Star Blizzard (COLD RELIC) abuses stolen cloud data to breach US orgs
Russian group COLD RELIC (Star Blizzard) reused data from one US breach to compromise another; enforce phishing-resistant MFA and audit passkeys.
- Published
- Updated
- Ecosystem
- CloudEnterprise software
- Exploitation
- Yes, Mandiant investigated a late-2025 and a 2026 intrusion at two US organizations, and Google reports new phishing operations as recently as November 2026
What happened
Google Threat Intelligence Group (GTIG) and Mandiant describe a series of linked intrusions by the Russian espionage group COLD RELIC, formerly tracked as COLDRIVER and UNC4057 and also known as Star Blizzard, Callisto and Gossamer Bear. Google Threat Intelligence says the US Department of Justice has assessed the group operates within Russia’s FSB. SecurityWeek likewise reports a formal US association with the FSB.
Case 1, late 2025. According to Google Threat Intelligence:
- The group compromised the cloud productivity environment of a US-based organization, using a residential proxy network located close to the victim.
- It authenticated with stolen Super Admin credentials plus an MFA push notification fatigue attack. After the user accepted an MFA request, it registered a device model not previously seen in the tenant.
- It kept access with passkeys, MFA backup codes and rogue MFA devices.
- It downloaded the entire cloud storage drive, both by mass sharing files to an attacker-operated email account and by direct download.
- It queried admin audit logs and viewed, modified and deleted Alert Center notifications to remove evidence.
- It exported full employee lists and started a domain-wide Google Workspace Customer Takeout export.
- Access lasted several months. It included OAuth inbox sync to external email clients and re-authentication with attacker-enrolled passkeys that survived password and session resets.
- After removal, the group used email accounts impersonating the victim’s employees to target staff and associates in the same professional circles.
Case 2, 2026. Google Threat Intelligence reports the following:
- The group impersonated an employee of the Case 1 organization and contacted a second US organization over an encrypted messaging app.
- That organization used Microsoft Entra ID as its federated SSO identity provider for Google Workspace.
- The group shared phishing links designed to steal Entra ID credentials and tokens, which redirected the victim to document lures matching the conversation.
- It then registered a new MFA device in the SSO solution and bulk downloaded files from the user’s cloud drive.
- Afterwards, the messaging account changed its profile name to a new persona. GTIG believes this was likely used to target other organizations in the same circles.
Google also describes a typical phishing link: a compromised website redirects to the real phishing domain, with a query parameter that often names a prior victim or lure. GTIG observed new phishing operations as recently as November 2026.
The organizations were not named. The Hacker News calls the first one unspecified, and the sources do not say how the Super Admin credentials were first obtained (not confirmed). Publication dates differ slightly: the headline list and Infosecurity Magazine place the Google blog on 17 November 2026, while the Google page itself shows 18 November 2026.
Am I affected?
The sources do not give indicators of compromise for you to match against. They describe who the group targets and which setups it abused:
- Google Threat Intelligence lists government officials, defense contractors and NGOs across Europe and North America as past high-value targets.
- The intrusions involved Google Workspace tenants and an Entra ID tenant federated to Google Workspace.
- People in the same professional circles as the two victims were targeted with impersonation, so unexpected contact from a known colleague on a messaging app that leads to a document link deserves scrutiny.
- Google says it issues Government-backed attack alerts when it detects COLD RELIC targeting Google accounts.
What to do now
The recommendations below are those Google set out, as summarized by Infosecurity Magazine, plus log details from Google’s own post.
- Enforce phishing-resistant MFA, particularly hardware keys, which cannot be intercepted or approved via fatigue attacks.
- Restrict and audit passkey registrations, and review MFA devices and backup codes enrolled on admin and user accounts.
- Enrich authentication logs with residential proxy and VPN IP reputation feeds, and apply conditional access policies to block or detect logons from known proxy networks.
- Apply cloud storage data loss prevention (DLP) policies to warn or block creating, uploading or externally sharing files that match sensitive patterns. Google notes the intrusions were helped by users storing credentials and other secrets in cloud storage.
- In Google Workspace, look for Customer Takeout requests (logged as CUSTOMER_TAKEOUT_CREATED in the Admin audit logs). Administrators can also disable Google Takeout for users at the organizational level, per Google Threat Intelligence.
- If an intrusion is identified, revoke all user sessions and force password resets for impacted accounts. Because attacker-enrolled passkeys persisted across password and session resets in Case 1, also remove any rogue passkeys, MFA devices and backup codes.
Sources
- Google Threat Intelligence: COLD RELIC: A Case Study in Impersonation and Compromise
- The Hacker News: Russian COLD RELIC Hackers Use Stolen Google Workspace Data to Breach Second U.S. Organization
- Infosecurity Magazine: Russian Espionage Actor Star Blizzard Focuses on Cloud Identities for Mass Data Theft
- SecurityWeek: Russian APT Poses as Employee of Hacked Firm to Breach Another US Organization