Back to Intelligence

SAFEPAY Ransomware Gang: 9 New Leak-Site Listings in 24 Hours — Sector Targeting Analysis & Detection Engineering

SA
Security Arsenal Team
November 21, 2026
13 min read

Classification: TLP:CLEAR | Publication Date: 2026-11-22 | Source: ransomware.live leak-site monitoring | Nature of data: Unverified threat-actor claims

SAFEPAY Ransomware Gang: 9 New Leak-Site Listings in 24 Hours — Sector Targeting Analysis & Detection Engineering

Executive Summary

On 2026-11-21, the SAFEPAY ransomware operation listed nine organizations on its dark web leak site in a single burst, spanning eight countries and seven sectors including Financial Services, Healthcare, Energy & Utilities, Technology, Hospitality, and Professional Services. All nine listings were independently observed by a second leak-site crawler, confirming the gang published these claims — but none of the listings constitute a confirmed breach. A leak-site posting is an accusation by a criminal actor, and only the named organization or its regulator can confirm or refute it.

The nine organizations SAFEPAY names are: tiic.com.ph (Other, PH), threeroyal.com (Other, TW), factotal.com (Technology, CL), offshore-ocs.com (Other, SG), relax.es (Hospitality, ES), pacifico.com.ar (Financial Services, AR), qrc-solutions.com (Professional Services, EC), santak.com.sg (Energy & Utilities, SG), and themedical.co.il (Healthcare, IL).

The operational significance for defenders is not any individual claim — it is the tempo and shape of the campaign. Nine listings in one day across three continents indicates a mature, high-throughput intrusion pipeline, most consistent with bulk initial access (edge device exploitation or purchased access) rather than bespoke targeting. Organizations in the affected sectors should treat this as a hunting trigger: search for pre-encryption staging indicators now, before a detonation timer becomes the discovery mechanism.

Sourcing & Verification

  • Corroboration: 9 of 9 listings were independently observed by a second leak-site crawler; 0 listings appear on a single source only. Multi-source observation confirms the gang made these claims on its leak site.
  • What this does NOT mean: Inclusion in this briefing reflects the threat actor's claim and is not confirmation that any named organization was breached, attacked, or is managing an incident. No tier in our data supports breach confirmation.
  • Disputes and denials: A named organization may dispute a listing. A denial is likewise not proof the claim is false — disclosure obligations vary by jurisdiction and sector, and not every incident is reportable. Neither silence nor denial settles the question.
  • Corrections: Security Arsenal will publish corrections as warranted and welcomes contact from any named organization at security@securityarsenal.com.

Threat Actor Profile — SAFEPAY

AttributeDetail
Group nameSAFEPAY
Known aliasesSafepay ransomware; branding overlaps with the broader "pay-to-unlock" family naming conventions (monitor for tag collisions with SafePay-themed phishing lures)
Operating modelOperates as a Ransomware-as-a-Service (RaaS)-style program with affiliate-driven intrusions; leak-site output volume consistent with a semi-closed affiliate structure where the core team controls the leak site and negotiation infrastructure
Extortion modelDouble extortion: data exfiltration prior to encryption, with leak-site publication as the pressure lever
Typical ransom demandsSector- and revenue-scaled; mid-market victims in financial and professional services have historically seen demands in the low-to-mid six figures USD, escalating for regulated-data holders (healthcare, finance)
Known initial access methodsExploitation of internet-facing edge devices (VPN concentrators, firewall management interfaces, virtualization management planes), phishing with macro/loader payloads, exposed RDP, and brokered access from IAB ecosystems
Dwell time before detonationConsistent with modern RaaS norms: days to ~2 weeks, with exfiltration completing before encryption; the pre-detonation window is the defender's only reliable intervention opportunity
Lateral movement toolingLiving-off-the-land binaries (PsExec, WMIC), RDP pivoting, Cobalt Strike / Sliver-class beacons, scheduled-task deployment of the encryptor across hosts, and Volume Shadow Copy deletion via vssadmin/wmic before payload execution

Profile caveat

Ransomware attribution from leak-site telemetry alone is inherently noisy. The behavioral indicators in this briefing are drawn from the gang's documented playbook and should be treated as hunting hypotheses, not fingerprints unique to SAFEPAY.

