attackers intentionally tried to make it difficult for defenders to realize their activities were part of a single campaign — for example, they ran different activities for different IPs and time frames. However, researchers noted the attack was conducted from specific IP address ranges.
"We observed the above activities from IP address ranges belonging to an external cloud provider, and then saw fraudulent subscriptions that shared common patterns in other cloud providers, giving us a more complete picture of the attacker infrastructure," researchers wrote.
They explained how the attackers used a worker structure in the virtual machines in which each VM only executed a specific operation, which is why activities came from different IP sources. The attackers also set up DNS records that appeared similar to existing company domains, so they would blend in with email messages and could be used in targeted phishing campaigns.
This research underscores how BEC attackers are investing more time and effort into evading detection by blending in with legitimate traffic using IP ranges that have a high reputation and ensuring the steps of their attack unfold at different times and locations, researchers noted.
As they learned how attackers were taking advantage of cloud service providers in this campaign, Microsoft's Digital Crimes Unit (DCU) worked with the Microsoft Threat Intelligence Center (MSTIC) to report its findings to cloud security teams so the malicious accounts could be suspended and the infrastructure taken down.