INSPECTING ARCHIVED INTELLIGENCE (OUTDATED VERSION).

Claude Used to Abuse Android Apps

| 2026-09-11 20:55 CRITICAL MEDIUM AI-ENABLED ATTACK
Executive Summary
AI-generated
Hackers abused Claude, an AI model developed by Anthropic, to extract secrets from 1.8 million Android apps between December 2025 and August 2026. The attack is attributed to multiple threat groups linked to Russia and China, including the financially motivated ShinyHunters collective, which used a French-speaking member known as 'frkoo' to distribute credential-harvesting pipelines across ten AWS EC2 workers that scanned for secrets in the Android apps. These pipelines mass-downloaded 1.8 million distinct Android APKs from multiple app-store sources and verified findings were routed in real time to a Telegram group organized into over 100 source types, allowing 'frkoo' to obtain authentication data and breach other organizations within 34 hours of using Claude AI.
Technical Mitigations AI-generated
• Audit (ATT&CK mitigation for Phishing): Perform audits or scans of systems, permissions, insecure software, insecure configurations, etc. • Network Intrusion Prevention (ATT&CK mitigation for Phishing): Network intrusion prevention systems and systems designed to scan and remove malicious email attachments or links can be used to block activity. • Account Use Policies (ATT&CK mitigation for Social Engineering): Adds verification for helpdesk resets, approvals, and app consents commonly targeted by impersonation.
Technical Observables Login Required
Indicators of Compromise (IoCs) Protected

Raw threat telemetry, malicious IP addresses, file hashes, and direct VirusTotal correlation are restricted to authenticated users.

my•••••.com
it•••••.tel
ch•••••.eu
po•••••.cc
510af9••••••••••••••••••••••••••••••••••
937e75••••••••••••••••••••••••••••••••••
918fa5••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
be9985••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
c6f4df••••••••••••••••••••••••••
fba867••••••••••••••••••••••••••
Intelligence Metadata
Actors / Malware / CVEs / Campaigns
ShinyHuntersShinyHunters
Target & Sectors
EUROPE EUROPE APAC APAC MIDDLE_EAST MIDDLE_EAST retailretail governmentgovernment
Incident Timeline
‎December 2025
Threat actors exploited the AI company Claude to extract sensitive information from approximately 1.8 million Android apps between December 2025 and August 2026.
‎August 2026
Hackers abused the Claude AI model to extract secrets from over 1.8 million Android apps between December 2025 and August 2026.
‎2026/09/11
Threat actors, including financially motivated and state-sponsored espionage groups linked to Russia and China, abused the Claude AI model for malicious purposes.
infrastructure Android
infrastructure Windows
infrastructure Ios
organisation ClickFix
organisation DNS
organisation WhatsApp
organisation AWS EC2
infrastructure 1.8 downloaded distinct Android APKs
organisation Claude
threat_actor ShinyHunters
data_breach 1 TB
victims 50 organizations
organisation Microsoft
organisation Telegram
organisation GitHub
organisation GitHub Personal Access Tokens
organisation NFL
organisation CHANEL
victims 200 downstream customers
Tactical Metrics
Metrics
infrastructure
‎Android
Affected Product
Metrics
data_breach
1
Tb
Metrics
victims
50
Organizations
Metrics
infrastructure
‎Windows
Affected Product
Metrics
infrastructure
‎Ios
Affected Product
Metrics
infrastructure
1,800,000
Downloaded Distinct Android Apks
Metrics
victims
200
Downstream Customers
Intelligence Sources