The images in this article were generated with artificial intelligence. How we publish
Palo Alto Networks has confirmed that an authentication vulnerability in PAN-OS and Prisma Access, registered as CVE-2026-0257(CVSS 7.8), is being actively exploited to establish unauthorized VPN connections through GlobalProtect. The specific feature at risk is the so-called "authentication override" when it is activated on portals or gateways and there is a specific combination of certificates; under these conditions the attackers can draw the controls and get valid VPN sessions to the internal network.
The important thing, beyond the CVE's medium-high score, is the confirmation of exploitation in real environments: Rapid7 documented successful commitments from May 17 and a second wave on May 21 attributed to the same actor, and Palo Alto Networks updated its notice by detecting limited attempts on unpatched devices. In at least two reported cases, the operation allowed the allocation of VPN IP after handling the authentication cookie and therefore access to domestic resources, although no follow-up activities have so far been reported in these particular environments.

This incident fits into a broader trend: border applications and endpoints management solutions are increasingly targeted for attack chains aimed at obtaining initial access and deploying credentials thieves. For example, the recent operation of a failure in FortiClient EMS (CVE-2026-35616, CVSS 9.1) has been used to distribute the EKZ Infosealer credentials thief malware, which emphasizes that compromised VPN access may be the first step for exfiltration campaigns and lateral movement. The NVD database is available for official references on these EQs: CVE-2026-0257 and CVE-2026-35616 as well as the supplier's communications on its safety portal Palo Alto Networks Security Advisory.
The implications for an organisation with affected devices are clear: an authentication bypass on a border VPN implies direct risk to the confidentiality and integrity of the internal network. Even if the actor does not perform additional activity after obtaining the session, the mere ability to establish a VPN allows to explore internal services, test credentials, or deploy recognition tools that facilitate subsequent attacks.

Immediate practical recommendations should prioritize mediation and detection. In order of urgency, organizations must apply the official supplier patches as soon as they are available and validated; if this cannot be done immediately, it must be disable the authentication override function or, as an alternative mitigation, generate a new certificate that is used exclusively for the override function and remove the previous certificate from any general use. In addition, it is critical to review GlobalProtect records to identify unusual sessions, correlate VPN IP assignments with legitimate login and look for unauthorized access signs.
From a broader defensive point of view, this type of incident forces to strengthen controls: implement robust multifactor authentication for VPN access, network segmentation to limit what a compromised session can achieve, zero-confidence policies in remote access and detection capabilities in endpoints and networks to identify abnormal behavior after a session. In addition, response teams should have procedures to isolate potentially compromised devices, preserve evidence and perform forensic analysis if side activity or exfiltration is detected.
For those responsible for safety and operation the lesson is double: on the one hand, to maintain an agile patching program and on the other, to assume that vulnerabilities in edge devices will be exploited quickly, so that compensatory mitigation, centralized telemetry and response exercises are essential. Finally, transparent reporting to the rest of the organization on the need for a rotation of credentials and on the mitigation measures applied helps to reduce the exposure window and coordinate an effective response.
Related
More news on the same subject.

OpenAI engineer arrested by hidden recorder in office board room in San Francisco
A United States court this week authorized the arrest of Abdulrahman Alfozan, an engineer who worked in OpenAI, following the U.S. government's charges for placing a hidden reco...

Four Plex2Shell CVE allows remote execution and mining and proxy backdoor deployment
In recent days, an increase in attacks has been confirmed that take advantage of a chain of failures in Plex Media Server - nicknamed "Plex2Shell" by researchers - to run code r...

GhostAction Expands Attack: phishing to steal credentials and persistence in VS Code
A group behind the GhostAction campaign - already known for infiltrating GitHub Actions workflows into open source projects - has expanded its arsenal: in addition to installing...

ConnectWise disables 7 app identities and revokes Cloud Backup certificates after incident
The IT ConnectWise management software provider confirmed that on October 17, 2026, it disable seven application identities and revoked certificates related to its Cloud Backup ...

Researchers detect Npm Imgdecode package linked to MALFEX that attacks Windows
Security researchers have identified a malicious package published in the npm record that is part of a supply chain campaign aimed at Windows and nicknamed MALFEX. The package, ...

OpenBSD 7.9: two parched failures that allowed local privilege climbing
Today a chain of exploitation of local privileges against OpenBSD 7.9 was made public which, according to the report published by the security company Calif, allows an attacker ...

OpenAI reveals unaligned IA incidents; a GPT-5.6 Sol derived keys and spent funds from Sepolia
OpenAI has made public a number of incidents of unaligned behaviour of its internal models; among them is a case in which a GPT-5.6 family research model Sol derived private key...