Current Campaign Analysis

Sector targeting (from the 2026-11-21 listings)

The single-day burst shows deliberate sector diversification rather than a vertical spray:

  • Regulated-data sectors (high extortion leverage): Financial Services (pacifico.com.ar, AR) and Healthcare (themedical.co.il, IL) — both hold data classes that maximize pressure under double extortion.
  • Critical infrastructure-adjacent: Energy & Utilities (santak.com.sg, SG) — consistent with the broader 2025–2026 trend of ransomware crews testing operational-pressure tolerance.
  • Supply-chain multipliers: Technology (factotal.com, CL) and Professional Services (qrc-solutions.com, EC) — firms whose compromise can cascade to downstream clients.
  • Hospitality: relax.es (ES) — hospitality holds dense PII/payment records and historically under-invests in detection.

Geographic concentration

There is no single geographic concentration: the claims span the Philippines, Taiwan, Chile, Singapore (x2), Spain, Argentina, Ecuador, and Israel. The pattern — LATAM + APAC + Southern Europe + Middle East — is characteristic of opportunistic edge-device exploitation, where victims are selected by exposed vulnerable infrastructure rather than by strategic geography. Notably absent are US and Western European core targets, which may reflect affiliate rules-of-engagement, payment-processing friction, or simply where the vulnerable estate was found this cycle.

Victim profile

Based on the sectors represented, the claimed victims skew mid-market: large enough to hold monetizable data and pay, small enough to lack 24/7 detection. Estimated revenue bands run from roughly USD 5M (hospitality, regional professional services) to USD 50–150M (regional financial and energy services). This is the classic RaaS sweet spot — organizations with cyber insurance coverage but immature SOC capability.

Posting frequency / escalation pattern

Nine claims in a single day (2026-11-21) out of nine listings across the last 100 postings indicates the gang is front-loading a batch publication event rather than drip-feeding. Batch drops typically follow one of two patterns: (1) a completed intrusion wave from a single initial-access campaign, or (2) a negotiation-deadline mass expiry. Defenders should watch for a second batch in the coming days; sustained batch cadence suggests an active access pipeline, while a single batch suggests a completed campaign.

CVE linkage — hypothesis only

We have no evidence linking any specific CVE to any specific named listing above. That said, the gang's documented reliance on edge-device and management-plane exploitation aligns with several vulnerabilities currently in CISA's Known Exploited Vulnerabilities catalog with confirmed ransomware use:

  • CVE-2026-50751 — Check Point Security Gateway improper authentication (IKEv1): perimeter VPN auth bypass; canonical ransomware entry point. Added to KEV 2026-06-08.
  • CVE-2026-20316 — Cisco Secure Firewall Management Center hard-coded password: management-plane takeover enabling rule manipulation and lateral access. Added 2026-07-29.
  • CVE-2026-59310 — Broadcom VMware vCenter path traversal: virtualization management-plane compromise; a single foothold enables hypervisor-wide encryption, the highest-leverage ransomware outcome. Added 2026-08-18.
  • CVE-2026-63077 — JetBrains TeamCity deserialization: unauthenticated RCE on CI/CD infrastructure, enabling supply-chain-adjacent access to technology and professional-services firms. Added 2026-08-05.
  • CVE-2026-48027 — Nx Console embedded malicious code: developer-workstation supply-chain vector, relevant to the technology-sector listing pattern. Added 2026-05-27.

Treat these as sector-level exposure hypotheses: if your organization runs any of these products, patch per KEV deadlines and hunt for exploitation artifacts regardless of whether you appear on any leak site.

Detection Engineering

The following detections target SAFEPAY's documented TTPs across initial access, lateral movement, and pre-encryption staging. Deploy them as high-fidelity hunts in the affected sectors immediately.

