// 4 ZERO-DAY · 5 CVE · 5 EXPLOIT IN THE LAST 24H
Arctic Wolf Labs confirms active exploitation of CVE-2026-0257 to deploy Qilin ransomware. Specific TTPs reveal shared infrastructure among affiliates.

Arctic Wolf Labs has reconstructed multiple intrusions from June 2026 that all originate from a single flaw: CVE-2026-0257, an authentication bypass with a CVSS score of 9.1 in the Palo Alto Networks GlobalProtect portal and gateway. Attackers obtained interactive VPN sessions without credentials, then expanded through lateral movement and credential harvesting before deploying Qilin payloads. The critical finding is not the individual breach, but the observed maturation: the first exploitation dates to May 17, 2026, according to Rapid7, while the massive ransomware phase concentrated in June, with a roughly six-week gap that allowed operators to refine the attack chain.

Key Takeaways
  • CVE-2026-0257 is an authentication bypass in PAN-OS GlobalProtect with a CVSS of 9.1/10 (critical), added to the CISA KEV on May 29, 2026.
  • Arctic Wolf Labs investigated June 2026 intrusions with consistent TTPs: win.exe payload in C:\PerfLogs\, persistence via registry Run keys with pattern *[a-z]{6}, and source hostname kali.
  • Post-compromise: LSASS credential dumping via rundll32.exe + comsvcs.dll masked as .odt, Active Directory extraction via ntdsutil.exe, lateral movement with PsExec and C$ shares.
  • Unit 42 and Rapid7 confirmed active exploitation from May 2026 but did not observe lateral movement or identify the Qilin link, which remains specific to Arctic Wolf reporting.

The Flaw: How the GlobalProtect Bypass Works

CVE-2026-0257 affects the GlobalProtect portal and gateway, the PAN-OS components that manage remote VPN access. The vulnerability allows a remote attacker to completely bypass perimeter authentication, obtaining a legitimate VPN session without valid credentials. This is not a local escalation or a configuration weakness: it is a privilege jump that turns the perimeter from a barrier into an invisible entry point.

The severity is quantified at CVSS 9.1/10 (critical), with a network attack vector, low complexity, no privileges required, and no user interaction needed. Affected versions prior to specific patches include PAN-OS 12.1, 11.2, 11.1, and 10.2, according to Arctic Wolf reporting. CISA added the vulnerability to the Known Exploited Vulnerabilities Catalog on May 29, 2026, confirming its exploited-in-the-wild status at the federal level.

From May to June: The Timeline Reveals Maturation, Not Chance

The temporal sequence is the most relevant data point for risk assessment. Rapid7 identified successful exploitation as early as May 17, 2026, via its MDR service, but explicitly stated it did not observe lateral movement in that phase. Unit 42, in its June 9, 2026 report, confirmed active exploitation without identifying post-access behavior or lateral movement. The lack of lateral movement observed by the two sources does not mean absence: it means the initial activity was confined to the entry point, or that post-exploitation techniques evaded the sensors available at the time.

Arctic Wolf Labs instead investigated full intrusions in June, with Qilin ransomware deployment. The jump from "silent" exploitation to a massive ransomware campaign indicates operators used the May-June period to perfect the chain: credential harvesting, persistence, lateral movement, defense disabling, and only then encryption. The Qilin RaaS (Ransomware-as-a-Service) model, which distributes the working exploit to multiple affiliates with their own post-compromise techniques, explains the tactical variations observed in June.

TTPs as Fingerprint: Shared Infrastructure Among Affiliates

Arctic Wolf documented an unusually specific TTP pattern that serves as a fingerprint for shared infrastructure or toolkit. The ransomware payload was consistently named win.exe and placed in C:\PerfLogs\, a Windows performance monitoring directory rarely subject to active scrutiny. Persistence was created via registry Run keys with pattern *[a-z]{6}: an asterisk followed by six random lowercase letters.

