Attackers now exploiting three critical FortiSandbox flaws, one with AI-built exploit

Threat-intelligence firm Defused reports that attackers are now exploiting three critical flaws in Fortinet's FortiSandbox, the appliance other Fortinet products rely on to judge whether files are malicious. Two (CVE-2026-39813, a JRPC API path traversal that bypasses authentication, and CVE-2026-39808, an unauthenticated command-injection that runs code as root) were patched in April; the third (CVE-2026-25089) only last week. All are unauthenticated and rated critical. Compromising a sandbox is especially dangerous because attackers can make it wave real malware through as clean. Notably, the exploit for one flaw appears to have been generated with AI and is likely faulty, yet attackers are trying it anyway.

Check

Identify FortiSandbox, FortiSandbox Cloud, and PaaS instances and their versions, confirm whether the web and JRPC API interfaces are reachable from untrusted networks, and review logs for unauthenticated command execution.

Affected

FortiSandbox, FortiSandbox Cloud, and FortiSandbox PaaS that are unpatched against CVE-2026-39813, CVE-2026-39808, or CVE-2026-25089, especially instances exposed to untrusted networks; all three need no authentication.

Fix

Upgrade FortiSandbox to the fixed releases for all three CVEs immediately, restrict management and API interfaces to trusted networks, and treat any unpatched appliance as potentially compromised pending review.