YAML
---
title: SAFEPAY-Style Pre-Encryption Volume Shadow Copy Deletion
id: 8f3a1c2e-4b6d-4e9a-b1f5-safepay00001
status: experimental
description: Detects deletion or resize of Volume Shadow Copies via vssadmin, wmic, or bcdedit — a near-universal pre-encryption step in SAFEPAY and peer RaaS playbooks. High-fidelity when correlated with non-admin context.
author: Security Arsenal Threat Intelligence
date: 2026/11/22
references:
  - https://securityarsenal.com/darkside
logsource:
  category: process_creation
  product: windows
detection:
  selection_img:
    Image|endswith:
      - '\vssadmin.exe'
      - '\wmic.exe'
      - '\bcdedit.exe'
      - '\powershell.exe'
  selection_cmd:
    CommandLine|contains:
      - 'delete shadows'
      - 'shadowcopy delete'
      - 'shadowstorage'
      - 'recoveryenabled no'
      - 'ignoreallfailures'
      - 'win32_shadowcopy'
  condition: selection_img and selection_cmd
falsepositives:
  - Backup software maintenance windows
  - SCCM/Intune remediation scripts
level: high
tags:
  - attack.impact
  - attack.t1490
---
title: SAFEPAY Lateral Movement — Remote Service Execution via PsExec or WMI
id: 8f3a1c2e-4b6d-4e9a-b1f5-safepay00002
status: experimental
description: Detects PsExec-style service creation (ADMIN$ writes, PSEXESVC) and WMI remote process execution consistent with SAFEPAY's encryptor push-and-execute lateral movement.
author: Security Arsenal Threat Intelligence
date: 2026/11/22
logsource:
  product: windows
detection:
  selection_psexec:
    - EventID: 7045
      ServiceFileName|contains:
        - 'PSEXESVC'
        - 'ADMIN$'
        - '%SystemRoot%\\PSEXESVC.exe'
    - EventID: 5145
      Share_Name|endswith: 'ADMIN$'
      Relative_Target_Name|endswith: '.exe'
  selection_wmi:
    EventID: 4688
    New_Process_Name|endswith: '\WmiPrvSE.exe'
    Creator_Process_Name|endswith: '\svchost.exe'
    CommandLine|contains:
      - 'process call create'
  condition: 1 of selection_*
falsepositives:
  - Legitimate admin tooling (PDQ, SCCM, management agents) — baseline authorized hosts
level: high
tags:
  - attack.lateral-movement
  - attack.t1569.002
  - attack.t1021.002
  - attack.t1047
---
title: SAFEPAY Initial Access — Suspicious VPN/Edge Auth Followed by Internal Tool Execution
id: 8f3a1c2e-4b6d-4e9a-b1f5-safepay00003
status: experimental
description: Detects execution of reconnaissance or tunneling tooling (net, nltest, AdFind, ngrok, Cloudflare tunnel) shortly after VPN logon from an unusual source, consistent with edge-device or credential-based initial access used by SAFEPAY affiliates.
author: Security Arsenal Threat Intelligence
date: 2026/11/22
logsource:
  category: process_creation
  product: windows
detection:
  selection_tools:
    Image|endswith:
      - '\nltest.exe'
      - '\adfind.exe'
      - '\ngrok.exe'
      - '\cloudflared.exe'
      - '\rclone.exe'
      - '\7z.exe'
      - '\winrar.exe'
  selection_net:
    Image|endswith: '\net.exe'
    CommandLine|contains:
      - ' group "domain admins"'
      - ' view'
      - ' localgroup administrators'
  condition: selection_tools or selection_net
falsepositives:
  - IT administration; archive tooling on file servers — tune by host role and service account
level: medium
tags:
  - attack.discovery
  - attack.t1482
  - attack.t1087
  - attack.command-and-control
  - attack.t1572
KQL — Microsoft Sentinel / Defender
// SAFEPAY pre-ransomware staging hunt — Microsoft Sentinel
// Looks for the kill-chain signature: suspicious auth -> discovery -> staging/exfil tooling -> shadow copy tampering
// Run over the last 14 days; tune allowlists for your environment.
let lookback = 14d;
let suspiciousProcs = dynamic(["vssadmin.exe","wmic.exe","bcdedit.exe","rclone.exe","7z.exe","adfind.exe","nltest.exe","psexec.exe","cloudflared.exe","ngrok.exe"]);
let riskyCmds = dynamic(["delete shadows","shadowcopy","recoveryenabled","ignoreallfailures","megasync","copy \\"]);
let StagingHosts =
    DeviceProcessEvents
    | where TimeGenerated >= ago(lookback)
    | where FileName in~ (suspiciousProcs)
       or ProcessCommandLine has_any (riskyCmds)
    | summarize StagingEvents=count(),
                ToolsSeen=make_set(FileName),
                FirstSeen=min(TimeGenerated),
                LastSeen=max(TimeGenerated)
        by DeviceName, InitiatingProcessAccountName;
let RecentLogons =
    SigninLogs
    | where TimeGenerated >= ago(lookback)
    | where ResultType == 0
    | extend IsNewGeo = iff(Location !in (dynamic(["US","CA"])), "Unusual", "Baseline")  // adjust to your baseline geographies
    | project UserPrincipalName, IPAddress, Location, IsNewGeo, TimeGenerated;
let ServiceCreations =
    SecurityEvent
    | where TimeGenerated >= ago(lookback)
    | where EventID == 7045
    | where ServiceFileName has_any ("PSEXESVC","ADMIN$","temp","appdata")
    | project Computer, ServiceName, ServiceFileName, Account, TimeGenerated;
StagingHosts
| where StagingEvents >= 2 or array_length(ToolsSeen) >= 2
| join kind=leftouter (ServiceCreations) on $left.DeviceName == $right.Computer
| project DeviceName, InitiatingProcessAccountName, ToolsSeen, StagingEvents, FirstSeen, LastSeen, ServiceName, ServiceFileName
| order by LastSeen desc;
PowerShell
# SAFEPAY Rapid Triage — run on suspect hosts and domain controllers
# 1) Scheduled tasks created in the last 7 days (common encryptor-deployment mechanism)
# 2) Volume Shadow Copy status (deletion = imminent or completed encryption stage)
# 3) Exposed RDP (initial access check)
# Usage: powershell -ExecutionPolicy Bypass -File safepay_triage.ps1 | Out-File triage_$env:COMPUTERNAME.txt

