The images in this article were generated with artificial intelligence. How we publish
The scenario described by the response teams and the companies that have investigated the campaign is direct and fast. According to public reports from industry firms, in mid-January, anomalous accesses, including the creation of administrative accounts with VPN access and the exfiltration of configurations of the equipment in question in a matter of seconds, began to be detected. The speed and uniformity of these intrusions make you think of an automated process that takes advantage of some new attack route or a way to draw the previous correction.
Fortinet publicly confirmed that the activity observed in recent weeks is similar to the holdings documented in December and that the company is working to identify and remedy completely the vector responsible for the new intrusions. In his technical note on the abuse of SSO in FortiOS ( see communication from Fortinet) the company recognizes that, although the known holding has so far affected FortiCloud SSO, the underlying problem is related to SAML SSO implementation in general.

The signs that have shared clients and analysts help to rebuild part of the attack: in several access records it is noted that the administrative accounts were created after a SSO login from the cloud-init @ mail.io mail address and from the IP 104.28.244.114, coincidences that were also identified by the security firm that published observations on the campaign. Fortinet, for its part, has listed commitment indicators that can be used for the search for evidence on potentially affected devices ( see IOCs shared by Fortinet).
To size the exposure, the Shadowserver surveillance group maintains a team tracking with FortiCloud SSO exposed on the Internet and its board shows about 11,000 publicly accessible devices with that activated service ( see Shadowserver panel). This figure explains why the addition of the CVE-2025-59718 to the list of actively exploited vulnerabilities of the U.S. agency CISA generated urgent measures: on December 16, CISA incorporated this failure into its catalog and ordered federal agencies to apply patches in time-bound ( CISA statement, entry into the catalogue).
In this context, the practical recommendations are clear and urgent. Fortinet has suggested limiting administrative access from the Internet by applying a local-in policy that restricts IP addresses authorized to manage devices; official documentation explains how to implement this policy in FortiGate ( more information about local-in policy). In addition, while delivering a final correction, the company advises to disable the option to log in administratively using FortiCloud SSO in system settings and to verify any engagement trace in the records.
If, when reviewing the machines, the commitment indicators associated with this campaign are found, the operational recommendation is to treat the equipment and its configuration as committed: change all relevant credentials, including LDAP / AD accounts that may have been exposed, and restore the configuration from a backup known as clean. Fortinet has published technical steps and resources in his community to guide the response ( see suggestions in the Fortinet community).

Beyond this particular incident, there are two lessons that should be internalized. The first is that centralized authentication solutions such as SAML / SSO, which offer comfort and control, also concentrate risk: a vulnerability in that chain of trust can give wide and rapid access to an attacker. The second is that having applied a patch does not guarantee that the exposure has disappeared; there are always cases where alternative operating routes arise or the correction does not cover unforeseen scenarios. Therefore, in critical environments, in-depth defense measures - access segmentation, IP whitelisting for administration, continuous log monitoring and restoration from clean backup - remain essential.
If you manage Fortinet in a business network, it is appropriate to act immediately: review the configuration of FortiCloud SSO, audit recent access, apply Internet management blocks and prepare response plans that include the rotation of credentials and safe restoration. Many of these instructions are detailed in the official sources listed in this article; review and follow them significantly reduces the likelihood of similar intrusion.
The situation is still developing and both security providers and manufacturers will continue to publish technical updates. To keep you up to date and compare information, check the publications of manufacturers and alerts of organisms such as CISA, as well as reports of response companies that monitor malicious activity in the field. In an environment where attacks are quickly automated and spread, active prevention and early detection are the best defense.
Related
More news on the same subject.

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...

OpenCode: possible exposure of AI suppliers in Console affects 1.4% of users
On the night of October 16 there was an unauthorized access to a system inherited from OpenCode which, according to the company, could have exposed keys from IA suppliers that u...

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...