Fortinet has disclosed a critical authentication bypass vulnerability that affects multiple Fortinet products. It has been actively exploited in the wild, enabling attackers to access customer devices by abusing FortiCloud single sign-on (SSO) trust mechanisms.
The vulnerability, tracked as CVE-2026-24858 and assigned a CVSS score of 9.4, impacts FortiOS, FortiManager, FortiAnalyzer, and FortiProxy when FortiCloud SSO authentication is enabled.Â
The company confirmed that attackers with valid FortiCloud accounts and registered devices were able to authenticate to systems associated with unrelated customer environments.
Fortinet said it implemented emergency controls at the platform level to limit further abuse.
Fortinet’s Product Security Incident Response Team (PSIRT) said the issue involves an authentication bypass through an alternate access path. This allowed FortiCloud SSO authentication to succeed under conditions that should otherwise prevent cross-account access.
While the company blocked FortiCloud SSO access from vulnerable versions, the vulnerability remains relevant for organizations that have not yet upgraded.
The company said it identified two FortiCloud accounts being used maliciously to exploit the flaw and locked those accounts on January 22.
Although FortiCloud SSO is not enabled in default factory settings, it may be activated during device registration to FortiCare, unless administrators explicitly disable the option permitting administrative login through FortiCloud SSO.
Fortinet said attackers who successfully authenticated via FortiCloud SSO were observed:
Once authenticated, attackers were observed carrying out multiple post-access actions.Â
Researchers marked the following accounts under indicators of compromise that have been neutralized:
The following IP addresses were observed being used for login activity:
These IP addresses were reported by a third party and not directly observed by Fortinet:
The malicious activity included the creation of admin accounts with common names such as backup, itadmin, support, and security. Hence, Fortinet urged customers to audit all admin users.
The attackers also rotated network infrastructure and leveraged Cloudflare-protected IP addresses, which further complicated detection and monitoring of malicious activities.
To reduce risk, Fortinet disabled FortiCloud SSO globally on January 26, and re-enabled the service on January 27 with safeguards. Under the updated controls, FortiCloud SSO authentication is now blocked for devices running vulnerable software versions.
Customers must upgrade to fixed releases for FortiCloud SSO functionality to work. Disabling SSO locally is no longer required, as it is handled server-side.
Fortinet confirmed the vulnerability was exploited prior to its mitigation steps. Researchers are also investigating additional products, including FortiWeb and FortiSwitch Manager.
Although the company’s platform-level controls now restrict the attack path, the flaw poses a risk to organizations that have not yet applied updates. The risk is driven less by active exploitation and more by slowed patch adoption.
The flaw exploited identity trust relationships between cloud services and on-premise infrastructure, allowing attackers to move laterally across customer environments without a direct device breach.
Customers must upgrade affected products to fixed versions and review administrative access for unauthorized activities. Organizations are advised to examine authentication logs for suspicious FortiCloud SSO activity.Â
The company emphasized that FortiCloud SSO will not function on vulnerable versions, making upgrades mandatory for continued use.