Write-Output "=== SAFEPAY Rapid Triage: $env:COMPUTERNAME — $(Get-Date -Format o) ==="

Write-Output "`n[1] Scheduled tasks created in last 7 days (non-Microsoft):"
Get-ScheduledTask | Where-Object {
    $_.Date -gt (Get-Date).AddDays(-7) -and $_.Author -notmatch 'Microsoft'
} | Select-Object TaskName, TaskPath, Author, Date, @{N='Action';E={$_.Actions.Execute}} | Format-List

Write-Output "`n[2] Volume Shadow Copy status:"
$shadows = Get-CimInstance Win32_ShadowCopy -ErrorAction SilentlyContinue
if ($shadows) {
    $shadows | Select-Object ID, InstallDate, VolumeName | Format-Table -AutoSize
} else {
    Write-Output "  !! NO shadow copies present — possible vssadmin deletion (pre-encryption indicator) !!"
}
# Check for shadow deletion events in System log (VSS EventID or vssadmin execution)
Get-WinEvent -FilterHashtable @{LogName='System'; Id=7036} -MaxEvents 50 -ErrorAction SilentlyContinue |
    Where-Object { $_.Message -match 'Volume Shadow Copy' -and $_.TimeCreated -gt (Get-Date).AddDays(-7) } |
    Select-Object TimeCreated, Message | Format-List

Write-Output "`n[3] RDP exposure check:"
$rdpEnabled = (Get-ItemProperty 'HKLM:\System\CurrentControlSet\Control\Terminal Server').fDenyTSConnections -eq 0
Write-Output "  RDP enabled: $rdpEnabled"
$nla = (Get-ItemProperty 'HKLM:\System\CurrentControlSet\Control\Terminal Server\WinStations\RDP-Tcp' -ErrorAction SilentlyContinue).UserAuthentication
Write-Output "  NLA required: $($nla -eq 1)"
$rdpOpen = Test-NetConnection -ComputerName localhost -Port 3389 -WarningAction SilentlyContinue
Write-Output "  Port 3389 listening: $($rdpOpen.TcpTestSucceeded)"

Write-Output "`n[4] Recent failed logons (brute-force signal, last 24h, top sources):"
Get-WinEvent -FilterHashtable @{LogName='Security'; Id=4625; StartTime=(Get-Date).AddDays(-1)} -ErrorAction SilentlyContinue |
    Group-Object { ($_.Properties[19].Value) } | Sort-Object Count -Descending | Select-Object -First 10 Count, Name | Format-Table -AutoSize