Deployed RAT tools included AnyDesk, Ngrok, LogMeIn, and MeshAgent. Credential dumping from LSASS occurred via rundll32.exe loading comsvcs.dll, with output masked as an .odt (OpenDocument Text) file to evade extension-based filters. Active Directory database extraction used ntdsutil.exe with the Install From Media method. Lateral movement leveraged PsExec and administrative C$ shares. Some attacks included data exfiltration to MEGA cloud storage via Rclone.

The hostname kali and overlapping IP addresses between exploitation phases and subsequent VPN sessions indicate the same operators, or a single access toolkit, rotated across infrastructures. This is not an anecdotal detail: it is the indicator that the exploit has been integrated into a reusable arsenal, not a one-off operation.

Immediate Actions

Priority actions derive directly from the documented facts:

  • Immediate patching: apply corrective PAN-OS updates for CVE-2026-0257 on versions 12.1, 11.2, 11.1, and 10.2, verifying specific build availability in vendor advisories.
  • Terminate active sessions: invalidate all existing GlobalProtect VPN sessions and force re-authentication, as the bypass allows previously established unauthorized sessions.
  • Hunt for C:\PerfLogs\win.exe: check for the presence of this specific file and the registry Run keys pattern *[a-z]{6} as compromise indicators.
  • Correlate May-June 2026 logs: search for VPN access from hostname kali or suspicious IP addresses in the period May 17 - June 30, 2026, preceding ransomware deployment.

The sources do not quantify the exact number of victims nor the geography of the June 2026 intrusions. It also does not emerge whether Qilin published exfiltrated data on a leak site for double extortion, nor whether a public PoC exists for CVE-2026-0257 beyond the generic Unit 42 reference to "post-PoC release monitoring."

"Arctic Wolf assesses with moderate confidence that exploitation of this vulnerability leading to Qilin ransomware deployment is ongoing, driven by widespread scanning activity and the RaaS model's tendency to distribute working exploits across multiple affiliates." — Arctic Wolf Labs

The Qilin Model and Exploit Proliferation

Arctic Wolf's observation that "variations in tactics observed during the June incidents may reflect different Qilin affiliates leveraging a shared exploit method while employing their own post-compromise techniques" is the key to understanding the threat. Qilin, active since 2022 and also known as Agenda, is not a monolithic group but a RaaS label under which distinct actors operate. The VPN bypass becomes a leveler: it lowers the skill threshold for initial access, shifting the value add to the subsequent lateral movement and encryption phases.

The risk is not confined to already breached organizations. It is systemic: every internet-exposed GlobalProtect instance with vulnerable PAN-OS represents a potential entry point for any Qilin affiliate with access to the exploit. RaaS distribution implies proliferation will occur faster than the average patching capacity of target organizations.

Frequently Asked Questions

Why is the six-week gap between exploitation and ransomware significant?

The May-June period allowed operators to test and stabilize the full chain: from VPN bypass to credential harvesting, from lateral movement to persistence. Rapid7 and Unit 42 observed only the initial phase, indicating post-exploitation evaded standard sensors or occurred on targets not covered by those vendors.

What is the relationship between CVE-2026-0257 and CVE-2026-0300?

None is documented. They are distinct CVEs with different vectors. CVE-2026-0300, a buffer overflow RCE, cannot be used to support claims about CVE-2026-0257 and is not cited by Arctic Wolf sources or Qilin exploitation confirmations.

No. Confirmation of the link between CVE-2026-0257 and Qilin deployment comes exclusively from Arctic Wolf Labs. Unit 42 confirmed active exploitation of the same CVE without identifying post-access behavior or specific ransomware.

Sources

Information verified against cited sources and current as of publication.

Sources


Sources and references
  1. cybersecuritynews.com
  2. gbhackers.com
  3. unit42.paloaltonetworks.com
  4. thehackernews.com
  5. cyfirma.com
  6. techradar.com
  7. security.paloaltonetworks.com
  8. cisa.gov