The Russian cyberespionage group known as Cold Relic is turning cloud breaches into a springboard for targeting a victim organization’s wider network of trusted relationships.

It’s a tactic that researchers at Google Threat Intelligence Group (GTIG) observed the attacker deploy with clinical efficiency in two separate compromises involving US organizations over the past year.

In the first intrusion, late last year, the attackers gained control of a cloud administrator account and harvested internal communications, employee details, and other organizational records. Cold Relic actors later used that material to impersonate an employee of the first organization for a successful social engineering attack on an individual at a second organization.

The activity, GTIG researchers said, shows how traditional social engineering is becoming more closely tied to attacks on cloud identities. Once inside an organization’s cloud environment, an attacker can gather the kind of internal information that makes a message from a supposedly trusted colleague or business contact much harder to spot as a scam.

Related:China-Nexus Actors Breach 10 Organizations in AI-Driven Attacks

The campaigns “demonstrate the convergence of human-centric social engineering and cloud identity abuse,” GTIG researchers wrote in a blog post. “By obtaining access to strategic personnel rosters and operational documents, Russian intelligence services turn legitimate business communications into attack vectors against trusted partners.”

A Tale of Two Breaches

Cold Relic (aka also known as Star Blizzard, Callisto, and Gossamer Bear) is a cyberespionage group that Google researchers have been tracking — and disrupting — since 2015. The US government has assessed the outfit as working within Russia’s Federal Security Service (FSB) and tasked with collecting information of political and strategic interest to Moscow.

Though low-profile, Cold Relic has notched up numerous high-value targets including defense contractors, government officials, and NGOs across North America and Europe. One of its favorite post-compromise activities, according to GTIG, is to steal emails and data that can help it establish “authentic pretexts” and convincing impersonations when reaching out to other potential targets.

In the first intrusion in late 2025, Cold Relic used stolen credentials for a super admin account to access a US organization’s cloud environment. The attackers routed their activity through a residential proxy located near the organization, so their connections appeared to originate from a familiar geographic area. The attacker repeatedly kept sending the actual account holder multifactor authentication (MFA) push notifications until the overwhelmed user approved one.

Related:Inside an AI Attack Against a US Retailer

After gaining access, Cold Relic actors registered their own device on the network and established additional ways to authenticate to it, including via passkeys, backup MFA codes, and additional rogue MFA devices. The threat group used the stolen Super Admin user privileges to exfiltrate the organization’s cloud storage, including full employee lists, email conversations, and organizational directories and contact lists. The threat actors then initiated several anti-forensics measures that allowed it to maintain access to the compromised network for several months.

In the second attack, earlier this year, Cold Relic impersonated an employee of the first company to contact an individual at a second organization through an encrypted messaging app. The threat actors sent the unsuspecting target phishing links disguised as documents relevant to their ongoing conversation.

After stealing the victim’s Microsoft Entra ID credentials and authentication tokens, Cold Relic used them to access the organization’s Google Workspace environment through its federated single sign-on system. The attackers then helped themselves to files from the compromised user’s cloud drive while also seeking out email archives, chat records, credentials, personnel information, and other data of potential intelligence value.

Related:Artex: The AI Pen-Test Tool Hackers Turned Into a Weapon

A Cold Relic Lesson for Organizations

“Cold Relic is highly adept at leveraging social networks and impersonation of key individuals to achieve a realistic pretense for reaching out to the targeted individual or organization,” Wes Shields, security engineer at GTIG, tells Dark Reading. “Once they have achieved their initial access goal, they move quickly to maintain persistence and exfiltrate data.”

The lesson for organizations is to train employees to be wary of unexpected messages, particularly when they arrive through communication channels a supposed colleague or business contact doesn’t normally use. If a message seems unusual, employees should verify the sender’s identity through a separate, previously established channel before responding or acting on it, Shields advises.

“In today’s hyper-connected world, it is impossible to have employees not be reachable via multiple channels, and attackers are taking advantage of this,” he says.

Joseph Spero, principal incident response consultant at Google, recommends other measures to limit the damage from identity-based attacks. These include enforcing phishing-resistant MFA, restricting and auditing passkey registrations, hardening MFA registration procedures, and auditing Oauth grants.

“Short-lived sessions and principle of least privilege will limit the potential scope of access when credentials are compromised,” he says. “When a compromise is suspected, mandate session revocation in addition to resetting passwords and MFA mechanisms.”





Source link

#

No responses yet

Leave a Reply

Your email address will not be published. Required fields are marked *