CybersecurityMalware & Ransomware
COLD RELIC phishing campaign targets public sector firms, Google warns
Google says the group uses stolen cloud data to impersonate trusted contacts and recommends stronger authentication.

Key takeaways
- Google says COLD RELIC remains active and has launched new phishing attacks in November 2026.
- The group uses stolen cloud account data to impersonate trusted contacts in follow-on attacks.
- Two cases examined by Mandiant involved US-based organisations, MFA abuse and data theft.
- Google recommends phishing-resistant authentication, session revocation and monitoring for unusual logins.
In this article
Google warns of active campaign
Google has warned that the COLD RELIC phishing and malware campaign remains active. The Russian-linked espionage group is still targeting public sector organisations and other high-value targets, according to TechRadar. Google says it saw new phishing activity earlier in November 2026.
The group is also tracked as COLDRIVER, UNC4057, Star Blizzard, Callisto and Gossamer Bear. Google says the US Department of Justice has linked it to Russia's Federal Security Service. The group has operated for over 10 years and has focused on data that may be useful to Russian government interests.
Google's Threat Intelligence Group says COLD RELIC relies on stolen accounts and internal data to support later attacks. Google says the group has targeted government officials, defence contractors and NGOs.
How stolen data fuels new attacks

Google described two cases examined by Mandiant, both involving US-based organisations. In one case, attackers stole the sign-in details of a Super Admin account. The account used multi-factor authentication, which adds an extra check beyond a password.
The attackers sent hundreds of approval requests until the target accepted one to stop the prompts. They then moved through the cloud productivity environment, downloaded files from cloud storage, exported employee lists and carried out a domain-wide data export. They also removed records of security investigations and email audits to cover their tracks.
The stolen material included emails, contact lists and details about the organisation. This helped the group strike a different company linked to the first victim. A year later, it pretended to be a staff member from the earlier victim and contacted another US-based organisation.
The messages contained links that tried to capture sign-in details and access tokens for Microsoft Entra ID. Those tokens act like digital keys that can grant access to services. Google says this method has become a regular part of COLD RELIC's operations.
Who is being targeted
Google says COLD RELIC mainly targets people and organisations with information that could be useful to the Russian government. The group has used these intrusions to support espionage goals.
The report also highlights the risk to public sector bodies and companies connected to them. Once attackers have internal emails, staff lists, resumes or event schedules, they can build convincing messages that appear to come from a trusted source. This makes employees more likely to click links or approve requests.
Google says the group collects these details to create believable background stories for future phishing. The approach can make a message seem routine, such as a request from a colleague or a known partner. That increases the chance that a target will respond without suspicion.
Why it matters
The campaign matters because a single compromised cloud account can expose a wider network of contacts and partners. Google says the group uses stolen emails and internal data to make later attacks more convincing. That can lead to more credential theft, unauthorised access and loss of sensitive information.
Public sector organisations may be especially exposed because they hold data of political or strategic value. The report also shows how attackers can abuse trusted business relationships. Employees may see messages that appear to come from a known colleague or partner, making the attacks harder to detect.
How organisations can respond
Google recommends stronger authentication. Organisations should use methods that resist phishing, such as hardware security keys, FIDO2 or WebAuthn authentication, and passkeys. Google Titan keys are one example of a hardware security key.
Google also advises security teams to watch for suspicious passkey and device registrations. If an account compromise is suspected, they should revoke existing sessions and remove OAuth permissions, which are approvals that let apps access accounts. Stronger controls over cloud access are also recommended.
Monitoring logins from residential proxy networks is another step. Such networks route traffic through home internet connections and have been used in the campaign. Google says these measures can help disrupt the activity.
Google says it will keep tracking the group's infrastructure and working to disrupt it. Readers should consult Google's Threat Intelligence Group report and their cloud provider's official security guidance for the latest defensive advice.
Sources
Spotted a mistake? See our corrections policy. Read how News by OnDevice works in our editorial policy.