Write-Output "`n=== Triage complete. Escalate any shadow-copy absence or unknown scheduled tasks to IR immediately. ==="

Incident Response Priorities

T-minus detection checklist (BEFORE encryption fires)

The pre-detonation window is where this fight is won. Alert and escalate on any of the following:

  1. Shadow copy tampering — vssadmin delete shadows, bcdedit ... recoveryenabled no, or sudden absence of restore points on servers.
  2. New scheduled tasks with SYSTEM privileges pointing to payloads in C:\ProgramData, %TEMP%, or user profiles — the standard encryptor push mechanism.
  3. PsExec/ADMIN$ writes to multiple hosts in a short window, especially from a single source host.
  4. Discovery bursts — nltest /dclist, net group "domain admins", AdFind execution — typically run in the final 48 hours.
  5. Data staging artifacts — large multi-GB archives (7z/RAR) on file servers, or rclone/MEGAsync/cloudflared binaries on hosts that have no business running them.
  6. EDR tampering — service stops, sensor uninstalls, or exclusions added to Defender/AV policy by non-standard accounts.

Assets SAFEPAY-style crews historically prioritize for exfiltration

  • Financial records, payroll, and banking detail (maximizes pressure on financial and professional-services victims)
  • Customer databases and PII (healthcare and hospitality pressure lever)
  • Legal/contracts and M&A material
  • Email archives of executive leadership
  • Backup repositories (targeted for destruction or theft to eliminate the no-pay recovery path)

Containment actions, ordered by urgency

  1. Isolate, don't power off — network-quarantine suspected staging hosts (preserve volatile memory and staging evidence).
  2. Kill the identity plane — force-reset credentials for any account seen in staging activity; revoke VPN sessions and tokens; disable suspicious service accounts.
  3. Block egress — deny outbound to known exfil tooling endpoints (MEGA, rclone remotes, tunneling services) at the proxy/firewall.
  4. Protect backups immediately — verify backup infrastructure is segmented and immutable; take backup management interfaces off the domain.
  5. Sweep the edge — audit VPN/firewall/management-plane logs for the KEV vulnerabilities above; assume the initial vector is still open.
  6. Engage IR counsel and review disclosure obligations — if your organization appears on a leak site, a claim alone may trigger regulatory clocks depending on jurisdiction; involve legal early.

Hardening Recommendations

Immediate (24 hours)

  • Patch or mitigate the five KEV entries (CVE-2026-59310, CVE-2026-63077, CVE-2026-20316, CVE-2026-50751, CVE-2026-48027). If patching is not possible today, take the management interfaces off the internet entirely — vCenter, FMC, TeamCity, and Check Point gateways should never be internet-reachable.
  • Deploy the Sigma rules above and run the Sentinel hunt query across the last 14 days.
  • Block shadow-copy deletion for non-backup accounts via AppLocker/WDAC rules on vssadmin.exe, wmic.exe, and bcdedit.exe.
  • Enforce MFA on all VPN and remote access, and audit active sessions for impossible-travel or new-ASN logins.
  • Verify backup immutability and confirm an offline or logically air-gapped copy exists and has been restoration-tested.

Short-term (2 weeks)

  • Segment backup and virtualization management planes onto isolated networks with dedicated admin accounts (tiered administration); a vCenter compromise should never be reachable from a user workstation.
  • Deploy egress filtering with TLS inspection or DNS-based blocking for exfiltration tooling (rclone remotes, consumer cloud storage, tunneling services).
  • Implement lateral-movement friction: disable PsExec-class remote service creation where not required, restrict WMI to admin subnets, enable LAPS everywhere.
  • Stand up canary artifacts — decoy files and honey credentials that fire high-fidelity alerts on staging access.
  • Run a purple-team exercise against the SAFEPAY playbook: edge access → discovery → staging → shadow deletion → encryptor push, and verify each stage produces an alert.

Related Resources

Security Arsenal Incident Response Managed SOC & MDR Services AlertMonitor Threat Detection From The Dark Side Intel Hub

Is your security operations ready?

Get a free SOC assessment or see how AlertMonitor cuts through alert noise with automated triage.