← Breach analysis by the Dual Vulnerabilities framework

Red October aka Inception Framework: Cloud Atlas cyber-espionage campaign (Red October / Cloud Atlas / Inception Framework, 2014)

2014 · O - Public administration and defence · Cyber-espionage campaign via spear-phishing with trojanized Office documents

Analysed 2026-11-06 by Hermes

Open this case in the tool →

What happened

Cyber-espionage campaign via spear-phishing with trojanized Office documents

Dual vulnerability type

No DV evidenced

Alternative: Expertise to Execution

Why this classification

(1) No strategic expectation can be established at the victim, because the row describes a campaign, not an organisation: the case file carries no receiver_name and its receiver fields repeat target categories (government/ministries, energy, telecommunications, finance, defence industry, other corporate targets) across ten countries, and none of the reporting names a single breached organisation. (2) What the sources document is entirely attacker-side: the Red October group returned in 2014 (Cloud Atlas, named by Kaspersky; Inception Framework, named by Blue Coat) with spear-phishing emails carrying trojanized RTF/Word documents exploiting a variation of CVE-2012-0158, implanting backdoors that ran command-and-control over WebDAV/HTTPS through free CloudMe accounts, and exfiltrated confidential data from targets in Russia, Kazakhstan, Belarus and elsewhere, later hiding behind proxy and cloud layers; the campaign was discovered by Kaspersky in August 2014, exposed publicly in December 2014, and disclosed by IT-security companies - no victim statement, regulator finding, recovery account or notification timeline exists anywhere in the record. (3) Under the framework's decision rule this is Unknown: the reporting confirms impact only, no lifecycle stage of any individual victim organisation is documented well enough to answer the origin x break test, and the case is atypical - a state-linked espionage operation against a diffuse population of state and critical-infrastructure targets rather than a single victim organisation. One candidate mechanism is noted but deliberately not load-bearing: the exploited Office vulnerability had been patched since March 2012 (MS12-027), so a patch-latency condition can be inferred at the population level, but it rests on the actor's campaign-wide pattern and no incident-specific victim evidence, and inferring a per-victim knowing-doing gap from it is exactly the reasoning the method rejects.

Sources

Notes

Data quality: the case file names no victim organisation; receiver_category/receiver_subcategory repeat the same seven-position target block about ten times and receiver_country lists Russia, Ukraine, Moldova, Belgium, Iran, France, Bulgaria, the United States, Turkey and Georgia, so the row encodes the campaign's target population rather than one victim - Country is left blank per the method spec for diffuse multi-organisation campaigns, and Sector codes the dominant target category (government ministries, plus energy, telecom, finance and defence industry) rather than one organisation's activity. Atypical case: state-linked espionage campaign against a diffuse target population, disclosed by IT-security companies (eurepoc_extra source_disclosure); no victim-side disclosure, regulator or legal response exists in the record, which is why the DV-Type is Unknown. The campaign exploited a variation of CVE-2012-0158 (patched March 2012 via MS12-027) and later CVE-2014-1761; that group-wide pattern could support an Expertise to Execution reading only at population level, not for any identifiable victim. The cited Symantec blog URL from the case file now redirects to the identical article on Broadcom's security.com blog domain.